MITIGATING DEMAND UNCERTAINTY THROUGH SUPPLY CHAIN STRATEGIES: THE CASE OF FOOD SMES IN THE HAJJ PHENOMENON

THESIS OF DOCTOR OF PHILOSOPHY
2015
CHAPTER 1: INTRODUCTION

1.1 Introduction
This chapter is centred on providing the reader with an overview of the thesis, beginning by discussing the research background and providing a definition of the problem, in addition to the scope of the research. The aim of the study, along with the research question and its main objectives, are articulated throughout the course of the chapter. The research gaps the study seeks to fill and justification for the study are also discussed. The chapter concludes by discussing the motivation for conducting the study (significance) and providing a definition of key terms. This is followed with an outline of the thesis organisation.
Small- and medium-sized enterprises (commonly referred to as SMEs) play an important role in the food chain throughout the Hajj season in the Kingdom of Saudi Arabia (KSA). However, SMEs are recognised as experiencing severe obstacles that have the potential to threaten their continuity, and the industry succumbs to the crisis of demand uncertainty throughout the short period of the peak season of Hajj each year. This problem is complex due to the constant increase in the number of pilgrims and the continuous changes in their needs, preferences and tastes. Demand uncertainty can ultimately result in an increase in production costs, long lead times, substandard service levels, and quality problems, especially in terms of food obsolescence.
There is a gap in the literature in regard to aligning sources of uncertainty with supply chain strategies in an effort to improve supply chain performance. More specifically, the impact of Supply Chain Integration (SCI) on manufacturing strategies, such as postponement practice (PP) to mitigate demand uncertainty (DUM), has not been explored in its entirety. This study investigates three fundamental issues: 1) how effective supply chain, internal integration and external integration can be applied, and how the interaction between them mutually manipulates the improvement of PP and MCC; 2) the volume of cooperation leading to the mitigation of demand uncertainty in maintaining the survival of SMEs to sustain their performance; and 3) how the environmental condition (i.e. competitive intensity) moderates the influence of SCI on this interaction. Based on the extended resource-based view of the firm, the strategic resources and knowledge come not only from within the organisation’s boundaries, but also from outside. Thus, a firm’s overall strategic capability may be embedded in a wider network of inter-firm exchange relationship. Contingency theory also argues that an organisation should align their practices, processes and strategies with their business environment. In consideration to the extant literature, a number of hypotheses were defined, demonstrating the correlation between SCI, PP and DUM. Subsequently, a conceptual framework was developed with the objective to verify the relationship amongst the constructs. Mixed methodologies were employed; qualitatively, 12 datasets of CEOs of SMEs in the sector were used to validate the conceptual framework. Content and face validity was accomplished with a group of academics and experts. A pilot study was carried out on a sample of 50 subcontractors, Hajj campaigns, pilgrimage institutions and food suppliers. Subsequently, online administration was conducted amongst SMEs to test the hypotheses. As a result, 239 responses were received from the SMEs in the food sector in the KSA. Partial least square (PLS) was used for analysis. Based upon the extended resource-based view (ERBV) of the firm, alongside contingency theory, the initial and final results of the pilot test and survey were seen to be steady with the ERBV, where SCI was viewed not only as having a significant direct and indirect effect on the PP and MCC, but also as playing a critical role throughout the employment of PP as an important strategy, empowering MCC to mitigate demand uncertainty. Likewise, both results were seen to be consistent with contingency theory; that is, a firm should coordinate their SCI activities, PP and MCC to their business environment, particularly with high competitive intensity to enhance DUM. In order to achieve competitive intensity, organisations are majorly focused on emerging markets and expanding their product lines. In the event that organisations begin targeting similar set opportunities, they risk bringing up competitive intensity for themselves, which increases the cost of growth. The cost of business will be noticeable when considering marketing speed, media inflation, the rate of innovation and trade spend in marketing, all of which are indicators of completion intensity. Improvements in supply chain efficiency, optimising strategies in marketing and extracting the best of return on investments from promotions by organisations also indicates competition intensity. Porter identified costs, competition and the ability to differentiate as some of the main determining factors of competition. Importantly, these are all tied up within uncertainty mitigation.
However, despite the fact internal integration has a positive direct and indirect effect on the PP and MCC through both SI and customer integration, customer integration also has been found to improve PP and MCC in a direct fashion. Supplier integration has a significant impact on PP; however, it seems to have no significant impact on MCC. Moreover, PP also has an effect in mitigating demand uncertainty, both directly and indirectly, through MCC. Finally, MCC also has been found to enhance demand uncertainty mitigation. Research indicates that the direct and indirect effects of all constructs increase when there is intense competition.
Keywords: Supply Chain integration (SCI), Postponement Practice (PP), Mass Customisation Capability (MCC), Demand Uncertainty Mitigation (DUM), Competitive intensity, Extended Resource-Based View (ERBV), Contingency theory, Small and Medium Enterprises (SMEs), Hajj operation.

1.2 Research Background
Every organisation, whether big or small, takes adequate steps to mitigate demand uncertainty. This arises as a result of the continuous globalisation and technological advancement, which subsequently occurs in the global market. Businesses have to deal with supply and demand in order to ensure customers are adequately served with goods and services, whilst at the same time maintaining a good inventory management system. When business has difficulty in predicting demand, it poses challenges to the supply chain: in such circumstances, inventory becomes difficult to control (Christopher, 2005). There are a variety of reasons responsible for the cause in demand uncertainty: for example, there could be an increase in the buying relative to the projections or decrease in buying relative to the expectations. If organisations experience problems in demand uncertainty, they could easily succumb to pressure during peak seasons (Ayers, 2001). Demand uncertainty, however, can be mitigated with the use of proper management methods. This research examines the mitigation of demand uncertainty in Saudi’s SME food industry.
The study is focused on analysing management practices that will improve the performance of the supply chain, especially on the demand side and under the condition of stiff market competition from both similar and dissimilar competitors. The SME food industry during the Hajj Season is competitive owing to the enormous demand needing to be fulfilled; however, demand never lacks its uncertain edge, thereby emanating robust competition in the supply chain. For example, during the peak season, there is a plethora of demand uncertainty owing to a wide array of customers’ preferences.
Unfortunately, this may result in the development of long lead times that could be considered substandard and deteriorative to service quality provision, as well as the cost of provisions. Therefore, if appropriate, managerial strategies, such as the capability of mass food customisation, Supply Chain Integration and postponement practices, are implemented in the supply chain, which could go some way to achieving and maintaining a sustainable competitive edge, as suggested by Amit et al. (2005).
Demand uncertainty can be detrimental to SMEs as a result of their performance weaknesses; this can result in the deterioration of quality, in addition to delays in delivery (lack of fulfilling timely orders). All of these factors could arise owing to the fact that demand uncertainty contributes to the emanation of supply chain uncertainty, manufacturing chain uncertainty and control uncertainty—especially during the Hajj season, when demand for SMEs food products skyrockets. This is owing to demand amplifications and end-customer demands, which directly or indirectly contribute to inaccurate forecast (Hult, Craighead & Ketchen, 2010). Importantly, this could be caused by a lack of making comprehensive decisions in the supply chain. With regards to demand uncertainty, the decision-maker in the supply chain lacks prudent characteristics that would appropriately guide his decisions. Firms in Saudi’s SME food industry could, during the Hajj season, lack convincing information pertaining to the market environment, current supply chain behaviour, processing capacities and effective control actions, meaning the making of conclusive decisions to enable them gain competitive advantage against their counterparts is not achievable (Boyle, Humphreys & McIvor, 2008).
Studies show that sources of demand uncertainty can be studied from the standpoint of the management of supply-chain uncertainty (Amit et al., 2005): for example, accurate forecasts can be used to make assumptions regarding the kind of relationship that may exist between customers and a number of SMEs in the food industry in the Hajj season. Appropriate steps can be undertaken so as to ensure that the supply chain is apt in meeting all customer demands. This should ensure its provisions are all rounded, fully satisfying customer needs and preferences. Without question, this could create a competitive edge and thus enable firms to strive in the stiff market competition, whilst also gaining and maintaining market control (Boyle, Humphreys & McIvor, 2008). Based on this, the study seeks to examine the mitigation of demand uncertainty in the Saudi SME food industry during the Hajj season by analysing Supply Chain Integration, mass food customisation and postponement practices as used by the SMEs in the industry.
1.3 Problem Definition
This paper primarily seeks to address demand uncertainty mitigation through management practices in an effort to satisfy all customers who go to Mecca to attend Hajj, which in turn leads to leverage on SMEs’ performance. Hajj attracts a wide variety of people who visit Mecca for a week of religious activities. Individuals come from different places and have different tastes, all needing to be satisfied. This paper looks at ways in which customer expectations can be met to a given quality level in the SME food industry in Saudi Arabia.
Demand uncertainty has been a burden to most SMEs in the food industry in Saudi Arabia. When demand uncertainty grips an organisation, it affects the way the supply chain is managed. When completing inventory management, organisations strive to minimise stock outs and avoid the costs associated with holding inventory in excess. When an organisation is unable to precisely predict demand, it runs the risk of over-buying and -selling at discounted rates in order to sell off excess (Christopher, 2005). At the same time, if an organisation buys less in order to prevent wastage, peak times lead to stock-outs. This erodes customer confidence and can result in the loss of essential customers (Ulrich, 1995).
Several management practices have been put in place by different organisations with the aim of helping mitigate demand uncertainty. The Just in Time inventory is used as one of the methods of helping mitigate the issues associated with demand uncertainty; this is a process where inventory data is shared between vendors and retailers. This sharing of data allows retailers to complete the rapid replenishment of stock. Here, the retailers maintain low inventory levels and make frequent but smaller orders (Christopher, 2005). Although this management method is helpful in mitigating demand uncertainty effects, there remains the risk of stock-outs when demand rises quickly (Ulrich, 1995). Throughout the Hajji season, the SME food industry in Saudi faces this problem, even though it has adopted the Just in Time inventory management system. The SME food industry in Saudi has had to face the bullwhip effect that results from demand uncertainty (Shalaby, 2012). According to Lee et al. (2007, p. 93), the bullwhip effect takes place when demand order uncertainties magnify as they move up the chain. According to these authors, distorted information from one end of the chain to the other can result in tremendous inefficiencies. These authors add that firms can effectively offset the bullwhip effect by understanding and accordingly addressing its underlying causes. This effect is based on psychology, where buyers from a company overreact to conditions where shops have stock-outs or excess inventory (Christopher, 2005). Whenever companies face this effect, they tend to over-buy after a stock-out problem and under-buy after an excess problem (Ayers, 2001). Either way, this reaction has always resulted in the opposite problem occurring. Without the adoption of adequate management techniques to mitigate the demand uncertainty, the SME food industry faces this bullwhip problem. In other words, if demand uncertainty is not well mitigated, it leads to several problems in the SME food industry.
One of the most prominent problems facing industries when hit by the effects of demand uncertainty is increases in production costs (Ulrich, 1995). The food industries can face this cost in the event there is an excess in stock. Here, when SMEs overestimate demand, they store more than what is needed. The food items then deteriorate due to its natural obsolescence, thus costing the company providing food services. Therefore, the SME food industry needs to put in place proper measures of mitigating the effects resulting in such costs. As a second effect, long lead times also can result from demand uncertainty in the food industry. Whenever uncertainty is not mitigated, the players in the industry suffer the effects of long lead times. Poor supply chain management practices lead to the elongation of the time between the placement of order by the customer and the delivery of the same. Long lead times result in customers having little confidence in the business (Christopher, 2005). Aside from the problems posed to customers, suppliers also pose a problem in long lead times in the supply chain. Demand uncertainty therefore leads to disconnection in communication between the seller, the buyer and the supplier of raw materials (Angerhofer et al., 2000). As another outcome, poor quality service provision may be experienced as a result of demand uncertainty in the SME food industry. In the event that the company cannot precisely predict the demands of the customer, stock-outs will occur. This means management will need to provide short-term solutions to the demand experienced during peak periods. When short-term solutions are offered, quality of service is compromised. The success of the goods and services industry in present times is determined by the level of customer satisfaction exhibited (Ulrich, 1995). With a poor supply chain management system, demand uncertainty effects cannot be mitigated, which in turn leads to compromised service quality (Prater et al., 2001).
Despite the shortcomings of demand uncertainty during peak times, SME food industries in Saudi have been on the forefront to initiate the management systems deemed useful in mitigating demand uncertainty. The industries have put in place several measures to deal with demand uncertainty, especially during Hajj, when demand is at its peak. The implementation of a proper supply chain system and the management of such means that the movement of materials, funds and information across the system is done in a correct way in order to avoid excess or stock-out. The management of supply chain provides the real time analysis of the flow of products and information across the entire system (Cohen & Roussel, 2005).
This research paper analyses three supply chain management methods adopted by the SME food industry in Saudi in an effort to mitigate the risks and effects. The paper analyses Supply Chain Integration, mass food customisation and postponement practices, as adopted by the SMEs food industry in Saudi Arabia. This study will centre on the Hajj season in the country. Mixed methods will be applied, where primary qualitative data will be gathered to augment the secondary data.

1.4 Research Scope
Uncertainty arising from the supply chain is an issue every practicing manager struggles to control (Simangunsong et al., 2011). According to these authors, this uncertainty arises from the complex nature of globalised supply chain networks, and includes potential for quality issues over supplying and delays in delivery. Although supply chain uncertainty arises from various sources of the chain, including internal manufacturing and supply-side processes, this study focuses on the uncertainty arising from demand-side issues (commonly end-customer demand). According to Simangunsong et al., the management of the supply chain is key to mitigating this type of uncertainty. These authors argue that inadequate risk management practices and policies can have severe repercussions on the performance of the organisation. Enhancing understanding of uncertainty and its management therefore remains an important consideration in today’s competitive and dynamic market, which is characterised by many challenges and changes that continue to unfold in this IT-driven global arena (Simangunsong et al., 2011). This is particularly important to SMEs in the food industry in Saudi Arabia, which are required to meet and satisfy the varying demands of customers during Hajj without compromising on cost and quality, whilst also ensuring they remain profitable.
In mind of the above, the scope of this study therefore is centred on analysing three supply chain management methods adopted by firms in the SME food industry in Saudi during peak times in an effort to help mitigate the risks and effects arising from demand uncertainty. The study will be limited to analysing Supply Chain Integration, mass food customisation and postponement practices, as used by SMEs in Saudi’s food industry. The study also will focus on the Hajj season, as this is the time when the industry faces demand uncertainty as a result of many customers from different parts of the world with different expectations concerning the taste and quality of food, and how it is served.

1.5 Research Question
How can SMEs in Saudi’s food sector mitigate demand uncertainty and increase their sustainability and competitiveness against larger firms operating in the sector through the adoption of managerial strategies, including Supply Chain Integration (SCI), Postponement (PP), and Mass Customisation Capability (MCC)?

1.6 Research Aim and Objectives
The aim of this study is concerned with investigating how SMEs in the food sector can mitigate demand uncertainty in an effort to increase their sustainability and competitiveness against large firms operating in the sector through managerial strategies by identifying the relationship, and accordingly examining and validating the impact of Supply Chain Integration (SCI), Postponement (PP) and Mass Customisation Capability (MCC) on mitigating Demand Uncertainty (DUM).
The purpose is to establish how all of these factors reduce costs and accordingly satisfy customers through quick response to demands.
In order to achieve this, the following objectives will be explored:
1. To review the literature in mind of developing and validating a theoretical model that establishes the effect of Supply Chain Integration (SCI), Postponement (PP) and Mass Customisation Capability (MCC) on Demand Uncertainty (DUM) under high competitive intensity.
2. To conduct semi-structure interviews in an effort to validate the conceptual framework and accordingly design a questionnaire instrument for gathering relevant data.
3. To apply appropriate statistical analysis so as to examine the developed model.
1.7 Research Gaps
Supply chain uncertainty is one of the issues with which every business manager struggles, stemming from today’s global supply networks that are quickly becoming increasingly complex (Simangunsong, 2011). These complexities include increased potential for quality problems, as well as delivery delays. Such uncertainties characterising complex networks are a vital problem and therefore important to understand. Whilst extensive research has been conducted in an effort to identify the specific factors giving rise to supply chain uncertainty, most have focused on the manufacturing processes and supply-side processes, leaving demand-side uncertainties (end-customer demand issues) under-researched (Simangunsong, 2011, p. 4494).
Research on the mitigation of demand uncertainty amongst SMEs is also scarce. Many sources of demand uncertainty, as well as mitigation through supply chain management, have not received sufficient attention (Simangunsong, 2011, p. 4494). Specific research on SMEs in the food industry that operates during Hajj season in Saudi Arabia is yet to be conducted; however, this is a key area significantly contributing to the economic growth of the Kingdom. Studies show that demand uncertainty is the most severe type of uncertainty in the supply chain, resulting from volatile demand and inaccurate forecasts (Currie & Shalaby, 2011, pp. 117–118). Volatile demand and inaccurate forecasts are likely to characterise SMEs that operate in supplying food during the Hajj season, yet studies on how such firms manage these risks are non-existent (Currie & Shalaby, 2011, pp. 117–118). Accordingly, this study will extend literature on management practices that can be used by SMEs in the food industry (in general, as well as in Saudi Arabia in particular) to mitigate demand uncertainty, hence filling the existing gap.
This study also observes another gap in the literature, as identified by Simangunsong et al. (2011, p. 4494), who recognise that, thus far, no attempts have been made to establish a comprehensive understanding of the various sources of uncertainty and how these can be brought into line with management strategies so as to improve supply chain performance. Lai et al. (2012, p. 453) also recommends the need for future studies in exploring the influence of Supply Chain Integration on manufacturing practices and strategies, such as modularity, product differentiation and postponement, which in turn have an effect on MCC development and the overall application of this strategy in a dynamic business environment. The objective of the current study is to fill these gaps in the context of SMEs in Hajj. This study will also form a basis for future research by identifying areas necessitating further inquiry.

1.8 Research Motivations
This study is motivated by the fact that SMEs in the food industry in Saudi Arabia face volatile demands and are likely to make inaccurate forecasts during the Hajj season as the volume of customers and their desires are difficult to predict. Based on the fact that demand uncertainty has the most severe effect on an organisation, there is a need to investigate how such firms can address demand uncertainty mitigation through management practices, allowing them to satisfy all customers visiting Mecca to attend Hajj, whilst at the same time leveraging their performance. Besides providing insight on management practices that can be used by practitioners in the industry, this study is also motivated by the fact that previous studies focused on SMEs in Saudi Arabia—and specifically on those operating in the food industry—are unavailable. The findings of this study therefore will contribute not only to practice but also to research and the overall body of knowledge on the mitigation of demand uncertainty through management practices for SMEs in the food industry. Further, the study is also motivated by the need to provide recommendations to SMEs on how they can mitigate demand uncertainty during seasons of high customer demand.

1.9 Definition of Key Terms
Supply Chain Management: Wong, Arlbjørn & Johansen (2005) define ‘supply chain management’ practices as ways of managing integration and coordination of demand, supply, and their relationships, with the aim of providing customers with satisfaction.
Supply Chain Uncertainty: This term is recognised as broader than ‘supply chain risk’, and refers to uncertainties (risks included) that may arise at any point within the supply chain network (Simangunsong et al., 2011, p. 4493).
Supply Chain Integration: Lai et al. (2012, p. 443) define Supply Chain Integration as the degree to which a firm strategically collaborates with its partners in the supply chain and accordingly manages inter and intraorganisation processes in an effort to attain the effective and efficient flows of information, services, money, products and decisions, with the goal of providing maximum value to its clients.
Postponement: Can (2008, p. 6) defines ‘postponement’ as the process of delaying product finalisation in the supply chain until orders from customers are received with the aim of customising products, as opposed to performing activities with the expectation of getting future orders.
Mass Customisation: Davis (1987) defines ‘mass customisation’ as a process where manufacturers tailor-make products to satisfy individual customer needs at the same prices as those of mass-produced items, whilst mass customisation capabilities are defined as the ability of a firm to offer a comparatively high volume of product alternatives for a comparatively large market that demands customisation without significant trade-offs in quality, cost or delivery (Lai et al., 2012, p. 443).
Mass Customisation Capabilities: Lai et al. (2012, p. 443) define MCC as the ability of a firm to offer a comparatively high volume of product alternatives for a comparatively large market that demands customisation without significant trade-offs in quality, cost or delivery.
Competitive intensity: According to Luhmann (2005), the competitive intensity of any organisation is affected by several supply chain factors. All of these factors help in demand uncertainty mitigation. Under Supply Chain Integration, there is customer integration, supplier integration and internal integration. All of these affect the postponement practises and Mass Customisation Capability. In turn, these help in the mitigation of demand uncertainty.
Demand uncertainty: Demand uncertainty is defined by various scholars as variations and fluctuations in demand (Chen & Paulraj, 2008; Lai et al., 2012).
Demand uncertainty mitigation: According to Amit et al. (2005, pp. 4236–4237), demand uncertainty mitigation can be defined as those actions reducing the adverse effects of the outcome of activities associated with the demand-side of the supply chain.
1.10 Outline and Organisation of the Thesis
This paper is organised into seven individual chapters. Chapter 1 provides an introductory section, delivering an overview of the research problem, its background, the research question, aim and objectives, its scope, the gap it seeks to fill, and the general motivation for its undertaking. It also defines the key words to be used in the thesis. In a nutshell, this chapter provides insight into the nature of the research and how the study will be conducted.
Chapter 2 provides a literature review, with emphasis placed on discussing existing literature relating to demand uncertainty and how supply chain management practices can be used in order to ensure its mitigation. The section reviews previous studies and findings on how management practices, such as Supply Chain Integration, postponement and mass customisation capabilities, can be used in demand uncertainty mitigation. Studies discussing the relationship between these practices are also reviewed. Moreover, this section identifies existing gaps in previous studies and how the present study seeks to fill them.
Chapter 3 discusses the theoretical framework. In this section, the theories and concepts guiding the study, and from which the study is developed, are discussed. The contingency theory and the Resource Based View (RBV) theory, along with their application in Supply Chain management, are considered. The contingency theory suggests that the most appropriate management strategy in a particular context depends on a set of contingency factors, possibly including uncertainty of the environment, whilst RBV suggests firms can gain sustainable advantage by developing and acquiring infrastructural resources, as well as knowledge-based capabilities that are difficult for competitors to replicate. This section discusses how the two theories complement one another in explaining organisational performance. The adequacy of the contingency theory and RBV in management decisions regarding strategies for improving firm performance is discussed and related to SC practices, namely SCI, PP and MCC in mitigating demand uncertainty. A conceptual framework illustrating the relationship between the main constructs is also highlighted and hypotheses developed based on the framework.
Chapter 4 outlines the methodological framework and research paradigms used to conduct the study. This section also provides justification for each method adopted. Methods used for data collection, including survey questionnaire, semi-structured interviews and archive data/document review, will be illustrated. The sample selection process also will be described, along with analytical procedures for both qualitative and quantitative data.
CHAPTER 2: LITERATURE REVIEW

2.1 Introduction
Increasing demand uncertainty has affected the operations in supply chain management. Supply chain management, in turn, affects the inventory management in any business. Inventory control and demand uncertainty are closely interwoven. Technology is rapidly evolving, intensity of competition is increasing, supply chain is becoming more complex and the market is facing turbulent times. All pf these factors have resulted in increasing demand uncertainty in industries ranging from cars and electronics to chemicals (Prater et al., 2001). Demand uncertainty can only be controlled if adequate supply chain management is put in place. The literature review will seek to look at the literature available on the subject; this section provides definitions of demand uncertainty, supply chain management and supply chain uncertainty. The literature also explores different management practices that can be used in order to reduce the effects of demand uncertainty to a business venture. Supply chain integration, postponement practice and Mass Customisation Capability all have been discussed under this section. The discussions will be mainly based on past scholarly articles that have been written on the subject, as well as different texts, which will shed insight in this area.

Figure 1: Generalised Supply Chain Diagram
One of the papers that provided invaluable information on the subject of uncertainty is that written by Eliot, Hendry & Stevenson (2011), which analyses uncertainty in the supply chain, the main cause of which is the increased complexity of the supply chain with the passing of time. An understanding of the workings of the supply chain is necessary in order to gain understanding of some of the concepts been discussed by Eliot and his colleagues in their paper. These concepts also have been discussed in the text by Kouvelis (2007) in which he defines supply chain, contending that the complexity in these supply chains may be as a result of increased globalisation. The effect has been creating a greater interlink between different companies and different supply chain (Kouvelis, 2007, p. 5). One major contribution of the paper is in defining supply chain uncertainty, which is done by providing examples of where uncertainty in the business may be experienced. Eliot also created a distinction between uncertainty and risk (Eliot, Hendry & Stevenson, 2011, p. 4493).
In a different paper by Field et al. (2006), the impacts of uncertainty on the performance of a business are investigated through the completion of a study conducted on 108 financial service employees. The paper begins by providing a discussion, along with the identification of the fact that uncertainty has negative effects on the company (Dixon, 2004, p. 169). It is cited that amongst these effects is a reduction in company productivity, along with reduced customer satisfaction. Field and his colleagues provided a greater understanding as to the definition of uncertainty by expounding on the current literary definitions available. In their paper, empirical evidence is used to provide the findings aimed as a guide to managers on the best way of mitigating the negative effects arising from uncertainty in the workplace. The paper also introduces the concept of uncertainty coping (Field et al., 2006, p. 149).
In consideration of the impact of uncertainty on a firm’s performance, which increased the validity of the study, attention was also directed towards examining the issue from a managerial perspective through the use of primary data. This is able to provide a connection between company performance, uncertainty, and the decision-making process of a company amongst different companies. This analysis also provides the necessary data so as to form conclusions that inform the manager of best practices when it comes to issues having to deal with the application of different thoughts in management. One issue addressed by the paper, which aids in the study of uncertainty, is the suggestion that the complexity of the processes created in trying to achieve mitigation have an effect on the company’s performance (Field et al., 2006, p. 149).
A different paper by Meydanoglu (2009) considers this issue of the complexity in the supply chain. Meydanoglu’s paper begins by providing a discussion on the various factors that increase the complexity of supply chains. The paper is written with the point of view that this increase in complexity affects the predictability of the supply chain. This has an influential role on the management of the supply chain (Gattorna, 2009, p. 190). It is further contended that this raises the need to create supply chains that are more resilient in their operations. In his proposition, he sought to use a computer system that would be used in the management of risks associated with the supply chain. This system is referred to as the Supply Chain Event Management (SCEM) system. From Meydanoglu’s consideration, an understanding is provided as to the importance of information technology systems in the business environment. This provides an indication as to some of the systems being used to improve on the efficiencies created in the supply chain (Pegel, 2005, p. 31). The study, however, considers only the theoretical framework of the application of the system without any supportive empirical evidence (Meydanoglu, 2009, p. 1).
Another point of interest created in the work of Field et al. (2006) is the realisation that there are differences between manufacturing and service industries. These differences have an influential role to play in the overall effectiveness of the modes applied in the management of uncertainty. Amongst the differences cited are the direct contact the service industry has with customers, and the lack of inventory in service industries, which together limit their use in buffering uncertainty. They also highlight the scope of the different management practices that go into managing uncertainty. Arguing that there are narrow-scoped frameworks and wide-scoped frameworks, depending on what is to be addressed by each of these frameworks, the limitation in the study of Field et al., as in others, is that the data is collected from one particular region with the use of single respondents within the particular firms chosen. This could have created bias in the pool of data collected (Field et al., 2006, p. 149).
Eliot, Hendry and Stevenson (2011) developed their analysis with the use of secondary research, notably by using literature already available on the issue. This analysis enables others to gain understanding into the direction of thought accompanying the study of supply chains and their management (Wang, Heng & Chau, 2007, p. 220). The paper by Eliot provides an invaluable contribution in the sense that it details information used in the development of the theoretical framework of this paper. Importantly, two distinctions that may be applied in the general consideration of the supply chain are made, namely the supply-side processes, comprising those involved in internal manufacturing processes, and the demand-side issues, comprising those concerning the client (Eliot, Hendry & Stevenson, 2011, 4493).
There are various areas of uncertainty apparent within the firm that have been discussed in the paper by Eliot and his colleagues (2011). Discussion has centred on the development of a theoretical framework that considers the information identified in previous researches. They write on what other research papers have identified as areas of uncertainty, and then are able to discuss various strategies that may be relevant in the management of the supply chain. The advantage provided by the paper is that there is a relative comparison of the different models that were developed in an effort to mitigate supply chain uncertainty. The main limitation of the paper is in the consideration of past research (Sapsford & Jupp, 2006, p. 142). There is no empirical evidence that has been provided, thus far, in mind of testing the theories proposed by the writers. The paper calls for further empirical research on the subject to allow for the study of the adaptation of the given theoretical models in the management of supply chains (Eliot, Hendry & Stevenson, 2011, p. 4493).
A different paper by Huo (2012) provides a greater understanding of what supply chain management constitutes. The approach of the scholar is comparable to that of Eliot et al. in the sense that he considers the definition that has been provided by previous research papers. He is able to expound on the knowledge of how the management of the supply chain may be considered by taking into account what he terms ‘systematic and strategic coordination’, indicating that the incorporation of these two concepts into the functions of the supply chain is integral (Huo, 2012, p. 596). Huo also provides a deeper understanding of Supply Chain Integration by improving on the discussion in regard to the decision-making process in the management of supply chains. The paper introduces the concept by reviewing the limitations apparent in previous research. Huo further works to dispel some of the misconceptions surrounding the management and integration of supply chains, with the paper also introducing the concept of organisational capability and demonstrating its connection to the resource-based view—a connection important in understanding organisational capabilities (Henry, 2007, p. 126).
Huo (2012) has examined the various dimensions constituting the relationships in Supply Chain Integration, and also sought to examine the effect of some of these dimensions on the performance of the company. The paper by Huo continues by defining what organisational capabilities are in the sense of the firm’s ability to undertake tasks directly or indirectly, related to the value addition to the goods or services offered by the company (Mourdoukoutas, 1999, p. 68). This is an integral part of the consideration of the integration of the supply chain, which is important in the management of supply chains (Winser, Tan & Leong, 2011, p. 445). The paper moves forward to look at the various strategies explored in organisational capabilities.
The research by Huo (2012) has provided an additional dimension by directly seeking to link the management of the supply chain with the financial gains made by the company (Drury, 2005, p. 479). This is whereby different companies have different metrics for measuring performance. The use of a specific metric is a choice left to the discretion of the managers of the company (Harbour, 2009, p. 26). As with most other researches that have been analysed, the paper is limited in the sense it considers only data from firms in China. One other limiting factor in terms of the external validity of the research is that the suggestions proposed in the paper majorly affect manufacturing but no other industries. However, the paper provides invaluable insight that may be used in future research and in the study of supply chain management (Huo, 2012, p. 596).
As a supply chain management technique, integration has been most valuable in cases where mass customisation is adopted; this may be described as a management technique that is applied mainly in areas where there is a need to limit demand-side uncertainty (Abdelkafi, 2008, p. 217). Zahg, Lee & Zhao (2012) conducted a study that sought to consider two issues as regards integration and mass customisation: the first was how mass customisation was influenced by the integration of suppliers, customer integration and internal integration; the second centred on examining whether certain environmental factors limited the effect of the integration of the supply chain. The authors provide insight and knowledge concerning mass customisation, and further consider its inclusion as a supply chain management technique (Zahg, Lee & Zhao, 2012, p. 443).
Zahg, Lee and Zhao (2012) contend that an increase in competition amongst various industries and companies necessitates the need for the customisation of products in order to ensure successfully competition. This is regarded a differentiated strategy (Carneiro, 2012, p. 6). According to this author, companies need to do this with minimal cost in order to improve their performance. In their paper, Zahg et al. (2012) point out that, in order for companies to gain a competitive advantage, they first must consider the integration and reconfiguration of both external and internal resources. They contend that the implementation of the steps outlined in these two strategies will need to be undertaken in a way that adapts to the changing business environment (Kew & Stredwick, 2005, p. 221). They used what they termed as ‘the extended resource-based view’ in an effort to develop a model that was used to study the economics of mass customisation and the integration of the supply chain (Zahg, Lee & Zhao, 2012, p. 443).
Zahg et al. (2012), through their study, were able to demonstrate how internal and external integration may be utilised so as to achieve greater capabilities by companies. These thoughts contribute to the thinking of how Mass Customisation Capabilities (MCC) may be effectively achieved by companies. Zahg and his colleagues are able to indicate that internal integration and customer integration create the greatest effect on MCC. Moreover, the scholars further came to the conclusion that the application of integration on the supply side has minimal effect on MCC according to their examination, whether directly or indirectly. They also show that uncertainty on the demand, in addition to the intensity of the competition, may work towards negatively affecting the overall ability of the company to benefit from mass customisation. Although there were limitations in the undertaking of the research, the paper still provides knowledge that is valuable in the examination of Supply Chain Integration and its effects on mass customisation strategy (Zahg, Lee & Zhao, 2012, p. 443).
A different paper by Huang, Kristal & Schroeder (2010) investigated the effectiveness of mass customisation be to a company by considering the influence of the processes adopted by the company. There are many different models that may be used within the strategy of mass customisation; these models have an influential aspect on the efficacy of mass customisation in mitigating demand uncertainty (Chandra & Kamrani, 2004, p. 24). There are three aspects Huang and his colleagues considered when looking at the mass customisation strategies: flatness, employee multi-functionality and centralisation; flatness refers to an organisational structure with a short chain of command due to limited managerial level; centralisation refers to a having decisions made at higher consolidated levels and based on amassed knowledge and information (Chandra & Kamrani, 2004, p. 24); employee functionality, on the other hand, refers to the ability of employees to perform multiple functions in mind of the skills they possess. The paper approaches the issue from a contingency theory perspective, which is one of the theories that may be used in the analysis of management strategies (Markman & Phan, 2011, p. 475). In their examination, Huang et al. (2010) considers two forms of customisation, recognised as full customisation and partial customisation (Huang, Kristal & Schroeder, 2010, p. 515). The research paper has increased external validity as it was conducted using a significant number of respondents. This respondent pool was taken from different manufacturing industries and different regions, thus allowing for the generalisation of the recommendations and the drawing of conclusions into other areas of the world.
In their presentation Huang, Kristal & Schroeder (2010) present the various theories currently available in the field of mass customisation. These theories are used to build upon the idea of mass customisation, and further present a number of real world situations in which this strategy is applied. The advantage provided by this is by enabling for the examination of the extent of applicability of the mass customisation as a theory. This allows for the analysis of the issues faced by companies in their implementation of mass customisation (Chandra & Kamrani, 2004, p. 92). Huang et al. (2010) also point out that most of the literature available on the subject of mass customisation has been limited to examining the structural and infrastructural, and the implementation of strategies as regards MC.
It is important to understand the difference between customisation and mass customisation. Mikkola & Larsen (2013), as an example, describe customisation as a continuum that encompasses five different classes, namely pure standardisation, segmented standardisation, customised standardisation, tailored customisation and pure customisation. The authors also identify four different approaches to customisation, including collaborative, adaptive, cosmetic and transparent. On the other hand, mass customisation involves customer co-design procedure of products and services, which satisfies the needs of all personal customers with consideration to specific product attributes. Consequently, its operations are carried out in a fixed solution space, exemplified by stable, flexible and responsive procedures (Can, 2008).
Huang et al. (2010) point out that past researchers have not paid much attention to the effectiveness of given MC strategies, which is what they sought to consider in their research (Huang, Kristal & Schroeder, 2010, p. 515). They are also able to show that these researches instead focused on the theoretical aspects of the study. In their study, they analysed 167 manufacturing companies in order to gain perspective relating to the empirical evidence. This is able to fill the gap in terms of the empirical evidence in mass customisation, accordingly building a foundation that allows one to conduct an explorative research (Moser, 2006, p. 66). The study has been built on previous research in the field of MC, and has been able to create links between organisation theory and operations management in the development of the conclusions and discussions; these validate the study by Huang, Kristal & Schroeder (2010). They indicate that, in order to achieve the best results from mass customisation, companies needed to utilise organic structures, and further indicate that this was effective for companies that adopted full customisation as opposed to partial customisation. They also concluded that there are other internal structures that may have a negative effect on the benefits potentially obtained from mass customisation (Huang, Kristal & Schroeder, 2010, p. 515).
A different paper written by Govindan et al. (2012) expounds on the knowledge with regards to the supply chain. Govindan indicates that disruptions may create a systemic effect over an entire industry; this is from the interconnected nature of the current industries. This interconnected nature has been a factor in the concept of outsourcing that has gained increasingly more prominence in the recent past (Wagner & Bode, 2009, p. 198). The discussion that Govindan has presented also contributes to knowledge towards understanding the increased concern directed to uncertainty that is experienced in the supply chain. It states that the effect may be disastrous as a result of the shorter product lifecycles, shorter technological cycles, and the complexity of the supply chain caused by the increasing use of distribution and logistics partners. These have also increased the risks eminent in the supply chain. This study classifies this risk into three main categories: demand risks supply risks and operational risks (Govidan et al., 2012, p. 3039).
A positive argument provided by the paper written by Govindan et al. (2012) is that it presents a view of the risks associated with the food industry. This has been done with the creation of models that look at the most pertinent risks potentially affecting food supply chains. It has also tried to establish a link between the interactions between the different risks identified. The paper has also studied the different ways in which these different risks affect the management decisions adopted by the company. The paper has also presents various classical methods that may be used in the management of risks. The paper has limited its analysis to a single company, which would limit the overall propensity to generalise the findings to a greater number of companies (Govidan, Diabat & Panicker, 2012, p. 3039).
The aim of this chapter is to provide a greater degree of understanding of the issues already addressed in the introductory section. This will be achieved with consideration to the various literature available on the different issues to be addressed in the paper. Examination will mainly focus on the issues discussed in previous research papers, with other literary texts playing a guiding role. The literature review will move on to consider demand uncertainty, as well as how other papers have sought to address the issue. This discussion then will move on to consider the management of supply chains and also the different management strategies that may be applied in management. This then will look at Supply Chain Integration and postponement strategies in the mitigation of demand uncertainty. The literature review will move on to look at the discussions that have gone into considering mass customisation in companies.
2.2 Background of Al Hajj
According to Long (2012), approximately 2 million Muslims perform the great Pilgrimage to Makkah (Hajj) each year. Moreover, around 58% of Hajj Pilgrims are from overseas, representing different tastes, needs, expectations and demands, especially in terms of food (Kaysi et al., 2010). Hajj season accounts for approximately 70% of Mecca’s total annual revenue and further accounts for approximately 3% of Saudi GDP. Saudi Arabia gained a reported of $16.5 billion from religious tourism in 2012. By 2030, it expects to witness an increase in the number of visitors at Makkah city during pilgrimage by 40%. The religion requires all Muslims visit once in their lifetime for as long as they are mentally, physically and financially able. For this period, the Saudi government is required to ensure that all pilgrims have adequate housing (usually tents), food, health, water, sanitation, public safety and security, as well as ground transport (Long, 2012). This author reports that Hajj has significant administrative, social, political and economic impacts on Saudi Arabia. However, the present study will mainly focus on the economic impact of the Hajj.
Long (2012) explains that, prior to the discovery of oil, Hajj acted as the economic backbone of the KSA. With the vast oil wealth and the revenues generated as a result, however, the Saudi government is no longer dependent on revenues from Hajj. However, Hajj remains a major source of income for the Kingdom’s private sector. Long explains that, besides the Hajj service industry, Hajj also is a major season for the consumer industry as well, similar to the Christmas season in Western countries, such as the United States and the United Kingdom. Those celebrating Hajj, particularly from developing countries, buy items that are not available or are highly taxed at home, including luxury items, such as jewellery and perfumes, as well as medicines. Hajj also creates employment for young Saudis, such as the young men who are usually trained and hired to accompany the Hajjis as they make the sacred journey.
Most significant is the impact of Hajj on the private service industry. According to Long (2012), Hajj administration, for many centuries, was controlled by ancient, family-organised associations that arranged food, transportation and lodging, and also guided the Hajjis through the religious rites and guided them to Al-Madinah. Since Hajj was recognised as the backbone of the Kingdom’s economy, the guilds took advantage and charged the pilgrims whatever they could bear. However, the Saudi government, which is the responsible custodian of the two Islam Holy places, took this responsibility very seriously, and now strictly regulates what guilds charge in an effort to ensure Hajjis are not overcharged. Presently, the guilds function more as public utilities. The responsibility of providing Hajjis personal care now rests with the Mutawwifs; in essence, they act as religious tour guide agencies for designated nations of origin. They look after the Hajjis from the time they leave for Saudi Arabia to the time they go back home again (Long, 2012).
Besides the guilds, the Hajj service industry has grown to include other private sector enterprises that provide catering services, transport and accommodation. This study is focused on SMEs that provide food services and how they manage demand uncertainty through management practices of supply chain.

2.2.1 Hajj Pilgrims
Currie & Shalaby (2012, pp. 117–120) state that Hajj is a yearly pilgrimage, and it is vital for those who practice the Muslim religion to visit at least once in a lifetime. The authors go on to point out that Hajj takes place over a period of six days, spanning the 8th to the 13th day of Thul-Hijjah (TH); consequently, the 12th and the last month of the Islamic calendar. According to this article, we are able to see that Umrah—the lesser pilgrimage, as they term it—constitutes visiting the Grand mosque only, and can be done at any time of the year. Pilgrims can choose to conduct their Umrah separately or collectively with the Hajj. It is noted that the times for Umrah are mainly before Hajj and during Ramadan, which normally occurs two months before Hajj.
The paper by Currie & Shalaby (2012) also identifies that, within the period of Hajj, a good number of people in Makkah are normally visitors, constituting a population of approximately 4.2 million or so in some years, with Hajj people known to travel more than residents. The authors state that research shows that most Hajj pilgrims travel in groups, which are basically known as Hamla. The Hajj pilgrims travel with a selected guide managed by their home region, together with managed accommodation. Similarly, it is concluded that, since Umrah occurs throughout the year, it is not that strict. The paper also elaborates that Umrah is carried out continuously, thus experiencing its peak during Ramadan.
2.3 Food Industry during the Hajj Season in Saudi Arabia
The great pilgrimage to Makkah, the Holy City of followers of Islam, is regarded as Hajj. Considered as an essential and traditional visit for all believers of Islam, the Hajj is one of the five pillars of Islam. It is mandatory for all believers who are physically, mentally and financially capable to visit at least once in their lifetime. Accordingly, each year, millions of Muslims perform the Hajj in Saudi Arabia (Long, 2014). Saudi Arabia secured $16.5 billion revenues from Hajj in 2012, representing 3% of its GDP. Given the worldwide participation and its sheer size, consistent food supply is one the fundamental essentials during the period of the Hajj season (Rashid, 2012). Saudi authorities forbid perishable food items, and allow only limited quantities of packaged and canned food into the country. A variety of fruits and dairy products, along with water, are widely available across Saudi Arabia during the Hajj season, with the exception of poultry and other meat products (Mousa, 2013). Hundreds of kitchens are spread across Mina and distributed amongst the tents hosting the pilgrims (Information Office, 2013). The pilgrims also rely on restaurants and other catering companies for their meals. Food is sourced via domestic/local and/or foreign channels through a well-connected market structure (Mousa, 2013). The diagrammatic representation of the Saudi food service industry is outlined in the following figure, which is exclusively adapted with slight modifications from the flow chart devised by Mousa (2013).

Figure 2: Diagrammatic representation of Saudi’s food service industry (Adapted from Mousa, 2013)

2.4 SME Development
The population of Saudi nationals is expected to reach almost 30 million by 2020. The Saudi labour force will expand to 8.8 million by 2020, increasing from 3.3 million in 2000. A total of 46% of the total population in 2001 were aged under 14 years old, whilst 74% of the total population are under 29 years old. With this noted, the Kingdom needs to provide these young people with appropriate job opportunities Otsuki (pp. 1–2).
Shalaby (2012) defines small enterprises as those firms that have up to 20 workers, whereas medium enterprises as those with 21–100 workers, as provided by the Chamber of Commerce and Industry. According to this definition, both small- and medium-sized enterprises must have a capital of less than 15 million Saudi riyals.
The global economies’ increased liberalisation during the past decade has had a significant impact on the expectations of customers and on competition for local and foreign business entities, hence paving the way for opportunities of SMEs (Amosalam, 2008, p. 5). According to a report by Capitas (n.d.), SMEs in Saudi Arabia are the key to unlocking its economy’s vast potential.
Alarape (2007, p. 4) reports that there has been a growing and increasing interest in Saudi’s small business sector throughout the course of the last decade. These authors attribute this growth to economic liberalisation policies implemented by the Saudi government in compliance with the requirements of the World Trade Organisation. Bundagji (2005, pp. 2–3) reports that small- and medium-sized enterprises (SMEs) form up to 90% of Saudi’s private enterprises. They are the major source of the Kingdom’s private sector investment, and as such, form the primary source of employment, alleviate poverty, develop new products, promote the innovation of new technologies, encourage entrepreneurial culture and are expected to contribute more than 50% of Saudi’s total industrial production in the near future. However, Al-Awwaad (2007, pp. 2–4) observes that, in spite of their importance, the development of SMEs in Saudi Arabia remains slow with a high failure rate. Studies claim that up to 80% of SMEs in Saudi fail within their first five years, whilst 80% of the remaining ones fail within the following five years, thus implying that 96% of SMEs in Saudi Arabia fail within the first ten years of their establishment. The report further explains that, although SMEs have a strong footing in the Kingdom owing to the fact they form 90% of all businesses, their contribution towards GDP remains low. This is an area of concern into which research should investigate in order to establish causes of failure and how these can be prevented. The current study contributes to this by exploring ways through which SMEs can manage supply chain uncertainty.

2.4.1 Factors Influencing the Development of SMEs in Saudi
Despite the significance of SMEs in the recent world economy, these firms continue to be influenced by various factors that inhibit their growth. In his discussion of challenges that face SMEs in Saudi Arabia, Shalaby (n.d., p. 8) identifies inability to gain financial access as the main challenge, and further identifies dependence on foreign resources—particularly foreign manpower and technicians—as a disadvantage not only to the firms but also to the locals as they do not benefit from new jobs created by the firms (p. 9). Limited market skills is also another limiting factor contributing to the failure of SMEs in Saudi. According to Shalaby, most Saudi SME managers do not have formal training in marketing or in product distribution; thus, they are unable to take advantage of economies of scale in their operations, hence the tendency to undergo high costs. The study also found that most SME managers in the manufacturing sector are not the owners, whilst entrepreneurs were found to manage their own enterprises. This study established a lack of commercial knowhow, legal status, feasibility study, exhibitions and training needs assessment as some of the areas requiring focus in order for SMEs in Saudi to be more successful (p. 11). According to Darling et al. (2007, p. 4), the major factors influencing the development of SMEs include finance, operational setbacks, the existence of good quality staff, limited communication with regulatory authorities, such as national chambers of commerce, and necessities on the regulatory operations of SMEs. However, these authors strongly point out that finance, to a great extent, has been viewed as a major factor influencing the development of SMEs. However, it is presumed that governments have come up with programmes and measures in an effort to try and trim down the influence of finance on the growth of SMEs. Al-Awwaad (2007, p. 2) further emphasises the findings of recent research that demonstrate that poor communication amongst SMEs and the government, regulation and limited access to good quality staff are increasingly becoming factors that influence the development of SMEs in the Kingdom.
Additionally, Darling et al. (2007) suggest that regulations and procedural requirements involve business costs that are continuously incurred by SMEs. This eventually influences SMEs in terms of development due to costs, and may push them to informal sectors, which, as a result, also deters economic growth. Nevertheless, Al-Awwaad (2007) identifies that the government has a greater influence on the development of existing SMEs and not SMEs that are starting up.
In an attempt to salvage SMEs in the Kingdom, the Saudi government formed the SMEs Development Centre (SMEDEC) in 2004, and later in 2006, the National Competitive Centre (NCC) (Shalaby, n.d.). According to Shalaby (p. 1), the goal of the SMEDEC is to support initiatives that focus on improving the overall competitiveness of SMEs using members of the Commerce and Industry Chamber. Another government support initiative in support of entrepreneurs in Saudi Arabia is the Saudi Industrial Development Fund (SIDF), the objective of which is to support the development of the Kingdom’s industrial sector through funding. A review of the sector by the NCC indicated that entrepreneurial activity in the Kingdom was low to the extent that the government’s goals of increasing the role of SMEs in the economy were not going to be achieved (Al-Awaad, 2007, p. 2). Given the important role of SMEs in Saudi and the fact that their development is slow, and that failure is very high amongst them, the current study significantly contributes to reducing failure by providing empirical evidence on how uncertainty can be mitigated—particularly in the supply chain and on the demand side.
Capitas (n.d.) reports that, although many governmental initiatives and programmes have been established throughout the Kingdom with the objective to help address the challenges facing SMEs and accordingly facilitate their development, there is a need for a new SME support programme, as the existing ones seem to be ineffective. According to the report, the existing programmes are too many with the replication of roles and redundancy. The report recommends the establishment of a national SME Authority, which serves as a lead agency or steward, which develops SME-enabling policies, identifies SME needs, and accordingly coordinates and oversees the provision of intended services and programmes to the industry (Capitas, p. 2).

2.4.2 SMEs Supporting Systems in Saudi Arabia
According to Otsuki (2002, p. 1), SMEs make up a large part of entire business entities worldwide, and are one of the most significant drivers in the national economy. Amongst the majority of developed nations, SMEs support the nation’s economy by more than 80% of overall industrial production. The author states that SMEs in Saudi Arabia also are expected to play a significant role in the nation’s industrial and economic growth. Otsuki (pp. 6–8) identifies the supporting systems in Saudi Arabia as follows:
i) Government policy: Otsuki (2002, p. 1) states that lucid government policy systems on SMEs, as well as special incentives, are necessary when striving to support the development of SMEs in Saudi Arabia. Support from the government, as the writer points out, includes marketing skills, advanced technology, finance, human capital and taxation reduction on SMEs.
ii) Mobilising national committee on SMEs: According to Otsuki (2002, p. 1), the implementation of the mobilising national committee, which was formed by royal decree on April 17, 2001, supports SMEs by managing the ministries of finance, industry, national economy, commerce, SIDF and Saudi Credit Bank and SAGIA. In order to quicken the process, Otsuki proposes the following: first and foremost, the existence of a legal framework that incorporates the laws and regulations on taxes, labour, minimum wages and many others; the establishment of a government authority that supports SMEs; third, a financial support system for SMEs in Saudi Arabia that may comprise loan services through commercial banks; the provision of manufacturing and business facilities; fifth, training of management, technology and marketing teams for employees of SMEs; and finally, the establishment of a national association of SMEs to facilitate the development of SMEs in Saudi Arabia.

2.5 Supply Chain Management
Christopher (2005) regards supply chain as a link of networks, involving several companies operating across different processes that end up producing value in the form of services and products for their customers. Through this definition, upstream and downstream linkages prominently feature as the necessary elements of the supply chain to work properly. According to Mentzer et al. (2004), supply chain is a set of three or more entities that are directly involved in the flow of products, information and finances—on either an upstream or a downstream basis. Krajewski et al. (2007) give an alternative definition of the supply chain, considering this as the set of links involving suppliers of materials and services.
According to them, the link spans from the transformation of raw materials into services and products. For purposes of clarity, this thesis adopts the definition provided by Lambert, Stock & Ellram (1998), which states that the supply chain consists of interdependent companies acting together in order to control, improve and accordingly manage the material, information, product and service flow from the point of origin to the point of delivery. The main aim of the linkages is to provide satisfaction to the needs of customers whilst also ensuring all involved incur the lowest possible costs. According to Lambert & Cooper (2000), supply chains are different and vary from one company to another. A company manager may look at his firm as the main point of the company. This point of view is important because one company can belong to several linkages of the supply chain. Therefore, it becomes essential that the company understands its point of focus. Mills, Schmitz & Frizelle (2004) identify with this point of view.
According to the authors, four perspectives should be adopted whilst examining the supply chain, including upstream, downstream, dynamic network and static network. When adopting the upstream point of view, the purchaser needs to manage the relationship with the supplier. The downstream point of view indicates the need for a supplier to supply the customer with the correct product. Essentially, there are circumstances where the managers adopt the auditor role in the supply network. This is the static network, which typically comprises the analysis of the supply chains in which the company is involved when doing business. In a dynamic network, the manager works as a strategist: he looks for opportunities that can help to improve the position of the company in a network that already exists. Dynamic systems help in the creation of a strategic and long-term view. This set is prone to changes, where delivery frequencies and inventory levels can be changed (Angerhofer et al., 2000; Ramanathan, 2013). It should be noted that, despite the fact that all approaches to defining and analysing supply chains have been clearly explained, there is a lack of consensus in regard to how the state of the supply chain should be defined. Different authors have attempted to give their own definition. Kelton et al. (2003), for example, looks at the supply chain state of a system as a snapshot showing all relevant details. In considering the purpose of this thesis, the definition provided by Carvalho et al. (2012) can be adopted. This definition approaches the state of supply chain as a specific arrangement of the entities of supply chain and the related links dealing with information flows, management policies, lead times and performance level.
An examination by Li & Schulze (2011) indicates that an understanding of the various elements of the supply chain is needed if one is to develop the best strategies in their management. They begin their paper by introducing the various aspects of the supply chain, which involves raw materials, unfinished and finished products. They state that it is a process that starts with the suppliers, and progresses to include the manufacturing plants, distribution centres and, finally, retail stores. This is in line with the definition provided by Lambert, Stock & Ellram (1998), although Li & Schulze (2011) argue that, in developing management strategies, there is a need to consider all factors that affect the different areas of the supply chain.
The study carried out by Li (2008) adds to the definition of the supply chain; this is by defining it as a logistics network. Previous texts have looked at logistics network as part of the supply chain (Li, 2008, p. 215). Li considers that influencing factors that determine the logistic network would lead to better service for the customer. They provide a summary of papers that have been written in the past on the subject. The paper by Li & Schulze (2011) provides various significant time frames through which there have been developments in the way in which network designs may be done. Their consideration to past papers on the subject provides the ability to show that, over time, there have been simplifications made in terms of the assumptions made of the supply chain. This has made it simple to implement the strategies suggested under the logistics network (Li & Schulze, 2011).
Liu & Schulze (2011) contend that, in creating a flexible network, a company is able to better mitigate the risk factors accompanying uncertainty. In looking at the network of the supply chain, the most suitable strategy proposed is looking at the performance of different characteristics within the supply chain; this will be dependent upon the specific industry and the goals to be achieved by the company. The paper by Liu & Schulze further points to the fact that companies seek to gain a competitive edge over rivals through the management of their supply chain. This is a factor that has been identified in numerous other texts (Van Weele, 2005, p. 181).
A study by Vijayasarathy (2010) also considers the issue of the supply chain and how it may be managed. This study adds to the knowledge presented by Li & Schluze (2011) by considering the supply chain as a constituent of many different dimensions. Vijayasarathy further argues that supply chain management should involve the management of each of these dimensions. It also adds to the discussions by introducing what is referred to as ‘Supply Chain Integration’. This factor will be discussed later on in the literature review. The discussions are presented by a consideration of data from 276 firms (Vijayasarathy, 2010, p. 489).
The study’s consideration of the different dimensions of the supply chain provides a different look at how best supply chains can be managed. The presentation by Vijayasarathy (2010) provides a more in-depth understanding of the supply chain; this also aids in understanding the different strategies that may be applied in the management of supply chains. Vijayasarathy further provides evidence from previous research papers that have looked at the subject of supply chain management, with the author further demonstrating that each of the studies reviewed has idealised the supply chain as a combination of various factors. Vijayasarathy has also created one point in terms of integration that would require further study, which is in terms of the curvilinear relationship between integration and company performance. One area in which the study by Vijayasarathy falls short relates to the use of a limited number of respondents in the development of the conclusions. In the study, the response rate was that of 9%, compared to the higher rates of 15%–20% documented in other research methods (Vijayasarathy, 2010, p. 489).

2.6 Management Practices
Lambert, Stock & Ellram (1998) are some of the authors who have written in the area of supply chain management. In their work, they state that the success of supply chain management depends on the ability of individual companies in the network to overcome their recession and accordingly adopt an approach that is a process in nature. According to the study review conducted by Mentzer et al. (2000), supply chain management is looked at as the systemic and strategic coordination of the traditional functions of business and the tactics used across the businesses within particular companies. Wong et al. (2005) define supply chain management practices as ways of managing integration and coordination of demand, supply, and their relationships, with the aim of providing satisfaction to consumers. This satisfaction should be carried out in an effective and profitable manner. Supply chain practices have to be aligned with supply chain strategies; these are those that become paramount in mitigating the demand uncertainty effects during peak times (Cohen & Roussel, 2005).
In the study of the management of the supply chain, Li et al. (2005) chose to investigate the differences in the management of supply chains across different companies. Examination was carried out from the perspective of the competitive advantage firms are afforded, as has been done before by other studies (Wolf, 2008, p. 12). The paper developed knowledge on the issue of how a given supply chain can be managed. Li and his colleagues were able to identify six major supply chain management techniques, which included strategic partnerships with suppliers. This was where the company seeks to form long-term working relationships with their suppliers. The other was in the building of a good relationship with the customer so as to ensure a long-term and satisfactory relationship. They identified the need to understand the extent to which information is shared with suppliers. There is also the need to consider the quality of information. Companies might practice lean internal operations where wastage is kept to a minimum. There is also postponement involving the moving forward of certain operations. The study involved examining 196 organisations to which structural equations were modelled in order to validate the theories developed (Li et al., 2005, p. 618).
Li et al. (2005) are able to identify the reasons leading companies to consider the management their supply chains. They show that the management of supply chains by companies was a response to their need to improve the performance of the company and similarly improve the supply chain. They identify that, despite the recognition by a company executive concerning the importance supply chain management has on performance, the implementation and adoption of these strategies are poor. This was attributed to the complexity of supply chain management a factor attributed to playing an influential role in the strategies applied (Gattorna, 1998, p. 239). A notable aspect of the study conducted by Li et al. (2005) can be seen in terms of its ability to develop a conceptual framework using research papers that had been written earlier. This is considering that these previous papers were considering only specific elements of supply chain management. Therefore, the paper is able to connect upstream, internal and downstream supply chain points (Li et al., 2005, p. 618).
The authors state that the findings of the paper will be used in the development of future research, especially concerning the connection supply chain management has to other business areas. Li et al. (2005) conclude their study by considering the idea that the scope covered in the paper is limited in terms of the measures applied in their study of the measurement of the management of supply chains. Accordingly, they call for future research to include more measures. It also identifies the various industries in which different supply management techniques may be applied (Li et al., 2005, p. 618). In their examination, they are able to identify certain factors that may be considered mitigating factors in the application of the certain management techniques to the supply chain. A limitation in their research that is worth noting is that the paper sought response from only one respondent in a firm rather than from multiple respondents—a factor that has many limitations, including the fact that the information may not be collaborated (Bajpai, 2011, p. 31). Qi, Zhao & Sheu (2011) extend this analysis with consideration to the relationship apparent between the strategy a company implements and how this is influenced by the external uncertainty faced by the company. In their examination, they chose to use 604 questionnaires, which were distributed amongst Chinese firms. Their paper classifies strategy into three major categories, namely differentiation, cost leadership and focus. This follows early thoughts in management that identify these as the major strategies companies can adopt in maximising their returns (Schermerhorn, 2010, p. 218). The study by Qi et al. (2011) further indicates and contributes to the growing number of evidence emphasising the company’s need to be competitive as the guiding principle behind the need is to manage supply chains. Qi et al. contributes to the discussions by indicating that supply chain management largely involves two main strategies of being lean and agile. Any company therefore would need to undertake an analysis based on contingency theory, in line with which best suits their needs (Qi et al., 2011, p. 371).
One factor that may be commended in the 2011 work of Qi et al. is the development of a conceptual framework that hypothesises the environmental impacts on the supply chain strategies and competitive strategies adopted by companies to improve their business performance. This conceptual framework then undergoes testing by use of empirical measurements, which creates a connection between theoretical knowledge and the practical knowledge on the subject. This allows for a better argument of the findings in the paper (Zikmund, 2013, p. 47). Another positive is that the analysis is carried out across different regions of the Chinese market, where the large number of respondents lends both external and internal validity to the data collected. This is from the realisation that the year of the research was during a time when China played an important role in world trade, as well as the fact that most of the companies in China directly or indirectly have an influence on the supply chain of most of the world’s companies (Coyle & Coyle, 2009, p. 104).
The paper written by Qi et al. (2011), however, offers only one side of the strategies undertaken in supply chain management by considering that the company will adopt a cost benefit approach. This excludes differentiation and focus as strategies. Differentiation is one of the approaches a company might apply when looking to mass customise. Hence, they conclude by indicating that future research will need to include consideration to this aspect (Qi et al., 2011, p. 371).
Another side of the argument for the need of the management of supply chains is provided by Kocoglu and others in their paper. They contend that the importance of firms to manage their supply chains arises from the integration of such supply chains. They argue that the integration of supply chains leads to a need for better communication, both within and outside the supply chain. In order for this to work best, there is the need to manage the supply chain so as to achieve the best communication structure within it. This is a different argument to that given in most texts that look at integration as a management strategy requiring effective communication to be effective (Dam & Skjøtt-Larsen, 2006, p. 28). This integration involves both internal and external concerns of the supply chain. The study by Qi et al. (2011) came with the realisation that this has been an area in which minimal research has been conducted in the past. As a result, the scholars sought to build a relationship between Supply Chain Integration, information-sharing and the performance of the supply chain. The study is carried out using empirical evidence from 158 manufacturing firms in the Turkish region (Kocoglu et al., 2011, p. 1630).
In their paper, Kocoglu et al. (2011) argue for integration as a means of achieving competitive advantages similar to most other researchers. The researchers’ opinion is that this advantage is provided by cutting down turnaround times on products, with the researchers further contending that the increased integration and sharing of information improves performance by allowing for effective use of resources and capabilities by the companies. The research was able to conclude that information-sharing with suppliers, customers and within the firm was able to improve the relationship of all those involved, which had the effect of improving the company’s performance overall. This is a factor that has been the focus of much discussion in previous texts (Yao & Song, 2001, p. 162). However, the paper has one limitation in that it has considered only specific environments and industries that limit the external validity of the data collected (Kocoglu et al., 2011, p. 1630).
A different paper by Mikkola considers the management of supply chains with the use of three management techniques. These are customisation, postponement and modularisation. The paper introduces the concept of information into the management of the supply chain. This is an aspect that may have been overlooked in most other papers on the subject of managing supply chains. This is a role that is important in the management of supply chains (Mentzer, 2001, p. 292), building on the concepts discussed by other researchers who have been considered in their studies. In the paper, they indicate that the increase in flexible production has improved and been necessitated by the need to customise products. They also indicate that identified by Govindan, whereby increased outsourcing may be credited to the need to build relationships in the supply chain that would lead to better service for the customer (Mikkola & Skjøtt-Larsen, 2004, p. 352).
Importantly, the paper of Mikkola & Skjøtt-Larsen (2004) provides an example of Dell computers where the integration of different supply chain management methods leads to success in different areas of the supply chain. The researchers indicate that the information played a vital role in the efficiency and working on this model, notably between Dell, the company and its suppliers. This exemplifies the application of strategies discussed that look beyond the empirical and theoretical evidence provided in other papers. The paper also has created a theoretical connection between postponement and mass customisation with the use of modularisation. Mikkola & Skjøtt-Larsen also look at how the application of these three strategies affect companies, and further contribute to the discussions on mass customisation by indicating that the creation of modular products improves the overall customisation process. Their paper also indicates that, through this process, there is new information that is gained by the company. The paper presents knowledge on the different kinds of customisation available in different literature and the different strategies utilised in customisation (Mikkola & Skjøtt-Larsen, 2004, p. 352).
Importantly, however, it remains that the paper of Mikkola & Skjøtt-Larsen (2004) lacks in the fact it does not provide empirical evidence to support the models developed in the paper. This development of a model only works to examine the effect of management, but does not put forward the best practices. The other factor is that the simulation only considers the use of modularisation as the link between postponement and mass customisation; various other strategies that could have had an influential role to play were not considered. Nonetheless, the paper has provided more information on the connections that needs to be established in managing supply chains. These are with the improvements in communication that increases and eases the flow of information, both within the organisation as well as with other elements of the supply chain (Mikkola & Skjøtt-Larsen, 2004, p. 352).

2.7 Supply Chain Uncertainty
The term ‘supply chain uncertainty’ is sometimes used interchangeably with ‘supply chain risk’. It is, however, worth noting that the two terms have different meanings. Although some authors argue that the terms ‘risk’ and ‘uncertainty’ have different meanings (Hillson, 2006), others argue that the distinction is so minor that it is not necessary to distinguish between the two (Li & Hong, 2007). In situations where a difference is suggested as apparent, however, the key reason is associated with the type of outcome to be expected. According to Hillson (2006), risk relates to only issues that may result in negative outcomes, whereas uncertainty is associated with both negative and positive outcomes. ‘Supply-chain uncertainty’ therefore is broader than ‘supply-chain risk’, and refers to uncertainties (risks included) that could arise at any point within the supply chain network (Simangunsong et al., 2011, p. 4493). This definition fits with that given by van der Vorst & Beulens, who define it as ‘decision making situations in the supply chain in which those making the decisions do not know what to decide as they are indistinct with regard to objectives, lack information or understanding of the supply chain and its environment, lack information processing capability, are unable to predict with accuracy the impact of possible control measures on supply-chain behaviour or lack effective control measures’ (Simangunsong et al., 2011, p. 4494).
Current published literature on supply chain management tends to be too broad or focused on other areas of supply chain management, such as supply chain flexibility or performance metrics (Simangunsong et al., 2011, p. 4494). Although a number of studies recently have modelled quantitative approaches to managing uncertainty in the supply chain, reviews that adopt a broader look into sets of approaches to management of supply chain uncertainty remain scarce. Moreover, although extensive research exists on the area of supply chain risk, this study does not include important contributions to the uncertainty literature. Moreover, studies that attempt to identify and understand the various sources of uncertainty and how these can be brought into line with management practices so as to enhance supply chain performance are also scarce. Instead, previous research has adopted a broad approach towards the concept of supply chain management, and has focuses on mainly supply chain risks, rather than uncertainty (Li& Hong, 2007; Hult et al., 2010). Others have investigated supply chain uncertainty in general, without delving the specifics of the types of supply chain uncertainty (Lai et al., 2012; Flynn et al., 2010). Demand uncertainty is regarded as the most common and severe type of uncertainty, whereas studies specifically seeking how it can be mitigated are lacking.
Supply chain uncertainty, according to Simangunsong et al. (2011, p. 4493), is a problem with which every manager wrestles. According to these authors, this uncertainty derives from the increasing dynamic and complexity nature of global supply chain networks, and further includes potential delays in delivery, as well as quality issues. They emphasise that such uncertainties are important and should be clearly understood. This article argues that, although extensive research has examined the specific sources of uncertainties that are supply chain-related with regard to internal manufacturing processes, demand-side issues or supplier-side processes, there are other distinct sources remaining that have not received sufficient attention. In this study, Simangunsong et al. (2011) conducted a literature review through which they identified 14 sources of supply chain uncertainty, including the bullwhip effect, which has received much attention in previous research and those lacking in sufficient research, such as parallel interaction, which was only discovered in the recent past. The study also analyses approaches and strategies to managing these sources of uncertainty. It identified 10 approaches that look to reduce uncertainty at the source, and 11 approaches that seek to deal with it, and thereby minimising its effect on performance. The study uses concepts of contingency and alignment to build up a model of supply chain uncertainty, which uses concepts from the literature review to demonstrate alignment between sources of uncertainty and management strategies (p. 1493).
In their discussion of strategies for managing uncertainties, Simangunsong et al. (2011, p. 4495) classify these strategies into two main groups: 1) reducing uncertainty strategies, including any uncertainty management approach that enables firms to reduce uncertainty at its source, such as applying a suitable incentive or pricing strategy to reduce customer demand fluctuation; and 2) strategies for coping with uncertainty, such as those strategies that do not seek to alter or influence the source of uncertainty, but rather aim at identifying ways of adapting and therefore minimising the impact of uncertainty on the firm’s overall performance (p. 4495): for example, firms could develop advanced forecasting methods that enable them to make better predictions of customer demands, and hence reduce forecasting errors. In this case, whilst demand uncertainty is not altered, better forecasting enables firms to anticipate variations in customer demands, and in so doing reduces the impact of the uncertainty (p. 4495).
Simangunsong et al. (2011) identify mitigation as the third concept in uncertainty management. Uncertainty mitigation is defined by Simangunsong et al. (2011, p. 4495) as any action that seeks to lessen the adverse effects of the outcome of activities associated with the supply chain. Mitigation is a common concept in risk management literature, including that of Copacino (1997), Lie et al. (2012) and Luhman (2005). Simangunsong et al. (2011) assume that risk mitigation strategies are similar to coping with uncertainty strategies in their perspective, and hence categorise such approaches under the category of coping with uncertainty. This study lists the following as some of the key uncertainty management strategies identified in existing literature: lean management, supply chain flexibility and agility, supply chain integration, and risk mitigation (p. 4496).

Figure 3: summary of literature on supply chain uncertainty (Adapted from Simangunsong et al., 2011)

With regard to sources of uncertainty, Simangunsong et al. (2011) identify three main sources from the literature, namely supply, manufacturing processes and demand uncertainty (p. 4496). The authors explain that supply and demand uncertainty influence manufacturing process uncertainty, which consequently affects the timely delivery of orders. Of the three types of uncertainty, Simangunsong et al. (2011) suggest that demand uncertainty, which stems from inaccurate forecasts or volatile demands, is regarded as the most severe. This suggestion is also supported by Lai et al. (2012). Simangunsong et al. (2011) split demand uncertainty into demand amplification and end-customer demand, giving rise to four sources of uncertainty (p. 4496). The authors further identify a fifth source from the literature as ‘control uncertainty’, which is seen to relate to the capability of a firm concerning information flows, as well as decisions that transform customer orders into raw material requirements and production plans (p. 4497). They point out that the supply chain uncertainty circle is made up of four quadrants: supply side, manufacturing processes, control systems and demand side. The model also suggests that reducing such uncertainties will lead to a reduction in costs. According to these authors, the reduction of uncertainties is achieved through an integrated supply chain, which is recognised as having minimal uncertainties in each of the four identified areas, and thus is a means of combating supply chain uncertainty. Another source of uncertainty—that of parallel interaction—is derived from the supply chain complexity triangle. This uncertainty is related to complexity, which results from the way in which customers interact with multiple potential suppliers (p. 4498).
In a similar study, Burgess et al. (2006) also outline the supply chain complexity triangle as an effective means of studying, and subsequently controlling, uncertainties. This model introduces a sixth element: uncertainty arising from parallel interaction. This latter element paints a clear picture of the relationship that might exist between customers and a number of potential suppliers. Essentially, this fact increases supply chain uncertainty and can be notably detrimental as it can have an adverse effect on the performance of the supply chain (Burgess et al., 2006, p. 120). On the other hand, the micro/macro model is presumed to be more efficient than other models in handling demand uncertainty by curbing supply chain uncertainty (Burgess et al., 2006). According to these authors, this model identifies more specific sources of uncertainties, and further outlines specific tactics of overcoming the uncertainties. A seventh element, which is decision complexity uncertainty, is incorporated into the model. These elements explain the existence of multiple goals emanating from multiple objectives, all of which need to be carried out in order to reduce demand uncertainty (pp. 120–121). This emulates a complex triangle of events that can be solved by avoiding the indulgence of some of the presumed activities as it will increase the cost of production and subsequent time wastage (Burgess et al., 2006, pp. 120–122).
From the review of various models, Simangunsong et al. (2011) identify 14 sources of uncertainty, which accordingly are grouped into three categories: 1) uncertainties arising from the focal company (internal organisation uncertainty), which include manufacturing process, product characteristics, controls, decision complexity, organisational issues and IT complexity (p. 4498); 2) internal supply chain uncertainty, which stems from within the area of control of the company in focus or its supply chain partners, and includes demand amplification, parallel interaction, end-user demand, order forecast horizon, supplier, and chain configuration, facilities and infrastructure (p. 4498); and 3) external uncertainties, which arise from factors outside of the supply chain and over which the company has no control, comprising competitor behaviour, government regulation, macroeconomic issues and natural disasters, such as earthquakes, floods and hurricanes (p. 4499).

Figure 4: Models of supply-chain uncertainty (Adapted from Simangunsong et al., 2011)

2.8 Demand Uncertainty
Demand uncertainty is defined by various scholars as variations and fluctuations in demand (Chen & Paulraj, 2008; Lai et al., 2012). According to Lai et al. (2012, p. 447), high demand uncertainty implies that manufacturers face an unpredictable market that is characterised by changeable demand and which reflects a lack of knowledge in relation to changing trends, as well as preferences. These authors explain that, under such conditions, managers may have to deal with a lot of ambiguous, contradictory and conflicting market information, making it challenging to decide on how process and product designs can be adjusted so as to meet customers’ requirements. Demand uncertainty results from forecasting errors, variations in volume, and the composition of demand and changes in customer needs, as well as from irregular orders (Amit et al., 2005, p. 4236; Lai et al., 2012, p. 447). Lai et al. (20120) argue that an environment of demand uncertainty, external knowledge and knowledge transfer between specific supply chain partners becomes more valuable and even critical owing to the fact that such vital knowledge can reduce this uncertainty by promoting better preparation.
On the other hand, competitive intensity refers to the extent to which an organisation faces competition in the market in which it operates (Lai et al., 2012, p. 447). These authors point out that the level of intensity is affected by the activities, abilities and resources of the competitors to differentiate, including quality, price competition and product imitation. In an environment less competitive, customers have no options but to stick with the service or product providers available, meaning manufacturers/providers are able to fulfil customer needs for customisation via incremental investment in internal capabilities and resources.
Drawing from Simangunsong et al. (2011), definition of uncertainty mitigation, demand uncertainty mitigation can be defined as those actions that reduce the adverse effects of the outcome of activities associated with the demand side of the supply chain. According to Amit et al. (2005, pp. 4236–4237), the sources of demand uncertainty can be studied from the standpoint of the development of supply chain uncertainty. Various models have been formulated in an effort to enable individuals to gain clear understanding of these uncertainties: for instance, the supply chain uncertainty circle, comprising four quadrants, explains the root causes of uncertainty in the market. The quadrants comprise the control system, manufacturing process, supply process and customer demands (Amit et al., 2005, pp. 4236–4237). This model presumes that the reduction of the four uncertainties will result in a reduction of the overall cost of production, as well as apt performance of the organisation. As a result, ample room for the implementation of managerial strategies in an effort to control the supply chain and supply risk factors will be emanated, thereby significantly reducing demand uncertainty.
Whenever firms face demand uncertainty, they have the option to adopt either a reactive or a proactive approach in order to manage and accordingly mitigate its effects. In the case of proactive approaches, managers can use a postponement of differentiation, modular product design and make-to-order methods. In the use of modular product design, companies use modular product components and processes that are configured in unique ways (Ulrich, 1995). This is important because it helps in producing a high-product variety through a low component variety. This helps in managing demand uncertainty. Several scholars have carried out research, through which the importance of the modular product design has been exhibited through their findings. Companies use other approaches to deal with the effects of demand uncertainty; conversely, it is sensible to adopt the reactive method of dealing with issues of uncertainty, with companies suggested to consider proactive methods of dealing with this issue. Supply chain management practices are recognised as the best ways of dealing with the risks induced as a result of demand uncertainty across all industries; however, the food industry presents a singular set up that should be dealt with through the application of a singular approach.
One part of the issue of uncertainty is that raised by Liu, Shah & Schroeder (2012) in their paper, the study of which sought to examine the effects of demand and supply chain uncertainty in the achievement of mass customisation; this is grounded in the fact that these two factors are interrelated and have an influential role to play in the strategies undertaken in mass customisation (Abdelkafi, 2008, p. 111). The paper by Liu et al. (2012) has expounded on the knowledge in terms of the definition of mass customisation by adding the component of low cost whilst ensuring no reduction in quality or delivery. This develops arguments on the strategic models to be used in the management of mass customisation. The reason for this is that these factors have an influence on the uncertainty of both the supply and demand side (Liu et al. 2012, p. 677).
The examination carried out by Liu et al. (2012) emphasises the mitigating effects of demand side uncertainty on the ability of the company to meet the diversified needs of customers. The paper indicates that, as a result of the correlation between demand and supply side uncertainty, there is need for companies to consider their management strategies in an effort to deal with both in these instances. Liu et al. have also introduced the concept of production process uncertainty. This is mainly influenced by the two other forms of uncertainty considered due to its role as a connecting point (Choi & Cheng, 2011, P. 50). They therefore come to the conclusion that it is necessary for companies to consider these factors if they are to effectively adopt mass customisation. Liu et al. poses the argument that demand uncertainty is a result of the variability concerning the needs of customers (Liu et al., 2012, p. 677).
The paper has worked to provide some direction in terms of various measures companies adopt to improve on their ability to mitigate the uncertainties created through the differences in the clients’ request. This is one of the factors cited as a cause behind demand uncertainty. The paper by Liu et al. (2012) suggests the implementation of what is known as functional integration in the management of demand uncertainty. In this case, integration is aimed at increasing the amount of communication across the various departments. It also considers how this information may be better utilised to achieve the desired results of increased performance. Indicating that, in order for this to happen, there is a need for better integration (Liu et al., 2012, p. 677). This follows previous discussions that have indicated the success of this method in the mitigation of demand uncertainty amongst firms (Copacino, 1997, p. 10).
The study by Liu, Shah & Schroeder (2012) was able to provide the link that has been missing between theoretical knowledge on mass customisation and functional integration, and how these influence demand uncertainty. This contributes to the research of this paper as it proves the theoretical framework that may be used in the development of the strategies involved in the mitigation of the effect experienced as a result of demand uncertainty on the performance of companies. The link was established by using a large sample size data and accordingly utilising organisational information processing theory and the resource-based theory. The sampling was carried out across 266 firms, providing a large enough sample size (Liu et al., 2012, p. 677). A different perspective on the issue is presented in the paper, with the concept of competitive intensity presented, as well as how this is interrelated with demand uncertainty. In their paper, the scholars identified this as the competition that exists between companies with similar products. This indicates that competition affects the level of differentiation in the goods a customer is offered (Lai et al., 2012, p. 446).
According to Boyle, Humphreys & McIvor (2008), demand uncertainty can be handled in three main ways. These uncertainties may be mitigated by implementing managerial strategies, such as the incorporation of appropriate insurance policies within the supply chain, which curb the adverse effects of the detriments arising from the plethora of supply chain activities. This is implemented mainly when the sole objective is to prevent the occurrence of predicted supply chain risks: for example, environmental disruptions (Boyle et al., 2008, pp. 348–349). The SME food industry throughout the Hajj season can greatly benefit from this approach. By applying the necessary insurance policies, the sector can disregard the probability of incurring huge losses due to various environmental risks, such as a lack of adequate rainfall prior to the Hajj season.
Alternatively, effective managerial strategies may be adopted so as to reduce demand uncertainty: for instance, establishing the best tactic to numb the fluctuation of customer demands (Braunscheidel & Suresh, 2009, p. 120). The authors explain that a number of supply chains incorporate suitable pricing strategy in addition to other incentives in an effort to direct the attention of customers towards a specific preference, and thus decrease the margin of demand uncertainty in the market. Moreover, discrepancies caused by human errors in the supply chain are eliminated by incorporation of a bureaucratic decision-making managerial strategy or the use of automated processes to decrease supply chain uncertainties (p. 120). The SMEs food industry can immensely benefit by implementing this strategy throughout the duration of the Hajj season since there is a huge preference of customer demands.
According to Amit et al. (2005) the uncertainty developed during the manufacturing process also can be reduced via implementation of total quality control measures. This ensures the manufacturing process is in accordance with the standards and guidelines stipulated, thus fully meeting the needs of customers. This is especially vital for the SMEs food industry since it is very sensitive owing to its potential to affect the health of consumers; if poor hygienic conditions are apparent during the manufacturing process, a huge number of lives might be lost (Amit et al., 2005, p. 4236). The authors further explain that organisations can also opt to collaborate with customers, which will enable them to reduce decision-making uncertainties and thus decrease demand uncertainty. They point out that this is because a seamless supply chain is created where there is effective communication upstream to suppliers and downstream to customers. Throughout this process, functional and internal integration will be achieved, resulting in the emanation of a systematic supply chain, which will have been managed to reduce control, process and supply, as well as demand uncertainty. (Amit et al., 2005, p. 4237). This would prove very beneficial to the SMEs food industry, especially in the Hajj season, since it will not only enable its ability to curb demand uncertainties but also enable the steady growth of the industry.
Braunscheidel & Suresh (2009) further add that the uncertainties of the manufacturing process can be reduced by introducing a new design for the product being manufactured. This further meets most customer preferences, thereby reducing demand uncertainty. Moreover, demand, in addition to supply-related uncertainties, can be reduced by changing the design of the supply chain: for instance, by redesigning the configuration of the supply chain, including its facilities and structures (Braunscheidel & Suresh, 2009, pp. 120–121). According to these authors, organisations also can redesign the control of the supply chain; that is, changing the strategic objectives, decisions and operation activities. Furthermore, the chain information system, as well as the governance and organisation of the supply chain, should be redesigned (Braunscheidel & Suresh 2009, pp. 120–121). These authors further explain that managerial strategies enabling supply chains to cope with the demand uncertainty are implemented in the manufacturing as well as supply processes. These strategies neither alter nor influence the uncertainty; rather, they find ways of adapting to the situation and thus reduce the uncertainty. This is a very counterproductive managerial strategy that can increase the competitive edge of SMEs in the food industry during the Hajj season (Braunscheidel & Suresh, 2009, pp. 120–121). For instance, those involved with the supply chain invest heavily in reliable forecasting techniques that accurately predict the behavioural pattern of the demand; therefore, organisations are better prepared to handle demand uncertainty. This is owing to their prior anticipation of the demand variation, which enables them to adequately plan their manufacturing and supply processes to counteract the adverse effect of demand uncertainty (Burgess et al., 2006, pp. 705–708). In the case of SMEs in the food industry, a variation in production can be adapted to counteract the demand uncertainty, which greatly occurs during Hajj operation. However, Braunscheidel & Suresh (2009) argue that this is only possible if the supply chain is flexible. This is perfectly elaborated by the transformational system theory, which states that there should be flexibility across the entire supply chain. The input stage should be ready to incorporate any number of suppliers so as to meet all customer preferences. The same case should apply to the process stage, where apt flexibility needs to be incorporated into the labour and machinery sector, readily managing infrastructure, the workforce and equipment as deemed necessary according to customer needs (p. 121). Moreover, the customers should be flexible in the sense of being less sensitive to the delivery dates of products, so long as they fully meet their needs and preferences (Braunscheidel & Suresh, 2009, p. 121).
Simangunsong et al. (2011) identify several strategies of reducing uncertainty; these include new product design, supply chain redesign, and total quality control (TQC) (p. 4510). According to these authors, new product design and total quality control are effective in reducing process uncertainty, whilst supply chain redesign is effective in reducing demand and supply uncertainty. The authors identify elements of supply chain that need to be considered for redesigning as chain control: 1) decision functions that are responsible for managing the execution of strategic objectives and operational activities; 2) chain information systems; 3) chain configuration, such as structures, members involved and facilities; and 4) chain organisation and governance, including authorities and responsibilities (p. 4501). Besides redesigning supply chain infrastructure and configuration, two other strategies are suggested for reducing uncertainty: collaboration with key customers and suppliers, which helps in breaking barriers between the various stages in the supply chain that may reduce the uncertainty associated with decision-making complexity within the firm; and limiting the role of humans in the supply chain process by utilising automated processes or simplifying bureaucratic procedures and decision-making policies (Simangunsong et al. 2011, p. 4501). According to these authors, this could reduce human-related uncertainty.
The concept of collaboration is important in the context of the current study, which seeks to explore Supply Chain Integration as a possible management practice to mitigating demand uncertainty. According to Simangunsong et al. (2011, p. 4501 ), this concept has been further examined by studies that propose that the ‘seamless supply chain’ in which every member and entity in the chain is highly integrated into the system and acts as one will result in reductions in supply, process, demand and control uncertainty. Integration strategy in this context implies extending management systems downstream to customers, and upstream to suppliers, having first attained internal and functional integration (Simangunsong et al., 2011, p. 4501). These authors discuss the ‘well-trodden path’ concept identified in existing literature as a systematic way of achieving a seamless supply chain, where control uncertainty is first reduced, along with process uncertainty, then together in combination with supply uncertainty and, finally, with demand uncertainty. According to the article, this process requires waste elimination through the use of lean strategies, in addition to the synchronisation of material flows across the supply chain. The authors further add that, in addition to lean and agile supply chain, effective information-sharing is a crucial aspect of the collaboration strategy, with organisations commonly relying on the application of Information and Communication Technology (ICT) for this purpose. They note that ICT solutions could provide a suitable decision-making support system; in turn, this may reduce control uncertainty by improving the process and overall quality of decision-making. Conversely, the mismanagement of the information-sharing process, such as through inaccurate or delayed data, could cause difficulties in making the right decisions, hence increasing control uncertainty (Simangunsong et al., 2011, p. 4501). Accordingly, the authors recommend regular employee training, and the regular testing and review of procedures, backup, logging and recovery procedures as ways of reducing ICT complexity related to uncertainty.
Simangunsong et al. (2011) also identify pricing/incentives strategy as another approach to reducing demand uncertainty (p. 4501). Based on the reviewed literature, these authors argue that the revising process or application of controlled marketing promotions is effective in reducing the bullwhip effect in particular. Responsive stock replenishment was also identified as an effective method. This strategy involves ensuring that the period of planning is much shorter than the forecast horizon in an effort to reduce the uncertainty associated with innovative products, which tend to have a short lifecycle and a wider range of products. In this vein, a study conducted in the food industry demonstrated that the application of a stock replenishment cycle shorter than the minimum product lifecycle enabled the case company to satisfy demand and have sufficient time to sell off excess stocks in the event of end-of-product-life products.
Simangunsong et al. (2011) categorise the identified strategies as product design, lean operations, process performance measurement, shorter planning period, collaboration, ICT system, decision support system, redesign of chain infrastructure and configuration, and pricing strategy (pp. 4401–4403).
With regard to strategies for coping with uncertainty, Simangunsong et al. (2011) argue that supply chain inflexibility is one of the most effective approaches for coping with the sources of uncertainty. They argue that a transformation framework can be developed for flexibility by adapting the transformation system theory (processes, outputs and inputs). They further explain that an organisation can create input flexibility on the input stage by, for example, employing multiple suppliers; however, they note that this could increase supply risk, such as delivery reliability and quality issues, particularly when sourcing for critical items. They also add that the cost is higher when multiple suppliers are used; hence, the strategy needs a careful balance. These authors further explain that, at the process stage, machine and labour flexibility could be used to manage people, infrastructure and equipment uncertainty, whereas customer flexibility can be used at the output stage when customers are less sensitive to products or delivery dates (pp. 4503–4504). These authors also note that collaboration is included both as a strategy for reducing certainty, as well as for coping with it, as it involves sharing supply chain information to reduce uncertainty and to address it when it arises unexpectedly.

Figure 5: Uncertainty management (Adopted from: Simangunsong et al., 2011)
2.9 Managerial Practices for Mitigating Demand Uncertainty
2.9.1 Supply Chain Integration (SCI)
In the present intensely competitive business environment, many organisations have come to realise that the provision of the most excellent customer value at the lowest cost is not merely linked to the activities and processes within the organisation itself, but rather to the supply chain as a whole. According to Huo (2012, p. 596), supply chain management has been afforded significant attention in recent years from both researchers and academicians. For this reason, scholars and authors suggest the need for an integrated relationship between manufactures and their supply chain partners. In support of this, Flynn et al. (2009, p. 1633) argue that intense global competition has caused a number of organisations to venture into systematic advance to Supply Chain Integration (SCI). According to the study conducted by Mikkola & Larsen (2004), Supply Chain Integration, which accelerates supply chain management, has received incredible attention in terms of designing and developing relationships between supply chain members.
Flynn et al. (2010, p. 59) further point out that Supply Chain Integration (SCI) continues to be considered a relatively new concept in research that has been defined in different ways by scholars. Although new, SCI has received much attention from researchers investigating supply chain relationships, and collaborative relationships between manufacturers and their suppliers or customers. Whilst some studies have focused on supply chain management as consisting of dynamic relationships between various supply chain partners, others have studied it as a single system rather than attempting to divide it into its different fragmented subsystems. Flynn et al. (2010, p. 59) further point out, that although some SCI definitions stress on flows of materials through the various parts, other definitions focus more information, cash and resource flow. Such descriptions, however, are considered too broad in focus, despite the fact they touch on many of the key elements of SCI. These authors build their definition of SCI from existing literature on the construct, which includes the manufacturer (internal integration), and extending it across both ends of the chain (supplier and customer integration). Furthermore, this is built upon existing gaps in literature to develop what they term as a ‘parsimonious definition’ of the term SCI; therefore, SCI is defined by Flynn et al. (2010, p. 59) as the degree to which a manufacturing firm strategically collaborates with its partners in the supply chain and accordingly collaboratively manages -inter and intra-organisation processes. Lai et al. (2012, p. 444) further define it as the extent to which an organisation strategically works in partnerships with its supply chain partners and manages internal and external organisational processes with the aim of achieving effective and efficient flows of products, services, information, money and decisions, with the intent of delivering maximum value to its customers. According to Flynn et al. (2010), the goal of SCI, in this domain, is to realise effective and efficient flows of information, products and services, decisions and money in an effort to provide the customer with maximum value at low costs and without delays (p. 59). Lai et al. (2012) also define integration as the extent to which the distinct internal functions of an organisation have the ability to be in partnership with one another, synchronise intra-organisational activities, make strategic decisions and accordingly devise cross-functional integral relationships. According to the authors, the integration mainly involves cross functional harmonisation, joint decision-making and internal relationship management. Throughout the course of composing organisational activities into joint processes, the authors point out that it breaks down the traditional functional way of assisting the attainment and relocation of organisational knowledge into particular designs, processes and end products, which eventually pave the way to a more connected and coordinated internal reaction to marketplace changes and distraction (p. 444).
Organisations put in place processes intended to integrate suppliers, internal functional units and suppliers with the overriding aim of optimising the total performance of all those partners involved in the supply chain system. Rungtusanatham et al. (2003) state that customer and supplier integration is a valuable approach to getting external resources from suppliers and customers. When integration has been adequately completed, improvements in process, control, supply and demand, in the operations of the business, are witnessed (Towill & Christopher, 2002). This promotes information-sharing across the supply chain, as well as the promotion of integrative inventory systems. It goes a long way in terms of improving quality service delivery and responsiveness to the dynamic market (Lambert & Cooper, 2000; Lee & Whang, 2005). In addition to customer and supplier integration, internal Supply Chain Integration focuses on the development of products. This uses different terms, such as cross-functional teams and functional coordination (Vickery et al., 2003; Min & Mentzer, 2004). In the case of functional coordination, there is the measurement of collaboration and interaction within a given company (Kahn & Mentzer, 1996). Cross-functional teams, on the other hand, bring together issues of marketing, research and development, manufacturing and purchasing personnel together. This is important in reducing the costly design of products and maintenance, and duplication. It improves product reliability and enhances customer satisfaction.

Figure 6: A supply Chain network (Chandra & Kumar, 2001, p. 290)

Information-sharing, organisational coordination and product co-development are the main areas of interest in the mitigation of demand uncertainty effects. These three areas are important for empirical investigation owing to the fact they are vital in the design and development of products. Information-sharing involves the sharing of production, inventory, marketing and technological information across all areas of the customer and supplier (Fisher, 1997; Ayers, 2001; Stock & Lambert, 2001). Supply Chain Integration also looks at product co-development, where customers, suppliers and internal functional units make joint efforts to develop products. This refers to the joint design of products, production operations and process engineering, with the collaboration of key customers and suppliers. In integrated product development, there is close internal coordination from the stage of product design and process development, through to production and product launch. Different authors have stressed the co-development of products with various stakeholders. Song et al. (2009), for example, stress the need to include customers in the process, whereas Min & Mentzer (2004) emphasise the need to involve all internal functions whilst completing product development.
Some of the aspects of Supply Chain Integration were studied and presented by Flynn, Huo & Zhao in their paper (2010), during which they considered the fact that previous studies on the subject were inconsistent when examining their inability to accurately define Supply Chain Integration. The paper sought to redefine this term, and thereby provide further insight on the subject. The writers contend that previous papers have only sought to establish the relationship between the supplier, end user and firm; they have done this to the detriment of the other internal factors that play a role in influencing the supply chain and, by extension, the integration of the supply chain (Flynn et al., 2010, p. 58).
The paper by Flynn et al. (2010) examined three individual aspects of the firm, and how these three aspects affect performance; these are supply chain, business performance and operational performance. The paper applies a contingency approach, which uses hierarchical regression in an effort to determine the effect of certain given aspects of Supply Chain Integration on company performance. This statistical measure allows the study to explore the factors on the basis of importance (Gliner & Morgan, 2000, p. 219). The scholars also use cluster analysis under the configuration approach in examination of the patterns in Supply Chain Integration. The use of cluster analysis enables the researchers to analyse the structures within the company without the need to define a priori. In the case of the study by Flynn and his colleagues, a more informed analysis of the integration existent in the supply chain can be completed. This is owing to the fact it allows for the study of the relationship apparent between these different factors (Neider & Schriesheim, 2007, p. 130). These were done to try to develop a relationship between the supply chain and the performance of firms. Moreover, it also aimed at identifying the most effective strategy that could be used in improving firm performance (Flynn et al., 2010, p. 58).
Despite the comprehensive nature of the research, the writers identify certain items that would require further analysis. One of these areas is centred on considering customer and firm integration from a longitudinal point of view; this would require creating a study to cover a greater period of time. There is also the consideration that the firms under examination were all Chinese, meaning there is a need to complete a study that takes into account businesses from other geographical locations; this is from the consideration of the fact that different factors may come into play in different parts of the world. It should be noted, however, that the ideas presented in the paper offer knowledge on various strategies that may adopted in firms with different characteristics from those discussed. On the other hand, this provides a foundation upon which future studies of the issue of Supply Chain Integration may be examined (Flynn et al., 2010, p. 58).
Lai et al. (2012) identify the three basic forms of SCI as internal integration, supplier integration and customer integration: internal integration stresses the integration of internal functions and processes, whilst supplier/customer integration emphasises the importance of building long-term close and collaborative relationships with external partners (suppliers and customers) (Lai et al., 2012, p. 444).

2.9.1.1 Internal Integration:
In their expanded discussion of the types of SCI, Lai et al. (2012, p. 444) describe internal integration as the degree to which the various internal functions and processes of an organisation collaborate with one another, strategically coordinate internal organisational activities and decisions, and accordingly form integral relationships across the different functions. Such integration mainly involves joint decision-making, internal relationship management and cross-functional coordination. Kotcharin et al. (n.d., p. 1631) define internal integration as the extent to which an organisation can plan its organisational practices, procedures and behaviours into joint, synchronised and manageable processes, with the general aim of achieving customer desires and needs.
Lai et al. (2012, p. 444), observe that such integration majorly comprises cross-functional coordination, collaborative decision-making and internal relationship management. The authors go on to argue that, by structuring organisational processes and activities into cooperative processes, internal integration disintegrates the normal functional silos to aid the acquisition, as well as the transfer of organisational knowledge into explicit final products, designs and processes; consequently, this leads to a more coordinated and connected internal response to customer and market place changes as well as disruptions.
Flynn et al. (2009, p. 60) observe that, since internal integration breaks down functional barriers and brings about cooperation with the aim of meeting the requirements of customers, as opposed to operating within the functional silos related to traditional departmentalisation and specialisation, it is believed to be related to performance. Such writers also point out that, despite the fact manufacturers might sustain a functional organisation structure, orders from customers flow across functions and activities. They further argue that, when an order is delayed, customers are not bothered which of the functions led to the delay, but simply that the order will be realised. Flynn and his colleagues go on to explain that such a scenario calls for an integrated customer order fulfilment process, through which all activities and functions activities are harmonised. Thus, it is concluded that information-sharing, joint planning, cross-functional teams and collaboration are key elements of the internal integration process.

2.9.1.2 Customer Integration:
Supplier/customer integration refers to the extent to which an organisation can partner with suppliers and customers to structure its interorganisational practices, behaviours, processes and strategies into collaborative, manageable and synchronised processes in order to meet customer requirements (Lai et al., 2012, 444, p. 444).
Customer integration mainly involves customer partnership and the sharing of customer information, and the involvement of customers in product development and delivery (Flynn et al., 2010, p. 59). A different paper by Kotcharin et al. (n.d., pp. 1–2), defines customer integration in the same way. Chavez et al. (n.d., p. 1) describe customer integration as the combined planning, partnership and synchronisation of processes with the presence of major customers to accomplish mutually beneficial goals. This paper has also worked to provide some direction in customer integration in that information-sharing similarly has been looked upon as one more significant aspect of Supply Chain Integration. According to the authors, information-sharing refers to the determination to come up with tactical, strategic and operational information accessible to supply chain members. The scholars go on to provide examples: for instance, they point out that information-sharing can comprise tactical information, including production tactics, performance metrics of buyers and strategic information, such as sales forecasts from buyers. Consequently, information-sharing can comprise operational information: for instance, inventory holding information, which, when shared, can be of assistance in an effort to tone-down information distortion from all members of the supply chain.
Lai et al. (2012, p. 444) note that customer integration is important for manufacturers as it allows access to customer information, knowledge-sharing, the pursuit of joint development activities, speed-up of decision-making processes, reduction of lead times, and improved process flexibility. According to this article, such integration is crucial to manufacturers; besides helping them to obtain information about customer needs, it also enables them to gain a better understanding of their preferences and requirements in terms of what they prefer and why.
Flynn et al. (2009, p. 60) point out that customer integration is related to customer satisfaction, both directly and indirectly, based on its relationship of product development and innovation. The authors state that, in an integrated supply chain, growth of a strong strategic joint venture with suppliers will make possible their understanding and expectation of manufacturers’ needs, with the aim of better meeting its dynamic requirements (p. 60). Hence, this mutual exchange of information in regard to products, processes and schedules assists manufacturers in developing their production plans and ensuring the timely production of good, thus improving performance on the delivery of goods (p. 62). Flynn et al. (2009, p. 60) point out that, by developing a good understanding of the manufacturer’s operations, suppliers realise a high level of customer service, which eventually similarly assists manufacturers in improving their customer service.

2.9.1.3 Supplier Integration:
Supplier integration entails mainly supplier partnerships, supplier information-sharing and the involvement of suppliers in product development (Lau et al., 2012, p. 444). According to Flynn et al. (2010, pp. 59–60), developing close ties with suppliers enables service providers and manufacturers to gain better inputs from suppliers and accordingly to include their suggestions and recommendations into business operations. Lai et al. (2012, p. 44) argue that a successful ongoing partnership between suppliers and manufacturers is mutually beneficial to both parties, as it helps them in achieving their strategic goals. Zhao et al. (2008, p. 368) also emphasise that a closer business process alignment with a firm’s suppliers also facilitates the smooth delivery of various raw materials and components on a timely basis, thus enabling the manufacturer to reduce total lead time for delivery of customised goods.
Lau et al. (2012, p. 445) contend that closer binding with suppliers facilitates manufacturers in attaining better inputs from suppliers and integrating their recommendations and suggestions into business operations. The authors also point out that a closer business process association with suppliers encourages the smooth provision of numerous components of raw materials, ensuring timeliness in the process. Flynn et al. (2009, p. 60) further suggest that supplier integration has been presumed as related to product development performance and supplier communications performance. Conversely, however, some scholars seem not to have found a significant relationship between supplier integration and operational performance.
Lai et al. (2010, p. 771) suggest that suppliers also may grant familiar suggestions to manufacturers for product development in an effort to protect the value of their existing resources: for instance, knowledge concerning capacity and engineering. These authors argue that, by limiting themselves to information obtained from current customers and suppliers, manufacturers might trim down their potentiality of coming up with extremely innovative products in a competitive environment (p. 771).
Natour et al. (2011) also examined Supply Chain Integration and collaboration. The authors observe that supply chain collaboration over the years has been studied from four distinct perspectives. They identify these perspectives as uncertainty reduction, transaction cost economics, resource-based view, learning and knowledge. In their study, these authors reviewed past studies that have applied stakeholder theory and the theory of constraints to advance joint relationships in the supply chain. Their analysis indicates that previous studies provide conflicting insights as there are no extensively agreed upon conceptualisations of Supply Chain Integration and collaboration; this gap is holding back the progress of supply chain management research and practice. Despite this, positivist schools of thought, to a wider extent, have enlightened numerous conceptual frameworks proposed in the literature and the research efforts largely centred on structural integration. According to these scholars, whilst ‘joint’ in the supply chain is at times referred to as a means of accomplishing this structural integration, the terms ‘integration’ and ‘collaboration’ are not only used interchangeably, but also are interpreted with terms similar to information-oriented lining with supply chain processes. For this reason, it may be concluded that there seems to be no generally accepted theoretical or conceptual foundation for informing Supply Chain Integration and collaboration contributions. However, the authors seemingly argue that a few recent contributions have highlighted a more encompassing outlook of collaboration working on the soft aspects, such as goal congruence, decision synchronisation and incentive alignment, and the traditional hard perspectives of resources and information-sharing. Natour et al. (2011, p. 512), however, emphasise a main concern: that the majority of theoretical aspects proposed in preceding studies are not able to effectively tackle the underlying behavioural aspects, namely relationships, trust, politics and power, all of which, at present, characterise the supply chain environment.
Koçoglu et al. (2011, pp. 1633–1634) extend this analysis by stating that SCI has gained substantial attention with the transformation of manufacturing and supply strategies and intense globalisation. The scholars further argue that the theoretical groundwork of SCI may be traced back to Porter’s Value Chain model, emphasising the value-creating connections between chain members. They point out that, in the present day, the growing popularity of SCI over the last decade has demonstrated the linking of all supply chain members, where the corresponding alignment of partners’ objectives to move towards a shared structure of values is important for organisations to provide customers with excellent value (p. 1634). Another positive fact is that the effective connection of various supply chain activities, without forgetting the internal functions of an organisation with the external operations of suppliers, customers and other supply chain members, is pivotal in enhancing accurate supply chain relationships and accordingly assisting in the coordination of information flows from supplier to manufacturer and customer, as well as in the backwards flow from customer to manufacturer and supplier (pp. 1634–2635). In addition, these authors suggest that appropriate supply chain relationships, with a view on strategic partnership with supply chain partners, influence the flow of timely, accurate and quality information. On the other hand, they note that, although definitions regarding SCI make up the correspondence between integration and information-sharing, implying that SCI has an impact on the effective and efficient flow of information, very few empirical studies thus far have focused on the influence of the power of SCI on information-sharing and decision-making, with most having focused on supply chain performance.
In the research by Huo (2012), it is argued that SCI is vital to the success of organisations and supply chains (p. 597). The author points out that, regardless of its significant influence, SCI has not received the attention it deserves until recently. This paper further states that there are no commonly accepted sub-elements of SCI, and the relationships between distinct SCI dimensions are not consistently demonstrated in previous studies (p. 597). Furthermore, Huo further recognises that there is very little empirical evidence demonstrating how distinct SCI dimensions concurrently affect different types of organisational performance (pp. 598–599). This article also emphasises that several other studies bring out the sole roles of supplier integration or customer integration in improving performance when supplier integration or customer integration is considered separate, whereas the rest has centred on only the impact of internal integration on performance (p. 598). On the contrary, Huo argues that various recent studies take into account internal integration and external integration, which are recognised by such scholars as connected to performance. However, Huo states that these studies have limitations in the sense that they only consider one or two performance dimensions, with findings not seen to be in agreement. Moreover, they have afforded very little attention to the relationships between distinct types of performance and the mediating outcome between different types of SCI and performance varieties.
According to Koçoglu et al. (2011, pp. 1633–1634), Supply Chain Integration improves the extent of partnerships with external supply chain members, and in so doing, structures the organisation’s strategies, processes and practices into collaborative, synchronised, aligned activities in an effort to achieve inter-organisational information-sharing. This paper further explains that the ever-changing environment structure, as a result of the collaborative relationships between suppliers and buyers, improves the necessary technological and managerial resources to be implemented and utilised by multiple supply chain partners as competitive capabilities, as opposed to putting up with the cost of internalising these resources. Thus, SCI leads all participating parties towards an expanded resource base in order to link core aspects from heterogeneous sources of information into a common platform and accordingly attain information-sharing (p. 1634). According to Koçoglu et al. (2011), there seems to be some consensus in the literature that the eminent level of close relationships with supply chain partners brings about increased visibility of suppliers’ operational activities, which eventually allows transparency and a platform through which the information can be communicated between participating parties (p. 1634). The argument above concludes that SCI may be included in a firm’s infrastructure for the strengthening of information-sharing between supply chain members. Furthermore, the authors point out that SCI improves information-sharing through creating trust-based relationships. They further state that the deepening trust-based relationships between parties enhances the contract period between supply chain partners, and thereby attracts efficient conflict resolution, and encourages responsiveness of customers, flexibility and the flow of information through arousing the sense of belonging and determination to share (p. 1635). This study defines trust as the degree to which an organisation believes that its partner, with whom an exchange occurs, is honest and generous, and is recognised as an outstanding safeguard of long-term stability and the success of inter-organisational relationships (p. 1635). The reason behind this is that the growth of long-term secure relationships with important value-network members, which are significant to the functioning of the supply chain when considering the use of their power to affirm decisions, solutions and direct policies, is based on the level of confidence in relationships. In most cases, customers have an impact on the decisions of a manufacturer; similarly, the manufacturer looks for trust-based relationships with a customer owing to the fact that, with the increase in the degree of trust, the determination of the parties to share physical, financial and information-based resources also improves (Koçoglu et al., 2011, p. 1635). These authors further recognise that SCI enhances the involvement of customers in supply chain activities and similarly enhances the effort of supply chain members when it comes to information-sharing.
In addition, Koçoglu et al. (2011, p. 1634) further assert that SCI provides organisations with the opportunity to focus on their core competencies and particular vicinity of expertise by aligning itself with other supply chain members with different resources, technological knowledge and expertise. According to these authors, it is presumed that SCI refers to the implementation and exhaustion of collaborative and coordinating structures, processes, technologies and practices between supply chain members with the aim of building and accordingly maintaining a faultless channel for the accurate and timely flow of information, materials and, ultimately, finished goods (Koçoglu et al., 2011 p. 1634). Accordingly, this kind of structuring provides an option in cases where there is a lack of or restricted resources; this limits the cost of transaction and the ability to negotiate, hence paving the way for organisations to reap the benefits of utilising common resources and capabilities (Koçoglu et al., 2011, p. 1634). Furthermore, the parties involved have a better understanding of each other’s business in an improved way, and can help each other through flows of correct information that is well-timed in the realisation of higher supply chain performance. Koçoglu et al. (2011) conclude that SCI enhances increased specialisation, thereby allowing the flow of correct information in cases of need.
Drawing from the extended RBV of the firm, Lai et al. (2012, p. 444) argue that all three types of Supply Chain Integration (internal, supplier and customer integration) influence the development of MCC within a firm, with both internal and external integration promoting the strategic resources that are crucial for MCC development. According to these authors, internal integration in particular has established a platform for creating, assimilating and applying knowledge to product design. The article further explains that the synergistic effects of the joint efforts of various firm functions and departments provide strategic resources in an effort to address the complexity associated with customisation. The authors further note that organisations can attain and deploy supply chain resources in addition to knowledge through integration with suppliers and customers. They further explain that firms are positioned to acquire the strategic resources they can use to improve the essential elements of mass customisation, such as cost, flexibility, efficiency, product quality, delivery, flexibility and agility, by leveraging their capabilities and resources, and accordingly collaborating with external partners and agility. Lai et al. (2012) further add that those forms that are endowed with better internal socialisation and coordination also have more capability to acquire resources from external partners. Internal integration therefore is regarded as a core strategic resource for enabling external integration in the development of MCC.
Lai et al. (2012, p. 448) argue that demand uncertainty, together with competitive intensity, have contingent effects, and are the most important environmental conditions for the development of mass customisation. According to these authors, empirical evidence exists to indicate that the effects of Supply Chain Integration on a firm’s operating capabilities are moderated by environmental context. The authors further explain that rapid changes in demand require manufacturers to attain new knowledge that will be pivotal in guiding customisation as existing experience and knowledge becomes invalid. On the other hand, however, when the extent of demand uncertainty is low, manufacturers then can develop their mass customisation capabilities through relying on the existing resources and knowledge to design, produce and provide customised products. However, this task becomes a challenge when demand uncertainty is so high that the changes required cannot be addressed in this manner. In such a case, the firm needs to create an exchange and accordingly acquire new knowledge, as well as resources, by collaborating with external partners. Lai et al. (2012) further explain that combined effort by supply chain partners is necessary in order to deal with unpredictable demands and to develop customised products. This study provides an example by noting that a firm is able to attain accurate demand information on time by collaborating with customers; this leads to better decisions in making customised products to meet customer requirements (p. 448). Through improving collaboration with its supplies, the organisation can identify ways of exploring and increasing the range of possible solutions for meeting customer needs and reducing costs and lead times by improving joint processes (p. 448). Lai et al. (2012) emphasise that since challenges are more severe in an uncertain environment, it is important for manufacturers operating in such an environment to have closer collaboration with their external partners (p. 448). They argue that since internal integration drives and enables external integration, manufacturers operating in an uncertain demand environment can increase the impact of internal integration on external integration, and consequently develop greater mass customisation capabilities (p. 448). This implies that Supply Chain Integration and its dimensions (customer, internal and supplier integration) are related to mass customisation capability and may be effectively used in the mitigation of demand uncertainty.

2.9.2 Mass Customisation Capability (MCC)
Davis (1987) defines mass customisation as a process where manufacturers tailor-make products in order to satisfy individual customer needs at the same prices as those of mass-produced items. This can be approached from other perspectives. Mass customisation capability is considered as the ability of a company to come up with customised products on a larger scale but at a cost similar to those of non-customised goods. This method is very useful in dealing with demand uncertainty, which can plague a company. Responsiveness, volume effectiveness and cost effectiveness are some of the measures used to determine mass customisation capability (Dyer et al., 1998). Different sources point out that any firm can perfect its mass customisation capabilities by examining various elements; they include coordinating suppliers, postponing key steps in production and the implementation of modularity-based manufacturing (Salvador et al., 2004).
Lai et al. (2012, p. 443) define MCC as the ability of a firm to offer a comparatively high volume of product alternatives for a comparatively large market that demands customisation without significant trade-offs in quality, cost or delivery. According to these authors, MCC can be discussed under four main aspects: 1) customisation cost efficiency, which implies the ability to offer customised products at a price comparable to mass production; 2) high volume customisation, which is the ability of a firm to aggregate the individual demands of customers into large-batch common production and deliver customised products at volumes similar to mass production; 3) customisation responsiveness, which is the ability of firms to implement measures that reduce total lead time for the delivery of customised products and to deal with customisation demands quickly; and 4) customisation quality, which is the ability of firms to guarantee and manage a consistent quality of all customised products.
According to Can (2012, p. 16), one of the success factors for mass customisation is that the products should be customisable. Modularity, continuous renovations and multi-purposefulness are some of the methods identified as able to be used to increase customisability. The author, however, notes that some scholars argue that modularity is not necessary for mass customisation, but rather contributes to decreases in complexity and costs (p. 16).
In his discussion of the factors enabling mass customisation, Can (2012) explains that very short reaction times that are order-based are essential for mass customisation; hence, logical flexibility is a key enabler of mass customisation (p. 18). Physical flexibility is also viewed as essential in achieving a more agile production system. the author identifies the concepts of modularity, expandability, reconfigurability, reutilisation and scalability as higher level enablers of agility and flexibility. In his discussion, Can categorises flexibility as strategic flexibility that responds to changes in the firm’s external environment and operational flexibility, which in turn reacts to changes within the internal environment (p. 18). Based on a review of previous studies, Can argues that strategic flexibility includes product, mix, volume, production and expansion flexibility, all of which together enable the firm to respond in an agile way. Conversely, operational flexibility, such as process, delivery, routing, programming, labour and machine flexibility enable mass customisation (p. 18). According to this author, the relationship between agility, flexibility and mass customisation is strong.
Efficiency is also another identified enabler of mass customisation. According to Can (2012, p. 18), efficiency represents the mass side of mass customisation. The different definitions of mass customisation state that it should be cost-efficient or as efficient as mass production. A further review by Can emphasises that previous studies have established agile manufacturing, lean manufacturing, customer-driven design and manufacturing, and supply chain management as the processes and methodologies enabling mass customisation (p. 19).

2.9.2.1 Modularisation:
Modularity is defined as the level of module application by minimum interaction between various modules. It is the utilisation of the portions of modules with well-defined few interactions amongst them whilst including one or few functional elements in each of the modules (Can, 2012, p. 23).

Figure 7: Customer involvement and the modularity in the production-cycle (adopted from Can, 2012, p. 23).

Gershenson et al. (2003) identify various benefits associated with modularisation:
1. Products are easily updated owing to functional modules.
2. A smaller set of components increases product variety.
3. The use of components across product families increases component economies of scale.
4. Decreased lead time owing to the fewer components.
5. The decoupling of product functions facilitates simple design and testing.
6. The decoupling of product functions facilitates easy design and testing.
7. Differential consumption increases ease of service.
Moreover, modularisation also is characterised by some costs that need to be well managed, including:
1. Lack of performance optimisation as a result of the lack of smaller sizes and function-sharing.
2. Reverse engineering is easy, hence increased competition.
3. The re-use of components can result in static product architecture.
4. A lack of component optimisation can result in increased unit variable costs.
A paper written by Brun & Zorzini (2009) sheds more light on the issue of mass customisation, in which the authors present some of the history by indicating that the onset of mass customisation may be traced back to Davis in the 1980s through his book titled Future Perfect. Davis defines mass customisation as the ability of a company to supply products to a customer that are tailored to the tastes of the individual, thus enabling each and every customer to get what they want. Notably, the process also would be carried out at a cost comparable to that of standard goods and services (Boër, 2013, p. 7). Brun & Zorzin (2013) contend that, in so doing, agility, integration and flexibility are a component. They also contend that an element brought out in mass customisation is that of increases in cost; hence, companies will need to take this into consideration despite the need to ensure that there is no considerable increase in the price of final goods and services. Another point to be raised centres on that of optimum integration within the company, and how this influences the mass customisation capabilities of a company (Brun & Zorzini, 2009, p. 205).
The paper by Brun & Zorzini (2009) examined previous research papers, and from these has postulated two managerial methods that may be adopted in dealing with the issues arising from the adoption of mass customisation. These were modularisation and postponement techniques; as evidenced in earlier discussions, these were the subject of previous papers, one of which was that written by Mikkola & Skjøtt-Larsen (2004). Modularisation is a technique mostly applicable in the automotive industry, with great success a representation of the diagrammatic flow shown after this paragraph (Doran, 2004). The paper by Brun & Zorzini outlines that modularisation is product-focused, whilst postponement is a process-focused approach to managing the supply chain. The analysis of these factors was based on the literature review writers presented on the subject; this provides some of the background knowledge needed to examine the issue of mass customisation (Brun & Zorzini, 2009, p. 205).

Figure 8: Example of a modularization flow chart where value is added with each tier

However, the paper considered these factors by examining 20 companies, all of which were based in Italy. Two basic limitations that are to be argued for the application of this method are as follows: firstly, a sample of 20 companies is too small to make any valid argument as regards the entire industry; and secondly, the sample size was selected from one geographical region. Both of these factors together limit the external validity of the data collected. As stated previously, there may be many differing elements potentially influencing the way in which companies perform prior to and following the application of the suggested strategies—a factor identified in many previous texts (Sanchez, 2008, p. 161). Two areas that may determine the strategies to be applied by companies are complexity and customisation levels. Four main strategies were developed that were dependent on these two factors; these were identified as rigid, flexible, postponed and modularised structure. Despite the limitations of the study, however, the development of the theories in the paper will be useful in the consideration of mass customisation (Brun & Zorzini, 2009, p. 205).
A different paper went into consideration of the history and various issues associated with mass customisation as an integral part of the company. The paper by Fogliatto, Silveira & Borenstein (2012) examined some of the definitions identified in previous papers, with the introduction of the paper indicating some of the areas in which mass customisation as a strategy could be applied successfully so as to give companies a greater competitive edge. One such area was that of the food business. The paper is an expansion of another study (Da Silveira, Borenstein & Fogliatto, 2001, p. 1), which presents the developments that have gone into the development of strategies in mass customisation in the period spanning 2001–2010. The identification of these factors is indicative of how external environments have an effect on the internal strategies developed within the company (Fogliatto, Silveira & Borenstein, 2012, p. 14).
One aspect that was raised for consideration in the paper was pertaining to the economics of mass customisation; categorising the economics of mass customisation into economies of integration and the difference between the price it takes to customise and the premium charged for the product. The paper was able to build on previous papers that had identified six areas from where the success of mass customisation could be achieved. These were in the areas of increasing customer demands, allowing for increased competitiveness in the market place, creating value in the supply chain, improvements in the technologies applied in the management of mass customisation, improvements in the tools used in mass customisation, and improvements in knowledge-sharing between customers and the company, which have created an increased awareness of customisation (Fogliatto et al., 2012, p. 14).This is a factor that has been mentioned in other texts (Blecker, 2005, p. 26).
The paper contends that mass customisation as a strategy has become an integral part of the daily operations of most companies; nonetheless, there still remain certain issues that need to be addressed. The paper emphasises increased technological capabilities for these changes in order for the utilisation of mass customisation to be effective. The issues raised lie in areas such as the web-based tools and models available, which move away from the theory to the practical application of mass customisation. As with many other papers already considered, the paper provides worthy areas where future research can be undertaken in (Fogliatto et al., 2012, p. 14).
The study carried out by Lai et al. (2012), which sought to examine the way in which supply chain management integration is linked to mass customisation capability based on the resource-based view, significantly contributes to the mitigation of demand uncertainty through mass customisation and Supply Chain Integration. This study examined two issues: the joint influence of internal integration, supplier integration, customer integration and the interaction amongst them on Mass Customisation Capability development; and the moderating effect of environmental conditions (demand uncertainty and competitive intensity) on the impacts of Supply Chain Integration on Mass Customisation Capability (MCC) development. The authors used contingency and the extended resource-based view (ERBV) of the firm theories to develop a conditional indirect model, which they test using a dataset of 289 manufacturing firms from nine countries. The extended resource-based view (ERBV) is also referred to as the relational view or concept of dynamic capabilities, and makes the presumption that the resources required for internal usage also can be obtained from the external environment through collaboration and cooperation (Bohnenkamp, 2013, p. 6). RBV therefore asserts that, although firms should focus on in-house activities, they should also consider environmental changes that require access to external resources (Barney, 2013, p. 4). The results obtained by Lai et al. (2012) were found to be consistent with ERBV, as they demonstrated that internal integration has a significant direct impact on MCC and also plays a key and strategic role in central in developing customer and supplier integration. The study findings, however, indicate that, whilst customer integration was found to improve Mass Customisation Capability directly, the impact of supplier integration was found to be insignificant. Internal integration also was found to have a positive effect on MCC indirectly through customer integration. It was observed that the indirect effect amplified with intense competition and demand uncertainty.
According to Lai et al. (2012, p. 443), today’s modern-day business environment, which is characterised by fierce competition and rapidly changing customer needs has forced manufacturers to offer customised products in addition to services at prices that are reasonable and that do not vary significantly from those produced for mass production. These authors point out that Mass Customisation Capability has become an essential competitive factor in meeting and satisfying customer requirements in a cost-effective manner. They note that finding ways of enhancing MCC therefore is of valuable interest to both practitioners and researchers. This study also did not look at the mitigation of demand uncertainty, nor identify it as an area into which future research should research. The present study therefore is a response to this call.
Lai et al. (2012, p. 444) explain that manufacturing firms can improve their mass customisation capabilities by improving practices whilst also integrating resources within the organisation’s boundaries. The authors further emphasise that external interactions with suppliers and customers also can contribute significantly to MCC. They point out that, since MCC has been found to enable firms in attaining new and innovative forms and sources of competitive advantage, it has been regarded as an essential organisational capability. Lai et al. further suggest that, in order for firms to boost these capabilities, they need to find ways of building, integrating and reconfiguring internal as well as external resources so as to meet the rapidly changing requirements of the business environment. Therefore, the study examined how firms enhance MCC through resource integration and collaboration beyond their internal boundaries. The study draws from the knowledge-based view of the firm in terms of examining how collaboration and knowledge-transfer between buyers and suppliers play a role in the development of organisational capabilities.
Lai et al. (2012, p. 444) point out that a number previous empirical studies have investigated methods of developing Mass Customisation Capability by improving management practices and strategies such as customer involvement, postponement, modularity-based manufacturing processes, time-based manufacturing routines, organisational learning, sociotechnical work-design strategies, quality management and organisational structure design. These studies demonstrate that, in order for a firm to attain success in mass customisation availability, there must be complete understanding of the needs of customers and product availability, as well as the capability to manufacture products in an efficient and effective manner (Lai et al., 2012). This article further explains that there needs to be high levels of agility and flexibility in processes within a firm in order for MCC to be successfully developed. This implies that MCC development requires organisation to have effective internal integration across all functions and accordingly to fortify its degree of integration with suppliers and customers in order to effectively respond to changing customer needs and market requirements.

2.9.2.2 Customer Order Decoupling Point:
Olhager (2003, pp. 319–320) defines Customer Order Decoupling Point (CODP), which is also known as order penetration point, as the point at which products are linked to a specific customer order within the manufacturing value chain. Different positions of CODP specify different manufacturing situations, including Make to Order (MTO), Engineer to Order (ETO), Make to Stock (MTS) and Assemble to Order (ATO) (Can, 2012, p. 29).
According to Rudberg & Wikner (2004, p. 445), CODP is that point separating the decisions made under certainty from those made under uncertainty regarding customer demand.
Figure 9: The typical sequential approach to the CODP concept (Adopted from Rudberg & Wikner, 2004)

Rudberg & Wikner (2004) explain that, in the figure above, the speculation parts indicate the forecast-driven activities concerning customer demand that are carried out under uncertainty. The commitment part shows the customer-order-driven activities. Hence, the triangles between commitment and speculation point out the OCDP position in the value added chain. According to Yank & Burns (2003, p. 2078), the decoupling point is a critical element in the supply chain when considering positioning, which is an important decision in designing the supply chain. They describe OCDP as the point where the customer order and supply chain penetrates, which differentiates forecast from order-driven activities (p. 2078).

2.9.2.3 Mass Customisation and Customer Order Decoupling Point
In order to analyse customer participation in mass customisation, CODP is studied in two dimensions; that is, engineering and production dimension (Can, 2008). The paper goes on to state that the positioning of the CODP in mass customisation entails recognising the optimal balance, linking the forces of productivity and flexibility. Moreover, the shifting of CODP upstream in the flow of material makes the flexibility competitive and accordingly increases the customisation capability of the manufacturing system. Alternatively, when CODP is shifted downstream, there is much emphasis placed on overall productivity, and organisations advance in price competition. The author argues that, when it comes to positioning CODP, it ought to be taken into account that the marginal benefit from flexibility lessens as CODP is shifted more upstream, and the marginal benefit from productivity lessens as CODP is shifted more downstream. For this reason, the author suggests that balance across these forces is vital in terms of attaining mass customisation. It is further recognised that the degree of customisation ought to be appropriate to customer requirements and existing capabilities, whilst at the same time positioning the involvement of initial customer.

2.9.3 Postponement
Across the world, postponement practice has been applied to a large extent. According to Yang (2009, p. 4), the supply chain as a whole, ranging from product design and development, manufacturing and to end products, has been covered by postponement. The author goes on to state that postponement was initiated as a marketing strategy to lessen risk and uncertainty costs, which are linked to highly dynamic demands by postponing the creation of time, place, form and ownership utilities. The paper also emphasises that product designs could eventually be changed, not only to fasten the response to unexpected changes in demand but also in an effort to deal with supply matters comprising unexpected changes early on in the product cycle. For this reason, postponement could potentially provide the opportunity to alter the configuration of one product at the very last likely minute lest there is existence of disruptions in supply of a component.
In this section, the paper is aimed at examining the definition of postponement, the classification of the different types of postponement strategies, and providing an explanation on the way in which postponement is used as a tool for managing uncertainty mitigation.
Can (2008, p. 6) further defines postponement as the process of delaying product finalisation in the supply chain until orders from customers are received with the aim of customising products, as opposed to performing those activities with the expectation of getting future orders. According to the author, this definition implies that organisations can delay production, distribution, packaging and assembling until they receive exact orders from their customers. The writer also cites that logistics postponement provides opportunities to locate inventory in any other place at any given time, which accordingly reduces the risk of being at fault. It can be concluded that, if the company delays the exact order for product distribution to local or international warehouses, the chances of decreasing risks become high in delivering products more than or less than needed. The writer asserts that the major intention of companies to apply postponement, in most cases, is to reduce the cost of distribution.
Can (2012, p. 5) refers to postponement as a concept that brings together the responsiveness of the agile concept and the efficiency of the lean concept. Postponement is defined as the delaying of activities in the supply chain up to that moment that customer orders are received with the intent of customising the products, as opposed to doing so in anticipation of future orders (Can, 2012, p. 5). This definition implies that firms can delay the production, packaging, assembling, distribution or even purchasing of raw materials until exact customer orders are received.

Figure 10: Speculation-postponement strategy and a continuum of standardization-customisation
(Source: Yang et al., 2004)

According to the review conducted by Can (2012), simulation studies demonstrate that postponement results in decreased inventory levels and manufacturing lead times. The studies reviewed also identify a reduction in uncertainty, which stems from the short-term dynamics in the supply chain as another area of application of postponement strategy. According to Can, responsiveness is improved by the type of postponement used, rather than through delivery reliability. Further review of previous studies by Can (2012) emphasise that higher postponement application causes increased on-time delivery performance, which, in turn, leads to lower operational cost (p. 10). In his discussion of postponement as a strategy of uncertainty management, Can recognises that the postponement concept makes it easier for manufacturers to forecast aggregate demand as opposed to forecasting the demand of each finished product. Secondly, the delay period enables the manufacturer to obtain more accurate information with regard to time, quantity and place.
Van Hoek (1999, p. 19) recognise that the principle of postponement can be categorised into three categories; that is, form, time and place postponement. Form postponement involves delaying activities that decide the form and function of products in the chain until customer orders have been received; time postponement refers to delaying the forward movement of goods until customer orders have also been received; the final generic type of postponement, as defined by the author, is place postponement, which refers to the positioning of inventories upstream in centralised manufacturing or distribution operations in order to delay the forward or downstream movement of goods.
A different paper by Hoi et al. (2007) recognises that the main concept of postponement is to pull and not push the process of manufacturing, and consequently shift inventory from finished goods to semi-finished goods and raw materials (p. 375). These explain that purchasing postponement refers to when a large part of inventory comprises raw materials, whilst production postponement occurs when the majority of inventory consists of semi-finished products. Product development postponement, on the other hand, occurs when manufacturers do not design the products until they receive the order.
From his study, Can (2008, p. 9) poses the argument that various research methods in the literature, such as surveys, cases and simulation studies, are used to analyse the benefits of postponement. In his review simulation studies, the studies demonstrate that postponement led to a reduction in inventory levels and manufacturing lead times. The author’s major finding when postponement was applied showed that postponement decreased the uncertainty due to short-term changes in the supply chain. Consequently, Can recognises that postponement improves responsiveness, but not delivery dependability. However, a different study carried out by a different author from Can’s paper argues that the advanced application of postponement enhances the on-time delivery performance—or rather, customer service—and brings about lower costs of operation.
Can (2008, p. 11) states that postponement is seen as one of the strategies for tackling uncertainty. The author suggests the postponement concept revolves around two major concepts: the first idea the author identifies is that it is straightforward to predict aggregate demand compared to predicting the demand of each end product; the second idea is that more accurate information—that is, place, time and quantity—can be attained at some point in the delay period. As a result, he suggests that, by altering the business processes based on the postponement strategy, companies can obtain the missing information, which is the cause of uncertainty. Moreover, the author points out that the relationship between postponement and uncertainty in the integration of supply chain has been investigated. The author inspected the relationship between postponement and uncertainty, and how uncertainty can be tackled. It was found that there exist two levels of uncertainty, namely the low level of uncertainty, which includes place and time utility of the customer order, and individual demand forecasts of the end products, and a high level of uncertainty, which involves the quantity and time utility of production and what needs to be produced.
Cavusoglu et al. (2012, p. 478) evaluated the importance of relationship between postponement and information-sharing, and accordingly identified that there are two different strategies centred on decreasing manufactures’ uncertainty regarding demands. The authors also point out that production postponement and information-sharing strategies may replace, harmonise or otherwise clash with one another, depending on the degree of the rise in the unit production costs during the period production is postponed. The paper suggests that demand uncertainty has encouraged manufacturers to devise diverse uncertainty reduction strategies. The authors provide an example of a Make to Order (MTO) strategy. Usually, a manufacture has to lay down his price but postpones production until the demand uncertainty is tackled. Consequently, manufacturers might decide to cut-down demand uncertainty by acquiring information from other available sources. The writers argue that information can constitute external entity: for instance, an agency that deals with marketing research inside the supply chain; thus, they suggest that the basis of large amounts of supply chain efforts have always involved sharing demand information between retailers and manufacturers. This because it is believed that retailers let out better information concerning customers demand than manufacturers.
Cavusoglu et al. (2012) extend their study by pointing out that both strategies of production postponement and information-sharing constitute a similar aim of decreasing a manufacturer’s uncertainty regarding demand. A limitation in the research that is worth noting is that the literature on production postponement has investigated the influence of postponement strategies on retailers, whilst literature on information-sharing has openly displayed retailers as not receiving direct assistance; in some cases, they get hurt when they provide their manufacturer with private information. Accordingly, it is concluded that, contrary to the notion that postponement and information-sharing in most cases substitute one another since both strategies aim at reducing demand uncertainty, the paper elaborates that, based on the relationship between the two strategies, postponement and information-sharing can harmonise each other or otherwise bring out conflicting sides. Consequently, the authors state that these two strategies, from a retailer’s perspective, in most cases conflict with each other.
Min & Mentzer (2004) observed that products would tend to be differentiated when they approach the purchase point. Postponement is a value-added process concerning products, where the common requirements for processing are at a maximum. In postponement, the unique processing requirements for all product varieties delay as much as it allows the value-adding process. Song et al. (2009) also provide a discussion of this concept in the service industry. When management carries standards components and then accordingly moves customisation downstream, they result in flexibility in using the same materials to satisfy the needs of different customers. In this case, customisation is deferred to a later stage. In postponed manufacturing, there is the separation of product customisation from speculative manufacturing (Nyaga et al., 2010). Postponement confers dividends of a more responsive supply chain, which satisfies the needs of individual customers without the necessity of incurring higher costs in inventory and production (van Hoek, 2001; van Hoek & Weken, 1998). In supply chain management, postponement helps in reduced cost obsolescence, lower overhead costs and short product development cycles (Feitzinger & Lee, 1997). Logically, additional information can be collected so as to reduce any uncertainty during the delay. Postponement, however, is only beneficial in unpredictable environments. Where there is predictability, postponement is not very useful. Towill & Christopher (2002) conclude this well by stating that the best method of coping with uncertainty is to ensure understanding of the major causes identified in the supply chain. Mason-Jones & Towill (1998) and Davis (1993) provide different sources of uncertainty, namely from the process side, the control side, the supply side and the demand side. Supply chain management using postponement therefore can be useful in dealing with the demand side of uncertainty.
As indicated, there are various industry-specific factors that influence the supply chain management techniques applicable to any particular industry (Ganeshan et al., 1999, p. 559). The application of postponement in the food industry has been done to a lesser extent when compared to other industries. Previous papers have shown this to be mainly in the packaging industries within the food industry. The limited adoption of postponement in the food industry has also been emphasised in the paper written by Hoek (1999), which aimed at examining the application of postponement in the food industries. The research paper was undertaken in Netherlands, Belgium and Germany. The paper focuses on the management of supply chains in the food industry, and begins by defining postponement as the delay in production until an order has been placed by a customer. He contends that it also may be applied to the distribution and purchasing. Hoek further outlines the benefits of postponement as the ability to customise the products whilst saving on the inventory and logistics costs; this is done with increased flexibility on the side of the company (Hoek, 1999, p. 18).
The paper by Hoek (1999) sought to establish three things. It begins by providing an examination of the extent to which the current postponement techniques have been adopted by the food industry. This builds the foundation upon which the issue of postponement will be analysed. He also seeks to determine the extent to which the level of postponement affects the amount of outsourcing undertaken by the firm. There has been increasing thought that postponement, amongst other factors, has an influential role to play in outsourcing (Rushton & Walker, 2007). Hoek further seeks to establish the need for the reconfiguration of the supply chain if one is to gain advantages of postponement. The paper uses empirical evidence from a research project titled ‘World Class Logistics and Postponement in Europe’. The analysis was undertaken with consideration to four main sectors: food, electronics, automotive supply and clothing. This comparison allows for a wider analysis that presents a more complete picture of the application of postponement depending on the industry. This is also able to indicate the sector-dependent factors that influence the strategy applied in the managing of supply chains (Hoek, 1999, p. 18).
A more in-depth analysis of postponement as a strategy, as undertaken in the paper written by Yang & Yang (2009), aimed at improving understanding of postponement. Indicating that globalisation has created increased competitiveness in the market place, which reiterates that concluded in numerous other studies, companies have been forced to identify better competitive edge-achieving methods. Yang & Yang argue that, due to certain factors, the implementation of these strategies may be limited; this is an area that also has been explored in previous other texts (Kazmi, 2007, p. 555). One of these factors is in the complexity of the management strategy, which could increase the overall complexity of the company’s operations. Yang & Yang further argue that there are also additional costs associated with the implementation of these strategies. Such additional costs, at times, may not be easily passed on to the consumer. In the long-term, this will limit the competitiveness of the company. The paper further points out that the management of supply chains is largely a factor of the uncertainty that may stem from interruptions in the global supply chain (Yang & Yang, 2009, p. 1901).
In the analysis undertaken in the paper, the writer’s main focus was centred on analysing the supply chain through the use of accident theory. Accident theory is the proposition that the characteristics of a given system determine its proneness to accidents (Khosrowpour, 2006, p. 160). In the paper by Yang & Yang (2009), the main area of consideration was the uncertainty raised from disruptions in supply, and not that which is raised by variation in the demand of a good. They considered how postponement may be applied in this case in an attempt to mitigate this form of uncertainty. They also indicate that, in a similar manner, this will have a greater influence on the operations of the company similar to that caused by demand uncertainty. One contribution made to the discussion through the work of Yang & Yang is that strategy postponement has been favoured in companies where Just in Time manufacturing has been adopted (Yang & Yang, 2009, p. 1901).
Yang & Yang (2009) also have considered the amount of interaction apparent between the company and those in other areas of the supply chain; this is to ensure that the communication that is necessary for the efficiency of the postponement strategy is well implemented. Communication already has been identified as an integral and important part of the adoption of both postponement and mass customisation in the supply chain (Kumar & Krob, 2005, p. 36). In the paper written by Yang & Yang, the concept of what they refer to as ‘modular design and interactive complexity’ has been discussed, indicating that, in the implementation of postponement, companies would need to reduce the level of complexity existent in the supply chain. The end result is that there is an improvement in the economics of operations (Yang & Yang, 2009, p. 1901).
Despite the increased amount of information presented in the paper, the focus may be limited to a certain degree. One of these limitations concerns the realisation that there are other areas of the supply chain that may be affected by postponement as a strategy. The paper did not study these other areas. Another limitation is that, in the study, the data used was of a secondary nature; this was at the expense of empirical data that would have aided in affirming the theoretical knowledge presented. The knowledge presented in the paper, however, does contribute to the discussions on postponement as a strategy, with the writers concluding by stating that further research may be undertaken on the subject of complexity, with consideration to how this affects the performance of the company (Yang & Yang, 2009, p. 1901).
A different paper by Yeung, Selen & Zhang continues to explore the concept of postponement, with special focus centred on the application of the concepts in the Chinese economy. In their examination, they are able to provide empirical evidence that seeks to confirm their postulations. The provision of empirical evidence also builds on the previous research by Yang & Yang (2009), where analysis was aimed at establishing the operational characteristics that would influence the application of postponement, as well as how these characteristics would do this.
The paper indicates that, in the application of postponement, all levels of the supply chain will need to be considered; this is by analysing both the upstream and downstream constituents of the supply chain. They also indicate the various ways in which the supply chain may be characterised. Furthermore, they also are able to give five classifications of postponements that are dependent on the characteristics explored in the research: manufacturer-dominated, customer-dominated, virtual supply chain, balanced structure with loose suppliers, and balanced structure with no customer information (p. 331). Balanced supply chain structure refers to a structure wherein all actors have equal powers, with neither party in the chain dominating the others. In balanced structure, where there is no customer information, there is no information-sharing between customers and manufacturers, which means production based on order is difficult (p. 350). In the balanced structure with loose suppliers, the modularisation of products can occur, with no information-sharing between the manufacturer and the suppliers. This requires raw material to be re-processed and kept in the work-in-process format (p. 350). In the customer-dominated arena, customers are very powerful, and manufacturers can customise their production processes to meet the requirements of their special customers. Manufacturers respond faster to orders and save costs by keeping inventory as raw materials rather than in a finished goods format (p. 350). The manufacture-dominated supply chain is where manufacturers are powerful and accordingly seek to minimise costs and reduce risks through the postponement of production and ensuring inventories are kept in raw material format (p. 350). In the virtual supply chain, there is a highly developed information system that connects all parties in the supply chain. The manufacturer is able to postpone the design and production until the customer places an order. The manufacturer then outsources the production function for the order. The classification of postponement has not been considered by most of the previous studies on the subject, and therefore will be explored by the present study.
Figure 11: Supply (Upstream) and demand (downstream) supply chain. Adopted from Jacobs (2006).

Despite the increased information provided in the paper, the concern comes with the consideration that the paper examined only a single economy; this may provide a biased point of view, where the information provided might not be applicable in other economies of the world. The paper further analyses a limited number of companies (8 in total) of a given region, which further increases the bias in the findings (Yeung et al., 2007, p. 331). These two factors determine the effectiveness of the results. A limited sample size and the use of a single country to find the data needed in the study limited the diversity of the data and, as a result, the external validity of the conclusions drawn (Craig, 2005, p. 291).
Another important study on the application of postponement practices to mitigate demand uncertainty is that conducted by Cholette (2009), who developed a mathematical model for postponing channel differentiation as a way of mitigating demand uncertainty in winery sales channels. The results obtained from solving the model demonstrated that holding considerable portion of production at the packaging and labelling levels significantly improved product profitability. Cholette further investigated how different product restrictions and configurations can affect the level and extent of the postponement strategy, and accordingly identified the strategy as suitable over a range of costs, demand probabilities and other considerations (p. 3587). This study therefore provides evidence in support of postponement as a suitable strategy for mitigating demand uncertainty in the food industry.
Cholette (2009) explains that, in spite of the postponement concept being introduced into management literature in the 1950s, it has received less attention by practitioners (p. 3588). According to Cholette, the actual usage of the postponement strategy has been documented mainly in the automotive and other high technology industries, which have achieved successful adoption compared to other verticals (p. 3588). The author notes that the agribusiness has considerably lagged behind other sectors in adopting postponement practices; however, he attributes the lag in the adaptation to a lack of capability to employ processes that support and facilitate postponement, such as information coordination technologies, as well as the general incapability to modularise product design for enhancing component commonalities (p. 3588). Cholette also acknowledges that a firm’s ability to change processes to facilitate postponement—or to adjust the organisation’s mindset to consider the employment of such techniques—is an important issue that should not be ignored. The author identifies change management as an important barrier to the effective implementation of postponement. Cholette further concludes that, when well-implemented, postponement can be an effective strategy in mitigating demand uncertainty, particularly in an intensely competitive and globalised market environment.
2.9.3.1 Postponement and OCPD:
According to these authors, Decoupling Point (DP) is strongly linked to postponement, and requires the manufacturer to be very careful about where the DP is to be located. The level of postponement also is related to DP. Information flow and material flow perspectives can both be used in explaining how postponement can be used to locate the DP (Yank & Burns, 2003, pp. 2079–2081).
According to Can (2012, p. 39), the main reason firms postpone some of their operations is owing to a lack of customer order information. In his explanation, Can points out that, when needing to provide efficiency in utilising postponement as an uncertainty management strategy, some operations that require information about customer orders need to be moved downstream; in exchange, some of those that do not require customer order information or those that can be done in anticipation should be moved upstream. If not, the firm has to deal with high volumes of work in progress inventory, as well as long lead times, which might result in the stoppage of material flow. The point at which the customer order is linked to or seen to penetrate the system is CODP (p. 39). It is the point at which the missing information causing postponement is held. The relationship between postponement and OCPD therefore is very strong.

2.9.3.2 Lean and Agile Supply Chain:
According to Can (2008), the term ‘leagility’ refers to the combination of lean and agile concepts within the supply chain strategy by placing the decoupling point (DP) so that the organisation is best fit to respond to a volatile demand downstream whilst at the same time enabling level scheduling upstream from the DP. This simply involves efficiently producing customised goods in a manner that is similar to mass production. Agile supply chains are aimed to be market sensitive. According to Can, lean and agile SC increases the competitiveness of an organisation in the market. The leagility concept is illustrated in the figure below.

Figure 12: Lean, agile and leagile supply: Adopted from Mason-Jones et al. (2000)

According to Can (2008, p. 43), mass customisation, in nature, demonstrates a leagile SC, emphasising the importance of responsiveness and efficiency. Leagile SC essentially implies producing customised products in a manner that is efficient, as is done in mass production. The lean aspect of leagile supply demonstrates the mass aspect of mass customisation, whilst the agile aspect of leagile demonstrates the customisation aspect of mass customisation (Can, 2008, p. 43).
With regard to postponement, it is argued that postponement is an important factor that contributes to the achievement of agility through its contribution to the customisation of services and products, cross-functional efforts, and the use of information on customer order through the SC. Postponement therefore is regarded as vital for any agile strategy, and since it works towards moving CODP downstream, it increases the overall effectiveness and efficiency of the SC. Postponement therefore is considered a concept that contributes to efficiency (lean) and responsiveness (agile) (Can, 2008, p. 43).
Leagility is considered to be the link between MC and postponement. MC naturally requires a leagile SC owing to the fact that both efficiency and responsive are the key principles for the success of the MC strategy. Postponement, on the other hand, has been identified as contributing to efficiency and responsiveness; therefore, Can (2008) argues that postponement plays a role in the leagile SC of mass customisation. The MC strategy therefore has the capability of producing a market winner in all competitive priorities, through customisation (product design), customer satisfaction through the various appropriate production systems (regarding cost, quality and delivery), which are associated with the MC strategy (Can, 2008, p. 45).

2.9.3.3 The Figure of Market Qualifier and Winner Regarding Lean and Agile SC:
According to Mason-Jones et al. (200), the lean production concept can be applied across the value stream in an effort to eradicate waste and accordingly provide customers with real value. They explain that profit maximisation results from cost minimisation, which, in turn, is a direct outcome of elimination. The lean system framework enables a firm to attain waste elimination by smoothing the demand, thus resulting in a level schedule. Market situation also may call for cost leadership; in which case, the criteria for winning an order is cost, or may be demand service leader, where the winning criteria is service-level, or a combination of the two. As the figure below illustrates, quality, service level and lead time are market qualifiers for the agility concept. However, cost, which could be a market qualifier for the agility concept, is also a market winner for the lean concept. The connection between the two order winning criteria centres on the fact that, whilst service level needs to maximised, cost needs to be minimal so as to improve performance across the supply chain (Mason-Jones et al, 2000). According to these authors, the application of the lean tool in the SC is to address the benchmark for the lean market winner. This cost comprises all distribution, storage and production costs in the SC. Service level—also a market qualifier under the lean concept—also turns out to be a market winner for the agility concept.
Figure 13: Market Qualifier and winner, regarding to Lean and Agile SC

2.9.3.4 Lean Management:
Lean waste management identifies eight types of waste, each discussed below by Olivella et al. (2008). Notably, the 8 wastes of lean management include over-production, inventory, waiting, defects, motion, transportation, over processing and, incorrect use of staff and their abilities.
Over-production refers to when a firm produces more than is required by the customer. Over-producing impacts process chains, inventory and transportation costs, results in waiting, increased defects, and motion.
Inventory motion waste occurs when quantity of parts in stock are not being utilised in production and therefore take up valuable space. They could become obsolete whilst in stock.
Defects usually result in the reworking, re-processing and scrap as a result of products that are defective and need to be disposed or otherwise reworked, which results in costly processes.
Waste due to waiting results when tasks in the manufacturing process upstream and downstream are delayed due to delays in information, materials, equipment and operators. These delays could accumulate and result in losses.
Excessive transportation waste; including the unnecessary movement of items, finished goods, materials, parts and information from one place to another, all of which waste resources, time and money.
Motion–unnecessary motion is associated with staff and, more specifically, operators moving about the work space wasting effort and time. This can be caused by poor practices, standards and process design.
Over processing involves taking unnecessary steps throughout the production/manufacturing process. This wastes time, resources, materials and equipment.
Incorrect use of staff and their skills as well as abilities could result in missed improvement and learning opportunities. They should be used to eliminate the other aforementioned wastes.

Table 1: Adopted from Erriah (2013)
Type of WasteDescriptionExample within ManufacturingExample Symptom1T: TransportMoving the product to several locations. Whilst the product is in motion it is not being processed and therefore not adding value to the customer.Raw materials are made in several locations and transported to one site where a bulk intermediate is made. This is then transported for final product processing.
Packaging for customer use may be at a separate site.Movement of pallets of intermediate product around a site or between sites.
Large warehousing and continual movement of intermediate material on and off site rather than final product.2I: InventoryStorage of products, intermediates, raw materials, etc. all cost money.Economically large batches of raw material are purchased for large campaigns and sit in the warehouse for extended periods.
Queued batches of intermediate material may require specific warehousing or segregation especially if the lab analysis is yet to be completed or confirmed.Large buffer stock within a manufacturing facility and large warehousing on the site; financially seen as a huge use of working capital3M: MotionThe excessive movement of the people who operate the manufacturing facility is wasteful. Whilst they are in motion they cannot support the processing of the product.
Excessive movement of data, decisions and information.People required to move from one area to another in order to move the product along the manufacturing cycle.
People transporting samples or documentation.
People required to move work in progress to and from the warehouse.
People required to meet with other people to confirm key decisions in the supply chain process.Operators moving to and from the manufacturing unit but less activity actually within the unit.4P: PeopleOver use or under use of human resources.Not having enough staff to carry out duties.
Using over qualified people to carry out certain duties.
Using extra human resources to assist fully automated processesExcessive over-time.
Senior managers manning the manufacturing plant.5W: WaitingAs people, equipment or product wait to be processed, it is not adding any value to the customer.Storage tanks acting as product buffers in the manufacturing process- waiting to be processed by the next step.
Intermediate product which can’t leave site until the lab tests and paperwork are complete.The large amount of ‘Work In Progress’ held in the manufacturing process, often seen on the balance sheet and as ‘piles of inventory’ around the site.6O: Over ProductionProduct made for no specific customer.
Development of a product for no additional value.Large batch campaign, continuous large scale manufacturing processes.
Development of alternative process routes which aren’t used for the development of processes which don’t support the bottleneck.
Redesign of parts of the manufacturing facility which are ‘standard’, e.g. reactors.The extent of warehouse space needed and used.
Development and production organisation imbalance.
An ever changing process.
Large engineering costs/ time associated with facility modifications7O: Over ProcessingWhen a particular process step does not add value to the productA cautious approach to the design of unit operations can extend processing times and can include steps, such as old or testing, which add no value.
The duplication of any steps related to the supply chain process, e.g. sampling, checking.The reaction stage is typically complete within minutes yet we continue to process for hours or days.
We have in process controls which never show a failure.
The delay of documents to accompany finished product.8D: DefectsErrors during the process- either requiring re-work or additional work.Material out of specification; batch documentation incomplete.
Data and data entry errors.
General miscommunication.Missed or late orders.
Excessive overtime.
Increased operating costs.

3.0 Conclusion
Although a host of literature concerning the causes and appropriate managerial strategies of mitigating demand uncertainty is available, it is prudent that more research be done in the same field. This will result in the identification of more sources, which will prove beneficial to the supply chain, enabling organisations to completely counteract demand uncertainty. This is highly beneficial for the SME food industries, especially in the Hajj season, where there is a huge preference of customer demands.
In addition, further research will emanate more empirical evidence to verify the facts elaborated in the available literature concerning demand uncertainty. In addition, further study will enable individuals to classify the available sources of the uncertainties as general in terms of managerial organisation or have an industrial context. In the same way, no literature review has incorporated the three categories of uncertainty as the cause of demand uncertainty, with studies on classification of postponement also scarce. Therefore, there is need for contingency-based studies and research to be performed in an effort to determine whether such a scenario exists.
This study has fulfilled earlier calls for research with literature gaps, such as those made in the paper of Simangunsong et al. (2011), which sought to determine a comprehensive understanding of the many sources of uncertainty and how these can be aligned with management strategies in an effort to improve supply chain performance, thereby developing theory in this area. Previous research mainly has focused on the theory of the SCM paradigm in general, supply chain risk, and on narrower aspects of uncertainty, including supply and demand uncertainty only. Conversely, this study seeks to address the identified gaps by exploring ways of mitigating demand uncertainty through the discussed management practices.
Specifically, this research is a response to the call for future research by Lai et al. (2012) for future studies to examine how firms choose MCC strategies to mitigate demand uncertainty when there is high competitive intensity. Moreover, this study reveals some complex relationships between SCI and MCC. The mechanism of how SCI influences mass customisation strategies requires further investigation.

Figure 14: The Populated Model of supply Chain Uncertainty (Simangunsong et al., 2011)
CHAPTER 3: THEORETICAL AND CONCEPTUAL FRAMEWORK

3.1 Contingency Theory and its Application in Supply Chain Management
One of the theories that have been widely used in the analysis of management practices is that of contingency theory, which also will be used in the current study. Contingency theory asserts that an organisation should match its processes, strategies and practices to its business environment (Donaldson, 2001, pp. 23–24). In this vein, Donaldson describes this as a behavioural theory that argues there is no best way of managing, organising and leading a corporation or otherwise making decisions. Rather, it explains that the best course of action is dependent (contingent) on the internal and external situation. This theory maintains that the most effective organisational leadership or structural design involves the structure matching the contingencies. According to Flynn et al. (2010, p. 59), this theory employs a reductionist approach, where the organisation is treated as an entity that can be decomposed into various independent elements. This theory has been applied throughout the course of many empirical studies on management practices in supply chain.
Due to its application in the many different past research papers, such as that written by Huang, Kristal & Schroeder (2010), an examination of the contingency theory is required (Huang et al., 2010, p. 515). Its study will also stem from the realisation of its importance in the study of the management of supply chains; this is in the analysis of the strategies that may be applied in supply chain management. The transaction cost theory may also be used (Plaggenhoef, 2007, p. 69). An introduction into contingency theory is provided by the paper written by Donaldson (2006), which defines the contingency theory as the management approach that looks at the most efficient way of managing the supply chain, taking into consideration all the contingencies that may be present within a given operation. He indicates that there are various challenges, which are both empirical and theoretical in nature, in the investigation of contingency theory and its application to the study of supply chain management. The paper further presents the opportunities apparent in terms of contingency theory discussions (Donaldson, 2006, p. 19).
Another study that applied this theory is that by Lai et al. (2012) through the examination of the effect of SCI on MCC. These authors point out that previous studies have demonstrated that the value of a firm’s external resources increases in an environment that is dynamic and competitive. According to Lai et al. (2012), provide the example of when competitive intensity and/or demand uncertainty is high, postponement, a key strategy that enables mass customisation to be employed to cope with variability in the end product. Other than having efficient coordination across the firm’s internal functions, accurate and timely market-specific and component knowledge is also necessary when postponing differentiation. Improving internal integration and attaining knowledge from supply chain partners was considered more vital in this case for postponement strategy and Mass Customisation Capability development (Lai et al., 2012). These authors argue that Supply Chain Integration therefore is very important for MCC development, particularly in a dynamic competitive milieu.
Another important study that has applied this theory in the investigation of the impact of SCI on performance is that conducted by Flynn et al. (2010). In this study, the authors examine the relationship between three SCI dimensions. The study applied the contingency approach through the use of a hierarchical regression in an effort to establish the impact of individual SCI dimensions, namely internal, supplier and customer integration, and their interrelationships on firm performance. In their discussion of contingency approach to Supply Chain Integration, Flynn et al. (2010) draw from the contingency theory, and accordingly explain that the processes and structures of an organisation are shaped by the environment within which it operates (p. 59). Organisations therefore must match their processes and structures to their environment so as to maximise performance. These authors are quick to add that customers and suppliers form an important part of an organisation’s (particularly manufacturer’s) environment. These authors further mention structural contingency theory, which proposes that the success of how a firm performs depends on the degree to which the strategy it seeks to pursue matches or is aligned with the firm’s design (p. 60). Literature on strategic management refers to this alignment between a firm’s strategy and its performance as ‘fit’. When applied to Supply Chain Integration, this theory suggest that the individual types of SCI (its dimensions, i.e. customer, internal and supplier integration) should be aligned for best performance in order to be achieved (p. 60). This study therefore will apply structural contingency theory in its investigation of how demand uncertainty can be mitigated in the Saudi SME food industry during the Hajj season through supply chain management practices, including SCI, MCC and postponement.
Simangunsong et al. (2011) also applied this theory in their study. According to them, the manufacturing theory recognises that manufacturing strategy is significantly influenced by environmental uncertainty, which is also a key determinant of firm performance (p. 4501). These authors argue that this theory has been linked by various scholars to contingency theory; for that reason, it can be described as based on a contingency model. The contingency theory suggests that the most appropriate management strategy in a particular context depends on a set of contingency factors, possibly including the uncertainty of that environment (p. 4501).
Simangunsong et al. (2011) argue that, given the applicability contingency and alignment theories, manufacturing strategy theory can be modelled so as to provide a strong theory to guide future research in supply chain uncertainty, which includes a broader set of sources of uncertainty than those considered by previous studies. These authors developed a contingency-theory based model of supply chain uncertainty, which will be adopted in the present study.
Figure 15: Contingency theory-based model of supply chain uncertainty (Adopted from: Simangunsong et al., 2011)

3.2 Resource-Based View
In addition to the contingency theory, this study will use the Resource-Based View (RBV) theory, which asserts that, simultaneously, valuable, rare and non-substitutable and inimitable resources can be effective sources of superior performance, which may enable enterprises to attain sustainable competitive advantage (Ambrosini et al., 2009, p. 59). This theory has been used extensively in research in an effort to explain managerial practices and strategies centred on enhancing organisational performance. Important studies on the mitigation of supply chain uncertainty, such as that by Simangunsong et al. (2011), also have applied this theory in their research.
According to Vijayasarathy (2010, p. 490), the RBV theory is one of the perspectives most widely adopted in supply chain studies. From the perspective of RBV, it is suggested that firms can gain sustainable advantage by developing and acquiring infrastructural resources as well as knowledge-based capabilities that are difficult for competitors to replicate. According to this author, this theory has been used by scholars to explain why organisations seek integration with their customers and suppliers, and also in predicting the rewards and benefits of such integration to organisations (p. 490).
Wernerfelt (1984) explored the usefulness of analysing organisations from the resource perspective as opposed to the product perspective. The study suggests the application of concepts of resource–product matrices and resource position barriers in the development of entry barriers. The study uses these tools in an effort to explain the new strategic options that can be used by firms in their supply chain and which are seen to naturally arise from the resource perspective (p. 171). This study argues that both the resource and products perspectives of the firm have been well addressed in strategic management literature. According to the study, the concept of strategy traditionally has been expressed in terms of the firm’s resource position (its weaknesses and strengths), whilst most of the existing formal economic tools function based on the product-market side (p. 171). This author further explains that, although the two perspectives ultimately produce the same insights, it is expected that the insights yielded differ in ease depending on the point of view adopted. Wernerfelt (p. 172) argues that the resource point of view provides the basis tackling of a number of issues that are key in the formulation of strategy for diversified firms:
a) Which of the firm’s current resources should base diversification.
b) Which of the firm’s resources should be advanced through diversification.
c) In which markets and sequence diversification should occur.
d) The type of firms desirable for acquisition.
Wernerfelt (1984) refers to ‘resource’ as anything that may be regarded as the strength or a weakness of a particular organisation (p. 172), defining a firm’s resources as those assets (tangible and intangible) that are semi-permanently attached to the firm. Examples include in-house technological knowledge, skilled personnel, machinery, brand names, efficient procedures, capital and trade accounts, amongst others. Wernerfelt suggests ‘customer loyalty’ as one of the attractive resources a firm can possess (p. 174). In relation to customer satisfaction, Wernerfelt further explains that the nature and characteristics of the market for the resource creates the resource position barrier. He notes that it is much easier to initiate a position (be the pioneer) than to replace a person who already has it. In replacement, buyers pay higher prices than in pioneering. He cites first-mover advantages in access to raw materials or gaining government contracts as some of the related examples (p. 174).
The study conducted by Pertusa-Ortega et al. (2010) applied the RBV theory in examining the relationship between organisational structure, firm performance and competitive strategy. This study emphasises that strategy determines the structure of a firm. In their review of literature, Pertusa-Ortega et al. (2010) note that modern-day firms operate in a rapidly changing business environment that is characterised by volatile customer preferences and technological developments that transform business scenarios (p. 1285). They argue that, in such a context, the RBV is based placed on explaining the sources of competitive advantage for such firms. They further emphasise that the definition of a business firm with regard to the internal resources it possesses, as well as its capabilities, offers a better and more durable strategy than one based on the requirements and needs the business firms seeks to satisfy. These authors argue that, although the contingency theory suggests that external business environment and strategic decisions impact a firm’s organisational structure in its successful implementation, RBV stresses on a firm’s internal attributes and accordingly allows researchers to reconstruct the relationships between structure and strategy by evaluating the organisational structure as a precious resource and an important source of competitive advantage. In this study, Pertusa-Ortega et al. (2010) use the RBV theory to demonstrate that the organisational structure influences the amount and type of information that a firm can obtain and distribute, the knowledge created, and the adoption of strategic decisions; all of these affect the configuration of the strategy with which the enterprise competes in the market (p. 1293). According to the findings of the study, RBV complements the contingency approach in the explanation of organisational performance (p. 1294). This study concludes that RBV is adequate in management decisions concerning strategies for improving firm performance. As such, it is appropriate for guiding the current study.

3.3 Relationship between the Various Constructs
3.3.1 The Relationship between II, CI and SI
According to Zhao et al. (2011, p. 19), the relationship between II and external integration (SI and CI) remains unlimited, as most studies have discussed only the relationship conceptually without providing details of empirical evidence. However, most studies demonstrate that II positively impacts external integration, which is made up of SI and CI as its sub-dimensions through information-sharing and product development. Lai et al. (2012, pp. 446–447) explain that II facilitates the two sub-dimensions (SI and CI) of external integration (EI). According to these authors, resources encapsulated within the organisation provide a platform and a foundation for the acquisition of external resources. This argument proposes that II may help an organisation facilitate EI with its suppliers and customers (Lai et al., 2012, p. 446). In their discussion of the relationship between II and EI (SI and CI), Zhao et al. (2011, p. 19) argue that, from the point of view of organisational capability, a firm with a high level of internal coordination and communication capabilities is better positioned to attain a high level of EI with its suppliers and customers. They further explain that a company with a high level of absorptive capability is seen to have the ability to recognise the importance and value of new external information, and to incorporate and apply it to its commercial needs. A company that is able to interpret, disseminate, apply and evaluate new knowledge obtained from external customers and suppliers will be better able to understand their businesses, hence facilitating EI.
The effect of II on EI (SI and CI) can also be explained from the three major features of SCI: strategic cooperation, working together and information-sharing (Zhao et al., 2011, p. 19). With regards to information-sharing, it is obvious that a firm cannot share information with its external partners and integrate the data whilst also effectively sharing it with its internal functions and units if it does not have internal processes and systems, such as ERP systems. Well-established internal processes, systems and capabilities therefore enhance effective and timely information-sharing amongst suppliers and customers (Zhao et al., 2011, p. 19). They also facilitate the integration and sharing of information across the firm’s internal units. The same can be said of working together with external suppliers and customers, as well as in the development of strategic alliances and cooperation. II therefore facilities CI and SI.
H1: Internal integration has a direct impact on customer integration.
H2: Internal integration has a direct impact on supplier integration.
3.3.2 The Relationship between SCI with Postponement
The relationship between the various forms of SCI (II, CI and SI) with postponement is discussed from the perspective that information-sharing is key to production postponement. According to Cavusoglu et al. (2012), production postponement and information-sharing tactics may complement conflict or substitute one another depending on the degree of the increase in cost of unit production when production is postponed. Can (2008, p. 6) defines postponement as the process of delaying product finalisation in the supply chain until orders from customers are received; this has the aim of customising products, as opposed to performing those activities with the expectation of gaining future orders. Postponement basically is dependent on II, CI and SI owing to the fact that information is required from the various partners so as to establish when to hold production and when to continue.

3.3.3 The Relationship between SCI and MCC
According to Lai et al. (2012, p. 444), internal integration links various functions, allowing firms to establish strategic resources. According to these authors, manufacturers use cross-functional coordination and alignment so as to integrate resources across the firm and to deploy them in a systematic manner. Inter-functional relationship management also ensures that the process is cooperative, and that any conflicting departmental interests are resolved. It provides a platform for different departments functions and departments to merge their opinions and suggestions, and to integrate all resources through cooperation and working together (Lai et al., 2012, p. 444). The integrated procedures and operational routines, in turn, facilitate the creation and utilisation of resources and improve problem-solving, which then creates and increases organisational capability. According to the study, effective internal integration enables a firm to quickly respond to the customisation needs of its customers and accordingly effectively address the challenges associated with product complexity, flexibility and variety, and the costs related with the development of MCC.
Lai et al. (2012, p. 446) recognise that, although II facilitates the management of internal resources, successful MCC development also requires the application of external resources that may be obtained through supplier and customer integration. Information-sharing with customers and suppliers enables manufacturer to gain knowledge regarding demand, raw materials, markets and components. Close relationships and engagement with suppliers and customers during product design incorporates their knowledge and voices into the manufacturing process, thus leading to more efficient and effective customisation. Based on this discussion, Lai et al. (2010, p. 446) conclude that internal integration, customer integration and supplier interaction all affect MCC in a positive manner.

3.3.4 The Relationship between II and PP
Internal Integration (II) involves mainly joint decision-making, internal relationship management and cross-functional coordination. Kotcharin et al. (n.d., p. 1631) refers to II as the extent to which an organisation can plan its organisational practices, procedures and behaviours into joint, synchronised and manageable processes with the aim of achieving customer desires and needs. As such, it has a direct effect on postponement, enabling the organisation to make the correct decisions regarding when to hold production and when to continue after receiving orders from the customers. Therefore, we hypothesise the following:
H3: Internal Integration positively affects postponement.

3.3.5 The Relationship between II and MCC
According Lai et al. (2012, p. 446), the ERBV proposes that manufacturers can utilise internal and external resources towards capability development. The authors also mention that manufactures should incorporate both types of resources in an effort to come up with a hierarchy in which the extent of knowledge and resources is expanded as it shifts up the hierarchy. Therefore, the resources summed-up within the organisation create a foundation for the attainment of external resources. Lai et al. further argue that internal integration could possibly assist an organisation in facilitating external integration with both consumers and suppliers. The authors also mention that internal integration can increase the intensity of the capability and assist a manufacturer in creating a cohesive platform that breaks down internal subdivisions, tackles conflicts and reduces the obstacles facing supplier and customer participation. In this way, internal integration is presumed to have an indirect influence on MCC by enhancing external integration (p. 446). Internal integration therefore indirectly affects MCC through customer integration and supplier integration, and directly by enabling the firm to quickly respond to the customisation needs of its customers and effectively address the challenges associated with product complexity, flexibility and variety, and the costs related to the development of MCC. Therefore, we hypothesise the following:
H4: Internal integration positively affects mass customisation capability.

3.3.6 The Relationship between CI and PP
Customer Integration (CI) mainly involves customer partnership, the sharing of customer information, and ensuring customers are involved in product development and delivery. According to Lai et al. (2012, p. 444), customer integration is important for manufacturers as it allows access to customer information, knowledge-sharing, the pursuit of joint development activities, the speeding up of the decision-making processes, a reduction in lead times and improved process flexibility. This integration has a direct impact on postponement as it provides firms with timely information concerning customer orders, needs, preferences and requirements, meaning they only go on with production when they need to (once they have the orders) and then produce customised goods. Therefore, we hypothesise the following:
H5: Customer integration positively affects postponement.

3.3.7 The Relationship between CI and MCC
CI involves mainly customer partnership, sharing customer information, and customer involvement in design and delivery of products (Flynn et al., 2010, p. 59). CI allows manufacturers to access customer information, share this knowledge, speed-up decision-making processes, improve process flexibility, reduce lead times and pursue joint development processes and activities (Lai et al., 2012, p. 444). Furthermore, CI is important as it enables manufacturers to acquire information regarding customer requirements and also to gain a better understanding of customer needs and preferences. Therefore, we hypothesise:
H6: Customer integration positively affects mass customisation capability.
3.3.8 The Relationship between SI and PP
Supplier Integration (SI), on the other hand, entails mainly supplier partnerships, supplier information-sharing and the involvement of suppliers in product development (Lai et al., 2012, p. 444). According to Flynn et al. (2010, pp. 59–60), developing close ties with suppliers enables service providers and manufacturers to gain greater inputs from the suppliers and also to include their suggestions and recommendations into business operations. CI has a direct effect on postponement, facilitating the smooth delivery of various raw materials and components on a timely basis, thereby enabling the manufacturer to reduce total lead time for the delivery of customised goods once orders are in place. Therefore, we hypothesise the following:
H7: Supplier Integration positively affects postponement.

3.3.9 The Relationship between SI and MCC
SI mainly entails supplier partnerships, the sharing of information with suppliers, and involving them in product development (Lai et al., 2012, p. 444). This integration enables manufacturers to gain greater inputs from suppliers, and also to include their suggestions and recommendations within their business operations. This further facilitates the smooth and timely delivery of a variety of raw materials and components for mass customisation. Therefore, we hypothesise the following:
H8: Supplier integration positively affects mass customisation capability.

3.3.10 The Relationship between Mass Customisation and Postponement
According to Can (2008, pp. 5–6), the resolution to successfully deal with mass-customising involves postponing the role of distinguishing a product for a particular customer to the level of the most recent possible point in the supply network, which includes the supply chain, manufacturing chain and distribution chain of an organisation. Consequently, the author states that, in order for mass customisation to be enhanced effectively and a sustainable response garnered, organisations ought to incorporate product designs, manufacturing and logistics procedures, and supply network. For this reason, it is assumed that, for postponement in the differentiation of mass customisation to take place, suitable product design, processes and supply network are relevant. On the other hand, however, the connection between mass customisation and postponement can be referred to by the concept of leagality; as defined by Can, this is a combination of the lean and agile paradigm in a total supply chain strategy, positioning the decoupling point with the objective to best suit the necessity for responding to a volatile demand downstream, whilst at the same time presenting level scheduling upstream from the decoupling point.
According to Can (2008), postponement has been classified as a significant approach for determining the realisation of agility: for instance, in the course of its involvement in products and services customisation, as well as its involvement in customer order information through the supply chain, it can be concluded that postponement is viewed as a concept promoting both lean and agile factors in an organisation, hence helping address demand uncertainty. Therefore, we hypothesise the following:
H9: Postponement has a direct, positive relationship with mass customisation capability.

3.3.11 Contingent Effects of Demand Uncertainty and Competitive Intensity
Demand uncertainties, in addition to competitive intensity, are presumed to be key environmental conditions designed for MCC. Lai et al. (2012, p. 447) assert that there is empirical evidence to suggest that the influence of Supply Chain Integration towards operating capabilities possibly could be moderated by the context of the environment. For instance, when there is a rapid shift in demand, manufactures require new knowledge to control customisation owing to the fact the existing experience swiftly turns out invalid. On the other hand, however, when there is low level of demand uncertainty, manufacturers create their MCC through designs, productions, and the delivery of customised products, all of which depend on existing resources and knowledge (p. 447). Tackling demands that are unpredictable and subsequently developing customised products requires the united efforts of the partners in the supply chain, such as by working together with customers. Therefore, when firms increase collaboration with their suppliers, the organisation is positioned to explore and improve the variety of possible solutions for tackling customers’ needs and accordingly lowering costs and lead times through improvement in joint processes (Lai et al., 2012, p. 447). With this noted, it is concluded that, in an environment characterised by demand uncertainty, manufactures can improve the influence of internal integration on external integration and consequently can develop and improve MCC. Based on this, it is argued that demand uncertainty increases the indirect impact of II on MCC through customer and supplier integration.

3.3.12 The Relationship between II, SI, CI, PP and MCC with Demand Uncertainty Mitigation (DUM)
Drawing from the extended RBV of the firm, Lai et al. (2012, p. 444) argue that all three types of Supply Chain Integration (II, SI and CI) influence the development of MCC within a firm owing to the fact that both internal and external integration promote the strategic resources considered crucial to MCC development. The authors also mention that II can increase the intensity of capability and thereby assist a manufacturer in creating a cohesive platform that is able to break down internal subdivisions, tackle conflicts and reduce the obstacles regarding supplier and customer participation. In this way, II is presumed to have an indirect influence on MCC by enhancing external integration (SI and CI).
II, SI and CI also have been identified as having a direct effect on postponement, particularly through information-sharing. Postponement is defined as the delaying of activities in the supply chain up to the moment customer orders are received with the intent of customising the products, as opposed to doing so in anticipation of future orders (Can, 2012, p. 5). According to Flynn et al. (2010, pp. 59–60), postponement basically is dependent on II, CI and SI, owing to the fact that information is required from the various partners in order to establish when to hold production and when to it can be continued. Cavusoglu et al. (2012, p. 478) evaluated the importance of relationships between postponement and information-sharing through II, SI and CI, and found that they are different strategies to decrease manufactures’ uncertainty regarding demands.
According to Can (2008, p. 6), postponement also is directly related with MCC as it explains that mass-customising involves postponing the role of distinguishing a product for a particular customer to the level of the most recent possible point in the supply network, which includes the supply chain manufacturing chain and the distribution chain of an organisation. Therefore, we hypothesise the following:
H10: Postponement positively mitigates demand uncertainty positively.
H11: Mass customisation capability positively mitigates demand uncertainty.

3.3.13 Conceptual Framework and Hypotheses
According to Luhmann (2005), the competitive intensity of any organisation is affected by several supply chain factors; such factors help in demand uncertainty mitigation. Under Supply Chain Integration, there is customer integration, supplier integration and internal integration. All of these are seen to affect the postponement practices and Mass Customisation Capability. In turn, these help in the mitigation of demand uncertainty.

Figure 16: Conceptual framework used in the current study, adapted from Lai, Zhang, Lee & Zhao, 2012
3.4 Summary of Literature Review
Table 2: Key term sources
SubjectAuthorsSME Development
Development of SMEs in Saudi

Supply chain management

Management Practices

Supply Chain Uncertainty
Demand Uncertainty

Supply Chain Integration (SCI)

Mass Customisation Capability

Modularisation

Customer Order Decoupling Point

Mass Customisation and Customer Order Decoupling Point

Postponement

Postponement and OCPD

Mass Customisation and Postponement

Lean and agile supply chain

Contingency Theory and its application in SC
Resource Based View Otsuki (pp. 1–2), Shalaby (2012), Almosallam (2008), Bundagji (2005).

Al-Awaad (2007), Almosallam (2008).

Shalaby (n.d.), Otsuki (2002).

Christopher (2005), Mentzer et al. (2004), Krajewski et al. (2007), Angerhofer et al. (2000), Kelton et al. (2003), Carvalho et al. (2012), Li & Schulze (2011), Vijayasarathy (2010).

Wong et al. (2005), Cohen & Roussel (2005), Li et al. (2005), Bajpai (2011), Schermerhorn (2010), Qi et al. (2011), Kocoglu et al. (2011), Yao & Song (2001), Mikkola & Skjøtt-Larsen (2004).

Hillson (2006), Li & Hong (2007), Simangunsong et al. (2011), Li & Hong (2007), Hult et al. (2010), Lai et al. (2012), Flynn et al. (2010), Lie et al. (2012) , Luhman (2005), Burgess et al. (2006).

Chen & Paulraj (2008), Lai et al. (2012), Amit et al. (2005), Simangunsong et al (2011), Liu et al. (2012), Choi & Cheng (2011), Boyle et al. (2008), Braunscheidel & Suresh (2009), Amit et al. (2005), Burgess et al. (2006).

Huo (2012), Flynn et al. (2009), Lai et al. (2012), Rungtusanatham et al. (2003), Towill & Christopher (2002), Song et al. (2009), Min & Mentzer (2004), Flynn et al. (2010), Zhao et al. (2008), Lau et al. (2012), Koçoglu et al. (2011), Huo (2012).

Davis (1987), Lai et al. (2012), Can (2012), Fogliatto et al. (2012).

Can (2012).

Olhager (2003), Can (2012), Can (2008), Rudberg and Wikner (2004).

Can (2008), Yang (2009), Can (2008), Can (2012), Yang et al. (2004), Hoi et al. (2007), Cavusoglu et al. (2012), Song et al. (2009),

Can (2008), Yang (2009), Can (2012), Hoi et al. (2007), Cavusoglu et al. (2012), Nyaga et al. (2010), Hoek (1999), Yang & Yang (2009), Yeung et al. (2007), Cholette (2009).

Yank & Burns (2003), Can (2012).

Can (2008).

Donaldson (2001), Flynn et al. (2010), Huang et al. (2010), Lai et al. (2012), Flynn et al (2010), Simangunsong et al. (2011).
Ambrosini et al. (2009), Vijayasarathy (2010), Pertusa-Ortega et al. (2010)CHAPTER 4: METHODOLOGY

4.1 Introduction
The systematic and standardised approach that is followed in a research study to achieve the outlined targeted goals or objectives is termed a ‘research methodology’. Considered the substratum for conducting any research, research methodology supplies general principles that guide the researcher in a synchronised fashion (Dawson, 2002). A research study is considered a logical approach to systematically resolving a problem or issue, where the research methodology determines the logic and reasoning behind applying the methods necessary for conducting the research. Logically, it analyses why a particular method or technique has been selected by the researcher (Kothari, 2004). Research methodology plays an important role in establishing the outcome of the research, and comprises basic parameters, namely philosophy, approach, strategy and data collection methods. Identifying the correct principle in each parameter is essential for securing accurate and unbiased results.
Chapter 4 introduces the essentiality of the research methodology and accordingly highlights the research fundamentals connecting the theoretical paradigm of the study with its practical implications. The chapter brings forth the philosophical assumption that structures the study, the research strategy that outlines the nature of the relation between research logic, theory and practical materialisation of the study, and the methods used to collect and analyse the data for the study. Information on the planned procedures for conducting the study and extracting relevant results is provided in the subsequent sections of Chapter 4.
Section 4.2 details the research philosophy used in the study. Whilst various philosophies are outlined and discussed, this section justifies the research philosophy used in the present study using valid resources. In a comparable vein, Section 4.3 highlights the three types of research approach and justifies the use of a mixed-methodology approach. Section 4.4 focuses on the research strategy and validates the use of abductive strategy. Section 4.5 reflects different types of research design, and validates the use of multiple research designs to achieve the study’s objectives. Sections 4.6 and 4.7 significantly focus on the data collection methods, procedures and data analysis techniques for collecting primary data. Section 4.8 stresses on the ethical considerations followed in the current study, whilst Section 4.9 stresses on the role of theory in current research. Chapter 4 subsequently concludes in Section 4.10.
4.2 Research Philosophy
Research philosophy or paradigm is the underlying proposition focusing on the researcher’s assumption and the manner in which the world is viewed. Research philosophy can be comprehended as the process of acquiring knowledge to resolve certain prominent issues or problems to understand the social world in a better manner (Matthews & Ross, 2010). It establishes a relationship between the researcher’s description and explanation of reality, and the acquired and developed knowledge (Saunders, Lewis & Thornhill, 2009). The researcher’s perception of the world and reality produces a certain set of beliefs, which dictate the form and reality for completing the study. The philosophy underpins the research strategy and outlines the methods for conducting the research. Importantly, it is crucial to understand the research problem and choose the right philosophy (Bryman & Bell, 2011).
Research paradigms can be fundamentally categorised into three main groups: ontology, epistemology and methodology (Guba & Lincoln, 1994; Saunders et al., 2009). Ontology focuses on the nature of reality and defines the role of social actors. Social actors can either be objective or subjective. Thus, objectivism and subjectivism are two forces of ontology, highlighting the independency of social actors from the outlined social phenomenon (as in the case of objectivism) or the dependency of social actors from the outlined social phenomenon (as in the case of subjectivism). In subjectivism, social phenomenon is created as a result of the interaction of various social actors. Epistemology defines the relationship between the researcher and social phenomenon. Moreover, it defines the plausible nature of the study’s outcomes (Saunders et al., 2009). Methodology focuses on various methods the study needs in order to collect and analyse the data and accomplish the set goals and objectives (Creswell, 2009).
Each of these research paradigms are built on certain fundamental assumptions, which function as a basic set of beliefs of alternative inquiry to the study’s issues or problems. The four assumptions are positivism, post-positivism, critical theory and constructivism. The following table provides a brief outline of the four philosophical assumptions and their corresponding ontological, epistemological, and methodological paradigm (Guba & Lincoln, 1994).
Table 3: Four philosophical assumptions and corresponding ontological, epistemological, and methodological paradigm (Guba & Lincoln, 1994, p. 109)
Research ParadigmsPhilosophical AssumptionsPositivismPost-positivismCritical TheoryConstructivismOntologyNaïve realism- ‘real’ reality, but easily discernibleCritical realism- ‘real’ reality, but only imperfectly and selectively discernible Historical realism- ‘virtual’ reality shaped by values over time. Values include socio-political, cultural, economic and gender Relativism- constructed reality based on local and specific discerningEpistemologyDualist/objectivist; outlines true findingsModified dualist/objectivist; criticizes tradition/community; outlines probable findings Transactional/subjectivist; outlines value-mediated findings Transactional/subjectivist; outlines created- findingsMethodologyExperimental, verifies hypothesis, quantitative in most cases, easily manipulative through interventionsModified experimental, falsifies hypothesis, includes quantitative methods in most cases, manipulative and follows critical multiplismDialogic/dialectic Hermeneutical/ dialectical
A dominant assumption in the past 400 years (Guba & Lincoln, 1994) suggests that positivism elucidates naïve reality, where reality is observed and is easily discernible in its originality. The social phenomenon is studied in an objective manner, and further is recognised as independent of social actors. Studies based on positivism search for consistencies and casual relationships, and accordingly focus on the deductive principle to accomplish the study. Dependent on an existing theory, positivism develops and tests the hypothesis, and is experimental and manipulative in nature. The verification of hypotheses through quantitative methods—which test the facts acquired in the study—produces true findings and further develops the existing theory (Bryman & Bell, 2011; Saunders et al., 2009).
The principle underlying post-positivism lies in positivism itself; however, it adopts critical realism as its ontological assumption. In critical reality, ‘real’ reality is observed, albeit through imperfect and selective discerning (Guba & Lincoln, 1994). Whilst the existence of reality is seldom disputed in this philosophical assumption, it accepts the presence of differences in objects across different contexts. Post-positivism’s philosophical assumption follows a modified dualism or objectivism, and observes the study’s phenomenon with criticality and outlines probable findings (Guba & Lincoln, 1994; Bryman & Bell, 2011). Thus, post-positivism follows modified experimental methods for data collection, focuses on falsifying a study’s hypotheses, and, in most cases, includes quantitative methods. It is manipulative and follows critical multiplism (Guba & Lincoln, 1994), in which multiple stakeholders are involved to critique the research subject and questions in an effort to achieve the targeted results in an unbiased manner (Coward, 1990).
Critical theory goes against positivist assumption. Based on virtual reality, a reality defined by values such as socio-political, cultural, and economic and gender over time, effectuates the definition of social phenomenon and establishes a causal relationship between cause and effects within the boundaries of the social phenomenon in the critical theory philosophy. Human behaviour is explained based on theories (Bryman & Bell, 2011), and an interactive relationship is made apparent between the researcher and social actors/subjects of the phenomenon assessed. The research, as based on critical theory, is accomplished through dialogic/dialectic understanding, and the findings thus are value-mediated, given the presence of high subjectivity in this philosophical assumption (Guba & Lincoln, 1994).
Constructivism is based on local and specific discerning of the social phenomenon, and reality is ‘constructed’ in this assumption. Subjectivism principle is shared with critical theory and, unlike critical theory, the relationship between the researcher and social actors/subjects of the assessed social phenomenon is interactively linked to the created findings. The methods for conducting the study are hermeneutical/dialectical in nature, whilst constructivism is invariably linked with the local and specific values of the researcher and the subjects involved in the study (Guba & Lincoln, 1994).
Each philosophical assumption attempts to conduct the research study in accordance with the three major research paradigms: ontology, epistemology and methodology. The current study is conducted in lieu with post-positivistic assumptions. Post-positivist research has four crucial characteristics: a) research is broad; b) theory and practice cannot be separated and theory cannot be ignored merely in an effort to obtain facts (Ryan, 2006); c) motivation for conducting the research is central and crucial for the enterprise or sector or phenomenon under investigation; and d) the idea of conducting research with the use of only set or correct techniques and the subsequent categorising of information is inadequate (Schratz & Walker, 1995).
The main aim of the study is centred on addressing the phenomenon of demand uncertainty mitigation through management practices in order to ensure all customers who perform the Hajj at Mecca are satisfied, which in turn leads to leveraging SMEs’ performance in Saudi Arabia. The research study is broad since it interconnects the macro-economic and social endeavours of empirical research studies. Demand uncertainty mitigation, in and of itself, is a broad subject, the theory of which is understood to practically satisfy all Hajj pilgrims’ food supply through management practices. Mitigating demand uncertainty is invariably central to the government since uncertainty in demand could cause a crisis to the food industry during a peak period. In Saudi Arabia, the Hajji attracts different people from all over the world, which could be a cause of crisis if demand is not reliably predicted. Additionally, Saudi SME food suppliers need to understand the gross and subtle aspects of SCI in order to ensure their performance in the Hajj season is improved since they thrive under critical and complex competitive conditions. Given that the concept of uncertainty reflects both subjective and objective natures (Campos, Neves & de Souza, 2007), the idea of conducting research only with subjective or objective techniques is inadequate.
Considering these aspects, post-positivist research philosophy is apt for this study, where reality is critical of positivism (Eriksson & Kovalainen, 2008) and is established through social actors. Research findings are probable and are based on modified objectivism, with the research methodology focusing on elements in their natural settings. Research based on this philosophy is conducted to explain the research phenomenon, and the logic and purpose of the actions of the elements in the social setting (Guba & Lincoln, 1994). The following section focuses on the research approach in lieu with the post-positivistic research philosophy.

4.3 Research Approach
Whilst four distinct research philosophies are identified, research studies identify two common research approaches in an effort to clarify the research methodology or philosophy of methods: quantitative and qualitative (Eriksson & Kovalainen, 2008). A general distinction between the two approaches is provided in the below table.

Table 4: General distinction between quantitative and qualitative approaches (Saunders et al., 2009)
Point of ComparisonQuantitative approachQualitative approachOntology or Nature of RealityObjectiveSubjectiveResearch StrategyDeductiveInductiveResearch DesignExploratoryDescriptiveTypes of DataQuantitativeQualitativeData Collection MethodsExperimental and SurveyInterviews, case studies, ethnography, grounded theory, narrativeSample SizeLarge Sample SizeSmall sample size
The above table reflects the general distinction to be made between the two research approaches, where the following arguments can be made in this regard. Quantitative approach has its fundamentals embedded in positivistic philosophical assumption, and the qualitative research approach has its fundamentals in interpretivism or constructivism (Bryman, 2006). The research approach in positivistic philosophy is usually applied across large samples, where both quantitative and qualitative approaches are used. However, positivistic philosophy commonly deploys quantitative methods to collect the required data and achieve the objectives. The research approach in the interpretivism paradigm is usually applicable across small sample sizes, and in-depth investigations are conducted to collect qualitative data (Guba & Lincoln, 1994; Saunders et al., 2012).
From the structural arguments, it can be deduced that the nature of reality in the quantitative approach is objective, and a deductive research strategy is followed. The relationship between variables is measured in this approach, with the research design exploratory in nature. Numerical data is collected using experimental or survey methods across a large sample size using quantitative measures for data collection and analysis. Conversely, the nature of reality in a qualitative approach is subjective, and an inductive research strategy is followed. The context of the study or social phenomenon is discerned in this approach, and the research design is descriptive in nature. Non-numerical data is collected using interviews, case studies, ethnography, grounded theory and narrative, and suitable analysis is conducted across small sample size (Saunders et al., 2009).
Adding further, quantitative research approach is guided by a linear model that begins with a theoretical position, states hypothesis, and then progresses through the steps of the research design, measures the concepts, selects the site for carrying out research, selects the sample, and administers the instrument. Commonly, this instrument is used as a survey or questionnaire for the study (Easterby-Smith, 2003), with data usually analysed through the use of statistical techniques (parametric and non-parametric). The analysed data leads to findings and conclusions, and the researcher reverts to the theoretical position at the initial phase of the research and accordingly verifies its validity. This feedback loop, which subsequently leads the researcher back to the theoretical perspective, shows both deductive and inductive elements, and is ‘indicative of the positivistic foundations of quantitative research’ (Bryman & Bell, 2007, p. 157).
The characteristics of quantitative research can be outlined in an effort to understand the intricacies involved in this approach. Quantitative approach includes: a) a design that is determined before the project commences; b) the application of a single method or a combination of methods; c) a consistent approach, and d) the involvement of either a cross-sectional or longitudinal approach. Quantitative approach attempts to establish the ‘precise measurement of something… such methodologies answer questions related to how much, how often, how many, where and who. Although the survey is not the only methodology of the quantitative researcher, it is considered a dominant one’ (Cooper & Schinder, 2008, p. 164). To summarise, quantitative studies need and elicit quantitative information in mind of studying the particular phenomenon, and involves the usage of numbers and figures (Blumberg, Cooper & Schindler, 2005) to record the frequency of responses. They are highly objective and fail to gather the subtle nuances of subjectivism, which is apparent in qualitative design.
A qualitative design is at the opposite end of the continuum and is contradictory to the quantitative approach. It seldom focuses on numbers but rather employs a research approach that may evolve or adjust as the research progresses, and commonly uses multiple methods on a simultaneous basis. Qualitative design usually consists of longitudinal approaches, and generally is not concerned with consistency (Cooper & Schindler, 2008). The sample size in a qualitative design is usually small. A definition of a qualitative research, written in the last century, is still actively applied:
‘Qualitative research is multi-method in its focus, involving an interpretive, naturalistic approach to its subject matter. This means that qualitative researchers study things in their natural settings, attempting to make sense of, or interpret, phenomena in terms of the meanings people bring to them’ (Denzin & Lincoln, 1994, pp. 1–17).
Qualitative research design employs qualitative techniques or methods such as interviews, participant observation and diary/journal methods (Easterby-Smith, Thorpe & Lowe, 2002) and case studies for the completion of the research. In some cases, the qualitative research approach is synonymously viewed with the case study approach (Gall, Gall & Borg, 2009). According to Gall et al. (2009), a case study research ‘emphasises the fact that qualitative research focuses on the study of cases rather than of populations and samples’ (p. 31). Yin (1989) defines a case study as ‘an empirical inquiry that investigates a contemporary phenomenon within the real-life context, when the boundaries between phenomenon and context are not clearly evident, and in which multiple sources of evidence are used’ (p. 23). Thus, these qualitative techniques provide a range of different techniques that attempt to ‘describe, decode, translate and otherwise come to terms with the meaning, not the frequency, of certain more or less naturally occurring phenomena in the social world’ (Van Maanen, 1983, p. 9).
Whilst a qualitative approach attempts to gather data at a deeper level than the quantitative approach, the chances of a researcher’s personal involvement with the sample is relatively high. This can serve as a disadvantage to the study. A small sample size does not achieve research generalisability, which acts as an impediment to the application of research studies across different sectors of SME in the study context. Based on these arguments, it is clear that, rather than employing a single approach, combining both quantitative and qualitative approaches will ensure the initiation and expansion of certain research frameworks and constructs (Creswell & Plano Clark, 2011).
Additionally, the post-positivist research philosophy enables the researchers to investigate their own epistemology, take up a learning rather than a testing role, value problem-setting as opposed to problem-solving, focus on the specific findings rather than overall truth, and use critical multiplism and quantitative and/or qualitative methods to achieve the study’s objectives (Ryan, 2006). Considering these assertions, the current study follows a mixed-methodology to address the phenomenon of demand uncertainty mitigation through management practices in order to satisfy all customers who perform the Hajj at Mecca, which, in turn, leads to leveraging SMEs’ performance in Saudi Arabia. In this vein, mixed-methodology is essential for this study for the following reasons.
Mixed-methodology has emerged as a third methodological paradigm (Creswell & Plano Clark, 2007), and involves both quantitative and qualitative approaches for the collection, analysis, interpretation and presentation of the data (Teddlie & Tashakkori, 2009). A common approach applied in business and management studies (Saunders et al., 2012), mixed-methodology departs from the central assumption of conducting a study either on the basis of qualitative or quantitative constructs only. The central focus in mixed-methodology is the combination of various methods and techniques (Wheeldon & Ahlberg, 2012) to extract the benefits of both quantitative and qualitative approaches (Creswell, 2009).
The research process in mixed-methodology is guided by several key principles: ‘using fixed and/or emergent design; identifying a design approach to use; matching a design to the study’s problem, purpose, and questions; and being explicit about the reason for mixing methods’ (Creswell & Plano Clark, 2011, p. 54). Each of these key principles is elaborated in the following paragraphs: in fixed-mixed-methods, the use of qualitative or quantitative methods and techniques is predetermined and planned at the beginning of the research study. Contrariwise, in emergent-mixed-methods, the use of qualitative or quantitative methods and techniques emerge when conducting the research process (Creswell & Plano Clark, 2011).
There are multiple approaches to the design of a mixed-methods approach, where two categorical constituencies can be identified in this regard: a typology-based approach and a dynamic approach. In the case of the typology-based approach, mixed methods are classified, and a particular design is adapted in lieu of the study’s purpose and questions/objectives. In the dynamic approach, multiple components of the design are interrelated, aside from the research questions. Whilst research questions constitute as a central focus of adopting a design, other components, namely conceptual framework, methods and validity, are interconnected in this type of approach (Creswell & Plano Clark, 2011).
In a typology-based approach, mixed methodology is classified using various logical foundations (Creswell, Fetters & Ivankova, 2004; Tashakkori & Teddlie, 2003; Greene, 2007; Teddlie & Tashakkori, 2009; Morse & Neihaus, 2009; Leech & Onwuegbuzie, 2007). However, given the simplicity of Leech & Onwuegbuzie’s (2007) typological model, the current study adopts the mixed-methods process model of Leech & Onwuegbuzie (2007). Through this typology, three dimensions are present: the first layer identifies whether the study consists of fully or partially mixed methods design; the second layer identifies the timing of data collection in terms of whether it follows a concurrent or sequential pattern; and the third layer identifies whether each approach should be given equal or dominant status, considering their intensity of the application and use in the research process. Thus, choosing the relevant mixed-research approach invariably connects the design chosen with the research problem, purpose and questions.
Whilst mixed methods have received wide attention in recent years, given their dynamicity (Wheeldon & Ahlberg, 2012), explicitness regarding the purpose or logic behind mixing methods is essential in order to effectively use the time and resources. The reasons for choosing a mixed-methodology approach fall under one of these categories: triangulation, complementarity, development, initiation and expansion. Whilst triangulation corresponds to and corroborates the results using different methods, complementarity elaborates, enhances, illustrates and differentiates the results from one method with another. Development uses the results from one approach to develop the other approach, and commonly is employed to develop measurement instruments and accordingly implement sampling and implementing techniques. Initiation discovers new paradoxes and frameworks, and expansion aims at extending the horizon of the study’s inquiry with the use of different methods for different components of inquiry (Greene, Caracelli & Graham, 1989). Additionally, mixed- methodologies are used to obtain greater validity (triangulation), offset, completeness, process, different research questions, explanation, unexpected results, instrument development, sampling, credibility and context (Bryman, 2006).
With this as the theoretical foundation for the mixed-methodological research approach, the current study follows a fixed-methods approach and across-stage mixed method model, where qualitative and quantitative methods are predetermined and planned prior to conducting the research (Creswell & Plano Clark, 2011), with mixed methods applied across different research stages (data collection and analysis) (Johnson & Onwuegbuzie, 2004). A typology-based approach and partially-mixed sequential dominant status design is applied in this study. In partially-mixed concurrent design, the quantitative and qualitative phases are independent of one another, and are mixed only in the interpretation stage; however, they are conducted one after the other. The dominant status to the mixed-methods approach either can provide high importance to the quantitative or qualitative approach; the current study assigns dominant status to the quantitative approach, considering its objectivity and ease of data analysis and use (Wheeldon & Ahlberg, 2012; Saunders et al., 2009). In the current study, qualitative methods are employed first, followed by quantitative methods. A sequence exists in the methodological approach. Additionally, a quantitative approach is given dominant status, considering the post-positivist philosophy applied in the study. In a post-positivist philosophy, the study depends on a quantitative approach, and the addition of a qualitative approach benefits most researches. Thus, the current study is QUAN+qual in nature (Johnson, Onwuegbuzie & Turner, 2007). The research design is aligned with the study’s purpose and questions, with relevant research designs drafted, as explained in Section 4.5.
The reasons for mixing methods in the current study are expansion and completeness. The current study aims at validating a theoretical model based upon the extended resource-based view (ERBV) of the firm, and contingency theory, which establishes the effect of Supply Chain Integration (SCI), postponement (PP), Mass Customisation Capability (MCC), on Mitigating Demand Uncertainty (DUM) under high Competitive Intensity (CI). The study attempts to expand and further add to the existing literature on demand uncertainty under high competitive intensity, using the case of the Saudi SME food suppliers and Hajj season. A novel research in this context, the study adopts a mixed methodology in an effort to achieve completeness, where comprehensives in the area of inquiry are obtained using both qualitative and quantitative measures (Bryman, 2006).

4.4 Research Strategy
‘Research strategy’ can be defined as logics of enquiry, based upon which the research questions are answered (Blaikie, 2007). Whilst research philosophy and approach guide the researcher in conducting the research, they cannot be considered separate entities; rather, they belong to the multidimensional set of a continuum. However, debates persist regarding the appropriate identification and application of research philosophy and approach, which is aptly resolved in choosing the appropriate research strategy (Saunders et al., 2012). Whilst the positivists elucidate inductive or hypthetico-deductive process to explain patterns of behaviour, interpretivists emphasise establishing motivations that lead to the patterns of behaviour (Baker & Foy, 2008). Most social researches follow either inductive or deductive logic, from theories to patterns of observations, and vice versa (Trochim, 2006) to meet their objectives. Thus, choosing appropriate research strategy is crucial to navigate across the research study in a smooth and non-confusing manner.
On common grounds, two key reasonings can be identified as strategies guiding the research philosophy and approach: ‘inductive and deductive reasoning’ (Blaikie, 2007; Onwuegbuzie & Leech, 2005). In the inductive method (qualitative), inductive logic is applied. The researcher ‘begins with observations, seeks patterns in those observations, and generates tentative conclusions from those patterns’ (Rubin & Babbie, 2009, p. 39). In the deductive method (quantitative), deductive logic is applied. The researcher ‘begins with a theory, then derives hypotheses, and ultimately collects observations to test the hypotheses’ (Rubin & Babbie, 2009, p. 40). The following figure identifies the process involved in each of these approaches.

Figure 17: Overview of inductive and deductive reasoning (Wheeldon & Ahlberg, 2012)

In the inductive strategy, ideas available in the literature or otherwise through discourse from individuals are used to create new theory or concepts. Developed from the observation of facts on specific constructs or cases, inductive logic is more focused on exploratory research design and divulges in collecting in-depth data in an effort to explain the social phenomenon in question. Inductive reasoning reflects the attributes of the qualitative approach, and themes are identified from the collected data, which is used to build or refine generalised conclusions (Collis & Hussey, 2009; Blaikie, 2007; Bryman & Bell, 2007; Remenyi, Williams, Money & Swartz, 1998).
In the deductive strategy, theoretical constructs available in the literature are tested using a relevant hypothesis framework. Specificity plays an important role in deductive strategy, where a top–down approach is used to test the hypothesis with relevant data. Data collection, in this strategy, is quantitative in nature, and the study is deeply rooted in existing theories. More specific conclusions are effectively extracted from empirical study (Collis & Hussey, 2009; Blaikie, 2007; Trochim, 2006). Whilst these two strategies commonly guide the research philosophies, another important strategy or reasoning known to have emerged with mixed-methodology is abductive strategy.
Abductive reasoning is another crucial research strategy utilising the benefits of both qualitative and quantitative approaches to test or validate the research objectives. Based on the expertise or intuition of the researcher, abductive strategy is commonly applied in mixed-methodology; when this is done, abductive reasoning provides the research study with new extensiveness to comprehensively conduct the study in a robust manner (Wheeldon & Ahlberg, 2012). The following figure provides an overview of abductive reasoning.

Figure 18: Overview of abductive reasoning (Wheeldon & Ahlberg, 2012)

Given the application of mixed-methodology, the current study uses abductive reasoning. The mixed method research design and abductive strategy divides the current study into two stages: (1) the exploratory stage, where both qualitative interviews and quantitative surveys are conducted to explore aspects pertinent to the processes of strategy formulation, implementation and evaluation; and (2) the hypotheses-testing stage, where quantitative surveys will be conducted to examine research hypotheses on the relationship between Supply Chain Integration (SCI), postponement (PP), and Mass Customisation Capability (MCC), and Mitigating Demand Uncertainty (DUM) under high Competitive Intensity (CI). Thus, data for this study is both primary and secondary in nature, and will be collected using both qualitative and quantitative methods.

4.5 Research Designs
Research designs are ‘procedures for collecting, analysing, interpreting, and reporting data in research studies’ (Creswell & Plano Clark, 2011, p. 53). As the name suggests, research designs outline the structure for investigating the research study, acting as building blocks for organising the research study and report (Easterby-Smith, Thorpe & Jackson, 2008). They function as models for conducting the research and supply the following: a) the procedures and directions for research methods; and b) logic for interpreting the results (Creswell & Plano Clark, 2011). Research designs ultimately validate the logic behind using, a) particular method/s of data collection, b) source of information, and c) sample within the set time-constrained limits (Easterby-Smith et al., 2008). Thus, choosing the relevant research design enables the researcher to materialise the blueprint into a valid study. Research designs can be classified based on various constructs: methods of data collection, time, researcher participation and purpose of study (Blumberg et al., 2008). Considering that the commonly used classification is based on ‘purpose of the study’, the current study uses this construct of classification. Accordingly, there are three types of research design based on the purpose of the study: exploratory, descriptive and explanatory (Chisnall, 2001). The following table provides a general comparison of the three different research designs in literature.

Table 5: General comparison of the three different research designs
Research DesignsDescriptionApproach and StrategyData collection methodsSourceExploratory ResearchFocuses on discovering new insights and ideas and is usually conducted when study phenomenon has received less or no attention or information on the study phenomenon is very little or non-existing. Lack of previous literature on any particular study phenomenon demands exploratory researchUsually qualitative with inductive approachIn-depth literature review, interviews with subject matter experts and brainstorming using focus group interviews (Sekaran & Bougie, 2009; Hair, Babin, Money & Samouel, 2003; Saunders et al., 2009).Descriptive ResearchFocuses on elaborating on any particular phenomenon and is usually conducted on the basis of previous literature. This type of research design is applied when information about the characteristics of the study phenomenon is required to either ascertain specific facts or add further information to the existing theoryA specific set of scientific methods are used. Usually quantitative with deductive approach. Statistical techniques are typically used to conduct analysisMainly cross-sectional studies are conducted and raw data is obtained through surveys (Collis & Hussey, 2009;
Blaikie, 2007; Hair et al.,
2003; Malhotra & Varun, 1998).
Causal ResearchFocuses on identifying the cause-effect relationship between the study’s variables. The primary focus of the study is to test whether one cause has any effect on the otherUsually quantitative and statistical techniques are typically used to conduct analysis and summarize the dataExperiments (Wilson, 2010; Hair et al., 2003).From the general comparison of the three different research designs, it can be deduced that exploratory design generally is used when a study’s issue or phenomenon lacks prior knowledge and requires further insight to the idea under investigation. Descriptive design is commonly applied when prior knowledge on the study’s issue or phenomenon exists, and where the main focus of the study is centred on describing the study phenomenon in an in-depth manner. Casual or explanatory design focuses on establishing and accordingly explaining the casual relationship between the study variables. Given the subtle boundaries between these three study designs, a research study either can base itself on one of these designs or otherwise combine two or more designs in an effort to satisfy its purpose.
The current study aims at addressing the phenomenon of demand uncertainty mitigation through supply chain management practices in order to satisfy all customers who perform the Hajj at Mecca, which in turn leads SMEs’ performance in Saudi Arabia to be leveraged. Resultantly, the research design is divided into two phases; each phase is crucial to resolving the research problem and validating the relationship of the impact of Supply Chain Integration (SCI), postponement (PP), and Mass Customisation Capability (MCC) on mitigating Demand Uncertainty (DUM) under high Competitive Intensity (CI). Exploratory design is used in the first phase by completing a literature reviewing and conducting interviews to clarify concepts regarding aligning sources of uncertainty with supply chain strategies in order to improve supply chain performance. Descriptive-explanatory design is applied in the second phase, which obtains in-depth information on the impact of Supply Chain Integration (SCI) on manufacturing strategies, such as postponement practice (PP) to mitigate Demand Uncertainty (DUM) via cross-sectional sample survey. A total of 17 hypotheses are tested in an effort to understand how demand uncertainty can be mitigated in high season of customer demand:
H1: Customer integration positively affects postponement.
H2: Customer integration positively affects mass customisation capability.
H3: Internal Integration positively affects postponement.
H4: Internal integration positively affects mass customisation capability.
H5: Supplier Integration positively affects postponement.
H6: Supplier integration positively affects mass customisation capability.
H7: Postponement has a direct positive relationship with mass customisation capability.
H8: Postponement positively mitigates demand uncertainty.
H9: Mass customisation capability positively mitigates demand uncertainty.
H10: Internal integration has a positive indirect effect on postponement practice through customer integration.
H11: Internal integration has a positive indirect effect on postponement practice through supplier integration.
H12: Internal integration has a positive indirect effect on mass customisation capability through customer integration.
H13: Internal integration has a positive indirect effect mass customisation capability through supplier integration.
H14: Competitive Intensity enhances the indirect effect of internal integration on postponement practice through supplier integration.
H15: Competitive Intensity enhances the indirect effect of internal integration on postponement practice through customer integration.
H16: Competitive Intensity enhances the indirect effect of internal integration on mass customisation capability through supplier integration.
H17: Competitive Intensity enhances the indirect effect of internal integration mass customisation capability through customer integration.

The process flow of the current study’s research design is indicated in the following figure.
Figure 19: The process flow of the study’s research design

4.6 Data Collection
The question of the research methods appears following the definition of the research paradigm, its approach, strategy and design. The process of collecting evidence or information on a particular subject or phenomenon in order to validate the study’s assertion or accordingly generate new ideas or theoretical constructs is defined as data collection (Olsen, 2011). Thus, data collection methods—otherwise considered research methods—are techniques adopted by the researcher to generate, extract and analyse the data (Blaikie, 2007; Hussey & Hussey, 1997). A crucial stage of research, data collection, involves techniques centred on collecting primary and secondary data. The data is ‘collected afresh and for the first time, and thus happen to be original in character’ (Kothari, 2005, p. 95), and is termed primary data.
Primary data can be collected using numerous interactive methods. It can be collected either by quantitative method/s, qualitative method/s, or as a mixture of both quantitative and qualitative method/s. A few important primary data collection methods are as follows:
‘(i) observation method (ii) interview method (iii) through questionnaires (iv) through schedules, and (v) other methods which include (a) warranty cards; (b) distributor audits; (c) pantry audits; (d) consumer panels; (e) using mechanical devices; (f) through projective techniques; (g) depth interviews, and (h) content analysis’ (Kothari, 2004, p. 96).
The data collected from various sources, such as journals, books, articles and credible websites, is considered secondary data. In the current study, primary data is collected using qualitative and quantitative techniques. The qualitative and quantitative primary research methods used in this study are discussed in sections 4.6.1 and 4.6.2, and secondary data used in the study are detailed in the ‘Reference’ section of the study.

4.6.1 Qualitative Data Collection
Qualitative data can be collected using three primary methods: interviews, observations and written documents. Interviews provide data and knowledge from the respective participants in regard to their opinions and views concerning the study phenomenon. Direct quotations of the participants serve as primary data. In the observation method, the researcher adorns the role of observer, with the various aspects of the study phenomenon considered. Primary data in this method is obtained in the form of programme records, publications, reports and so on (Patton, 1990). Interview is another crucial primary data collection method, during the course of which the researcher personally communicates with the participants via telephone or face-to-face, or impersonally through telephone, email or any computer-mediated communication. Interviews may be structured, semi-structured or unstructured. The three interview types are differentiated on the basis of the structure of interview questions: if, for example, the questionnaire follows a structured pattern and the interviews generate quantifiable data, they are considered structured. There is limited interaction between the researcher and participants in the case of structured interviews, which are standardised and produce standard data. In semi-structured interviews, there is a unique mixture of structured and unstructured interview methods. Semi-structured interviews are flexible and commonly used by researchers (Dawson, 2002). Necessary probing questions will be used using the questionnaire tool, with new dimensions on the topic collected through this method (Saunders et al., 2009). Unstructured interviews lack any structure or questionnaire to collect data, with focus on obtaining in-depth information and opinions of the participants of the research study. These are usually carried out when the researcher aims at investigating matters or issues that have not undergone in-depth research. This approach is time-consuming and can deviate away from the phenomenon under examination in the study (Saunders et al., 2009).
Although structured interviews are standardised and commonly used in the quantitative approach, semi-structured and unstructured interviews are non-standardised and commonly used in qualitative research approach. The researcher uses a tool (questionnaire) with a set of questions to navigate across the interview process; however, the interviewer is at the privilege to modify and channel the interview if new information is obtained during the process (Saunders et al., 2012).
In the current study, the interviews function as a qualitative approach and serve a dual purpose: whilst they are used to acquire primary data for the study and supply direct quotations, they also are used as an effective means to validating the research model and items used in the questionnaire. They are used to achieve face and content validity for any research study. The semi-structured interviews serve as means for supplying qualitative primary data for this study.

4.6.1.1 Elaboration on Semi-structured Interviews and the Justification of their Application:
Interviews can be defined as purposed conversations (Burgess, 1984), and are commonly built around the rationale of garnering in-depth information on a particular subject in order to collect and analyse data, and thereby develop meaningful constructs (Mason, 2002). In semi-structured interviews, general interview protocols provide a framework for conducting the interviews (Recker, 2011). A questionnaire tool with checklist of issues or questions guides the researcher (Bryman & Bell, 2007). This data collection method has numerous benefits: for example, semi-structured interviews encourage two-way communication and are less intrusive (Recker, 2011), and also provide sufficient flexibility for the researcher to respond to participants’ information and accordingly steer the interviews in line with the pre-determined questionnaire (Bryman, 2004). Whilst semi-structured interviews can be used to confirm existing logic, the flexibility provided to the researcher during the interviews provides an opportunity for learning and producing new ideas. This non-structured nature of semi-structured interviews allows the researcher to discuss sensitive issues in a guided manner (Recker, 2011). The order of the questions in any semi-structured interview is pre-determined; however, given its flexible nature, the interviewer is permitted to clarify any ambiguity in the questionnaire and use multiple probes if required in an effort to encourage the participants from providing more information (Berg, 2009).
In semi-structured interviews, two types of question are present: structured and open-ended. Whilst structured questions collect factual data, open-ended questions collect actual or supplement data for the study (Stone & Collin, 1984). Open-ended questions allow the interviewer to supply unique responses and experiences in an in-depth manner (Wilkinson, Joffe & Yardley, 2004). However, the questionnaire prevents the interviewer from making any unnecessary deviations. Thus, conversations in this type of questionnaire are neither free nor highly structured. Considering the numerous benefits of semi-structured interviews, the current study considers this form of qualitative interviewing process most apt for this study.
The following section covers the requirements of the semi-structured interviews and the manner in which the requirements are fulfilled in the current study.

4.6.1.2 Semi-structured Interview Requirements:
The effectiveness of semi-structured interviews is based on various factors, including the number of participants, the length of the interviews, the questionnaire tool and the role of the interviewer in facilitating the interviews.

Number of Participants:
In the current study, 12 CEOs belonging to various SMEs across Saudi Arabia constitute as the participants of the semi-structured interviews. In order to obtain a reasonable number, depending on the sample pool size, the researcher set out to collect the participants for semi-structured interviews with the idea of 12, through the adoption of a snowball sampling technique. The size of the sample pool in qualitative interviews depends on time and resource availability. A total of 12 interviewees is reasonable for collecting qualitative data since this sample size provides the researcher with the ‘experience of planning and structuring interviews, conducting and partially transcribing these, and generating quotes for their papers’ (Adler & Adler, 2012, p. 10).
Snowball sampling can be defined as the selection of research subjects through referrals (Morgan, 2008). In the case of snowball sampling technique, study participants are recruited on the basis of the subjects’ social network. The participants may refer to other possible participants; thus, depending on their willingness to participate and the relevance of the study (Burns & Burns, 2008), interview participants are recruited. With the idea of identifying a total of 12 participants, the researcher used social connections to get approval from four CEOs. The researcher informed the four CEOs (with whom the researcher maintains a good relationship) of the need and importance of the study, where the social network generated the remaining eight participants for the study. Thus, the snowball technique is used aptly in this study to generate the sample pool for qualitative interviews. Informed consent is obtained through letter, and face-to-face interviews are conducted. Interviews are recorded through audio-recording devices, with the offices of the 12 participants located in Makkah and Jeddah City constituting as locations for conducting the interviews.

Length of the Interviews:
Selecting the appropriate length of interview is crucial when seeking to maintain the flow of interviews and accordingly prevent the participants and interviewer from losing focus of the research topic. From a general perspective, the length of one interview is between 45 minutes and 60 minutes (Eriksson & Kovalainen, 2008). The length of interviews in the current study is 60 minutes. Recording devices are used to record the interviews, aside from written notes after obtaining required consent from each participant.
Interview Questions:
The interview questions for the semi-structured interviews used in the current study are divided into two sections, namely Section A and Section B, serving a dual purpose by providing primary data and validating the content of the questionnaire. In order to build good rapport and make the participants feel at ease, each participant was asked to give information regarding their respective company’s background. The structured questions in the current study are the first four questions of Section A, which are concerned with collecting factual data.
The following questions were posed regarding the subjects’ company, constituting as Section A of the study:
[1] What kind of business does your company carry out?
[2] How many employees are there in your company?
[3] How many suppliers does your company have at this moment?
[4] In which countries are your suppliers based?
[5] In general, what do you think about Supply Chain Integration? Do you value the idea of having a close and good business relationship with your suppliers and customers?
[6] Do you have issues/problems with demand uncertainty mitigation during the Hajj season in regard to food provision? How you can mitigate it?
[7] What strategies are currently used by your company to mitigate demand uncertainty during the Hajj season regarding food provision?
[8] In general, what are the alternative activities and mechanisms that may be employed to deal with demand uncertainty mitigation during Hajj?
Section B focuses on the conceptual framework of the study and further attempts to validate the study’s conceptual framework. It aims at establishing face validity and content validity. The model is shown to the participants, who are asked to comment on the below model with certain specific open-ended questions.

Figure 20:

[1] Do you agree with the linkages between the four constructs? If not, please can you comment further?
[2] What do you think about the impact of Supply Chain Integration on postponement practices? Please comment.
[3] What do you think about the impact of between Supply Chain Integration on Mass Customisation Capability? Please comment.
[4] What do you think about the impact of between Supply Chain Integration on demand uncertainty? Please comment.
[5] What do you think about the impact of postponement practices on mass customisation capabilities? Can you give an example?
[6] What do you think about the impact of postponement on demand uncertainty? Please comment.
[7] What do you think about the impact of mass customisation capabilities on demand uncertainty? Please comment.
[8] What would you think if Supply Chain Integration or customisation capabilities or postponement was to be taken out of this model? How can firms mitigate demand uncertainty during the Hajj season when providing food to pilgrims?

The Role of the Interviewer:
The researcher performed the role of interviewer in the current study. Since semi-structured interviews involve CEOs of various firms, maintaining a high level of professionalism whilst extracting relevant information was a crucial aspect. However, the questionnaire tool added sufficient leverage to break the ice between the participants, and the interviewer’s enthusiasm and non-biased attitude ensured the high and equal participation of all CEOs (Bryman, 2008).

Data Preparation:
Data preparation is essential in qualitative studies. Data preparation paves the way for data analysis, where data preparation in the current study refers to the process of translation. The audio-recorded interviews are translated from Arabic to textual data. A direct translation method is used to translate the textual data from Arabic to English (Bernard & Ryan, 2010). Audio-recorded interviews are subjected to verbatim and word-to-word transcription, where the researcher played the role of transcriptionist in the current study. It is argued that oral-to-written translations reconstruct the data rather than producing a direct copy (Kvale, 1996; Fog, 2004). Considering this, the current study follows certain guidelines in an effort to ensure authenticity and accordingly maintain the representativeness of the transcribed data. All interviews are transcribed two days following the interview in order to indicate the subjective nuances projected or collected from interviews, aside from remembering and maintaining the identities of participants. Additionally, word-to-word written translation is believed to improve the general representativeness of the audio/oral recordings. Nevertheless, care was taken not to reveal the identities of the participants in the translated data. With this noted, the participants are identified as partici1, partici2, partici3 and so on in the transcribed data. The translated data is saved in Word documents, along with each participant’s code of identification.

4.6.1.4 Data Analysis:
Data analysis in a qualitative study is synonymous with content analysis, and three major approaches can be outlined in this regard: conventional content analysis, directed content analysis and summative content analysis. Conventional content analysis is based on observation, where codes are defined during data analysis. The source of codes in this analysis is the data itself. In the case of directed content analysis, the study begins with the defining of theory and codes before and during the data analysis. The source of codes in this analysis is the study’s theory or research findings. In summative content analysis, the study begins with relevant keywords, which are identified before and during the data analysis phase, whilst keywords are extracted from the literature review stage (Hsieh & Shannon, 2005).
Qualitative data analysis in the current study is accomplished using the direct content analysis method. Given that the study is based on theoretical concepts and conceptual framework, codes are based on Supply Chain Integration (SCI), postponement (PP), Mass Customisation Capability (MCC), mitigating Demand Uncertainty (DUM) and high Competitive Intensity (CI). Any text failing to fall into these categories will be assigned a new code in the analysis. The coding process for the current study is discussed in the following section.

Coding Process:
The process of coding the collected data is essential in order to ensure simple analysis. The process labels the collected data (Saunders et al., 2012), where three types of code generally are used in qualitative data: structural, themes and memos. Whilst structural codes focus on the features of the interview, thematic codes are centred on the collected text, whilst memos provide notes about the codes themselves (Bernard & Ryan, 2010). In the current study, a coding process is followed, based on thematic guidance for the literature on coding.
Thematic codes are developed based on the conceptual model; in this case, data is classified based on the study’s variables of Supply Chain Integration (SCI), postponement (PP), Mass Customisation Capability (MCC), mitigating Demand Uncertainty (DUM) and high Competitive Intensity (CI). The interview transcripts are read, and textual content is segregated from the mass data using discretion and understanding. The resulting data or units of data is in the form of sentences or paragraphs, which are coded based on the study’s variables. The available data then is ready for analysis (Bryman & Bell, 2011; Saunders et al., 2012).
4.6.1.5 Limitations of Semi-structured Interviews and Directed Content Analysis:
Limitations in the data collection can be attributed to two aspects: limitations in conducting semi-structured interviews and directed content analysis. Semi-structured interviews exhibit common limitations of any form of interview: wrongful representation of participants, lack of common and convenient time, meeting and location for all participants and A lack of planning and good organisation of group meetings (Malhotra & Birks, 2003). The lack of active participation from all members and the control of discussion by one single participant or few participants is another limitation of interviews (Eriksson & Kovalainen, 2008). In directed content analysis, the researcher is inclined to obtain supportive evidence rather than non-supportive evidence of the study’s theories, with overemphasis on a particular theory potentially limiting the research study’s focus to one particular phenomenon (Hsieh & Shannon, 2005).

4.7 Quantitative Data Collection
Quantitative data collection involves two methods of data collection, namely experiments and survey. Whilst experiments are commonly used in explanatory studies, surveys are applied in exploratory and descriptive studies (Saunders et al., 2012). Experiments aim at examining the cause and effect between two variables in a controlled setting, and are limited to the number of variables in the study. Commonly conducted in laboratory settings, experiments do not effectuate the generalisation of a study’s results (Blumberg et al., 2008; Saunders et al., 2012). Given the exploratory and descriptive nature of the current study, surveys with a questionnaire tool are used to collect primary quantitative data. The survey is a ‘system for collecting information from or about people to describe, compare, or explain their knowledge, attitudes, and behaviour’ (Fink, 2003). A close-ended questionnaire survey is designed for this study.

4.7.1.1 Questionnaire Development:
Questionnaires, for the current study, are developed from the completion of a relevant literature review and validated using focus groups. Additionally, a pilot study and purification measures validate the equivalence of the questionnaire tool. A questionnaire for the current study is developed using a dual process. In the first step, the specification and operationalisation of constructs is achieved. Following, scale and type of questionnaire is decided, with the questionnaire then developed. Each of these aspects is discussed below.

The Specification and Operationalisation of Constructs:
The specification of constructs refers to the definition of the study’s constructs. The current study comprises six constructs, namely customer integration, internal integration, supplier integration (SCI), postponement practices, mass customisation capability, and demand uncertainty mitigation. Each of these constructs is elaborated on in the literature review.
The operationalisation of constructs refers to the translation of concepts to measurable indicators (Saunders et al., 2012). Measurable indicators are obtained from a relevant literature review, and accordingly validated using focus groups. Measurable items of each construct and their corresponding codes are outlined below.

Customer Integration:
Supplier/customer integration refers to the extent to which an organisation can partner with suppliers and customers in an effort to structure its inter-organisational practices, behaviours, processes and strategies into collaborative, manageable and synchronised processes so as to meet customer requirements (Lai et al., 2012). The items for this construct are extracted from the studies conducted by Zhao, Huo, Flynn & Yeung (2008), Flynn, Huo & Zhao (2010), Swink, Narasimhan & Wang (2007), Narasimhan & Kim (2002), Swink & Nair (2007) and Cousins & Menguc (2006). The measurement items for assessing customer integration are mentioned below.
• CI1: We are in frequent, close contact with our customers.
• CI2: Our customers are actively involved in our product design process.
• CI3: The customers involve us in their quality improvement efforts.
• CI4: We work as a partner with our customers.
Internal Integration:
Internal integration can be understood as the degree to which the various internal functions and processes of an organisation strategically coordinate and collaborate with one another’s activities and decisions, and thus form integral relationships across the different functions (Lai et al., 2012). The items for this construct are extracted from the studies carried out by Huang, Kristal & Schroeder (2008), Zhao, Huo, Selen & Yeung (2011), Flynn, Huo & Zhao (2010), Braunscheidel & Suresh (2009), Droge, Jayaram & Vickery (2004), Koufteros, Vonderembse & Jayaram (2005) and Swink, Narasimhan & Wang (2007). The measurement items for assessing internal integration are as follows:
• II1: The functions in our plant are well integrated.
• II2: Our plant’s functions coordinate their activities.
• II3: Our top management emphasizes the importance of good inter-functional relationships.
• II4: Management works together well on all important decisions.

Supplier Integration:
Supplier integration mainly entails supplier partnerships, supplier information-sharing and the involvement of suppliers in product development (Lau et al., 2012). The items for this construct are extracted from the studies carried out by Zhao, Huo, Flynn & Yeung (2008), Flynn, Huo & Zhao (2010), Swink, Narasimhan & Wang (2007) and Cousins & Menguc (2006). The measurement items for assessing supplier integration are mentioned below:
• SI1: We maintain cooperative relationships with food suppliers.
• SI2: We maintain close communications with food suppliers in regard to quality considerations and design changes.
• SI3: Our firm key food suppliers provide input into our product development projects.
• SI4: We strive to establish long-term relationships with food suppliers.

Postponement Practice:
Postponement can be defined as the process of delaying product finalisation in the supply chain until orders from customers are received with the aim of customising products, as opposed to performing those activities with the expectation of achieving future orders (Can, 2008). The items for this construct are extracted from the studies completed by Can (2008), Yang & Burns (2003), Hoek (1999) and Cholette (2009). The measurement items for assessing internal integration are as follows:
• PP1: Our firm postpones final product assembly activities until customer orders are received.
• PP2: Our firm postpones final product-labelling activities until customer orders are received.
• PP3: Our firm postpones final packaging activities until customer orders are received.
• PP4: Our firm postpones the forward movement of goods.

Mass Customisation Capability:
Mass customisation is a process where manufacturers tailor-make products to satisfy individual customer needs at the same prices as those of mass-produced items (Davis, 1987). The items for this construct are extracted from the studies completed by Kristal, Huang & Schroeder (2010), Feitzinger & Lee (1997), Rungtusanatham & Salvador (2008), Huang, Kristal & Schroeder (2008), Liu, Shah & Schroeder (2006), Tu, Vonderembse & Ragu-Nathan (2001), Kristal, Huang & Schroeder (2010), Duray, Ward, Milligan & Berry (2000), Ismail, Reid, Mooney, Poolton & Arokiam (2007), Kotha (1995) and Da Silveira, Borenstein & Fogliatto (2001). The measurement items for assessing mass customisation capability are:
• MCC1: We can are highly capable of large-scale product customisation.
• MCC2: We can easily add significant food product variety without increasing costs.
• MCC3: We can easily add product variety without sacrificing quality.
• MCC4: We can customise food products whilst maintaining high volume.
• MCC5: Our capability for responding quickly to customisation requirements is very high.

Demand Uncertainty Mitigation:
Demand uncertainty can be defined as variations and fluctuations in demand (Chen & Paulraj 2008; Lai et al., 2012), and the process of mitigating demand uncertainty can be comprehended as demand uncertainty mitigation. The items for this construct are extracted from the study conducted by Zahra & George (2002). The measurement items for identifying demand uncertainty mitigation measures are:
• DUM1: Demand uncertainty is mitigated by providing our customers with products consistent with their nominated product specification.
• DUM2: Demand uncertainty is mitigated when our customers place orders consistent with their nominated delivery lead time.
• DUM3: Demand uncertainty is mitigated when our customers provide reliable forecasts as to their demands.
• DUM4: Our customers place orders consistent with their nominated delivery lead time.
• DUM5: We can provide products to our customer consistent with their nominated product specification.
• DUM6: Our customers provide us reliable forecasts on their demands.

Competitive Intensity:
Competitive intensity refers to the extent to which an organisation faces competition in the market in which it operates (Lai et al., 2012). The items for this construct are extracted from the study completed by Zahra & George (2002) and Jaworski & Kohli (1993). The measurement items for assessing competitive intensity are:
• CPI1: We are in a highly competitive industry.
• CPI2: Our competitive pressures are extremely high.
• CPI3: We do not pay much attention to our competitors.
• CPI4: Competitive moves in our market are slow and deliberate, with long time gaps between different companies’ reactions.

Scale and type of Questionnaire:
Questionnaire scale refers to the rating indicator assigned to the measurement items in any given questionnaire. Whilst scales such as Thurstone scale, Likert scale, Semantic differential scale and Guttman scale (Chisnall, 2001) are applied widely and used in operational management studies, Likert scale is used in the current study. Likert scale generally consists of points spanning 4–7 (Saunders et al., 2012). Whilst a four-point scale allows the participants to express their attitudes and beliefs either from a positive or negative perspective, a five-point scale, on the other hand, allows participants to express neutral feelings towards the measurement items (Malhotra & Birks, 2003). Considering this, a seven-point scale ranging from ‘strongly disagree’ to ‘strongly agree’ is used in the current quantitative survey questionnaire.
The type of questionnaire for any given study is dependent on the mode of communication between the researcher and participants (Churchill, 1995). There are three types of questionnaires: self-administrated, personal and telephone interviews (Blumberg et al., 2008); these can be used to collect quantitative data through the survey method. The self-administrated questionnaire refers to data collection through an electronic medium (email or web-based) or hand delivered (delivery and collection questionnaire) or mail questionnaire (Blumberg et al., 2008; Churchill, 1995; Saunders et al., 2012). Telephone questionnaires refer to data collected via telephone calls (Churchill, 1995). Personal interview questionnaires refer to data collected via face-to-face conversation (Churchill, 1995). By comparing the types of questionnaire, it was found that telephone interviews are costly and limited in length (Blumberg et al., 2008). Whilst personal interviews provide a high response rate, they are costly, and both personal and telephone interviews are subjected to interviewer bias (Churchill, 1995; Saunders et al., 2012). Considering the low cost and minimal research involvement in self-administrated questionnaires (Blumberg et al., 2008), an electronic medium is used, with data collected via email in the current study. The questionnaire is sent to the respective firms after identifying the relevant sample through the sampling process.

4.7.1.2 Sampling:
The sampling process can be understood as the process of extracting a representative sample from the study population (Johnson & Christensen, 2010; Gravetter & Forzano, 2012). The sampling process is a systematic and sequential five-fold process facilitating the research methodology. It includes: defining the study population; identifying the sampling frame; selecting the sampling technique to identify the sample elements; determining the sample size; and collecting the data from the elements (Burns & Burns, 2008).
Population Definition and Sampling Frame:
Population—otherwise considered as Universe—is the total group of people, objects or events from which the researcher obtains information. A sample, on the other hand, represents a larger population in the research study. It consists of a set of individuals from the population that guarantees the data for the study (Gravetter & Forzano, 2012). When the population is large, measuring each and every individual or object or event becomes a costly, tiresome and cumbersome process. In such situations, a sample is drawn. The representative sample is a relatively small set of the study population, and is a characteristic representation of the original population on all demeanours (Johnson & Christensen, 2010). Accordingly, generalisability is achieved. Accurate inferences based on the subset of individuals’ opinions, attitudes and behaviours can be drawn, without any biased view of the researcher. Statements on the population can be made based on the sample data (McCormack & Hill, 1997; Johnson & Christensen, 2010). Employees belonging to SMEs across Saudi Arabia, which supply food to Hajj pilgrims, constitute the population for the current study.
Sampling frame is a physical representation/repository of the study population, in its entirety, from which the sample is drawn (Sekaran & Bougie, 2009). Whilst it is difficult to identify all SMEs supplying food to Hajj pilgrims, given the lack of data consolidation and comprehensiveness of the Saudi food industry, SMEs across Saudi Arabia constitute the sampling frame of the study. A lack of appropriate sampling frame necessitates thoughtful process in selecting the relevant sampling technique.

Sampling Technique:
The process of selecting a sample is achieved using various techniques termed as sampling techniques. Sampling techniques can be categorized into two groups: probability and non-probability sampling. In probability sampling, each element in the population is given equal chance of being selected and the contrary is true in the case of non-probability sampling (Churchill, 1995). Four types of probability sampling methods are available in research methodology literature. They are: simple random sampling, systematic sampling, stratified random sampling, and cluster sampling. Three types of non-probability sampling can be outlined, based on available literature: convenience sampling, quota sampling and snowball sampling (Bryman and Bell, 2011).
Given the non-availability of the sampling frame, the current study applies convenience sampling to obtain relevant sample size for quantitative data collection (Malhotra, Agarwal and Peterson, 1996; Saunders et al., 2012). Considered as the most frequently used non-probability sampling technique, convenience sampling is easy and less time and energy consuming. However, its inability to accommodate all possible participants from the population is its potential drawback (Churchill, 1995; Saunders et al., 2012). Nevertheless, its ability to select sample from the available population renders convenience sampling as the apt technique for the current study.

Sample Size:
Sample size refers to the number of participants in the quantitative research study. Dependent on various factors such as size of target population, sampling error, confidence level and statistical techniques, choosing the relevant sample size is important to validate and test the research hypotheses (Saunders et al., 2012). The current study is based on SEM to validate the relationship between the study’s constructs. The sample size based on SEM is dependent on five crucial factors. The study’s assumptions are tested using multivariate analysis. Estimation technique is identified to estimate the sample size. Model complexity is analysed and data is screened for missing data and outliners. Average error of variance is estimated (Hair et al., 2010). A sample size of 200 is recommended for complicate model (Kline, 2005). However, sample size of 300 is also recommended (Hair et al., 2010). Given the complexity of the current study, the number of sample size was 245 and after excluding the 6 outliers, a sample size of 239 is obtained for data collection.

4.7.2 Language, Translation and Cultural Considerations
Oxford English dictionary defines language as, ‘method of human communication, either spoken or written, consisting of the use of words in a structured and conventional way’ (Oxford, 2014). Language is the manner through which people observe and describe the world around. It creates ideas in the minds which invariably ensure communication and interaction with the world around. Thus, language effectuates the origin and distribution of judgments which is critical in conducting any research study (Usunier, 1998). Since the study is conducted in Saudi Arabia and gathers the opinion and views of Saudi Arabians, the demand for Arabic language as primary language for the questionnaire is obvious. Given that the native language of Saudi Arabians is Arabic, the questionnaire is translated from English to Arabic.
Having established the need for translation, an overview of the different approaches for translation will enable the study to choose and justify the relevant translation approach. Direct translation, back-translation, parallel translation and mixed technique are four common approaches for translation. In direct translation, the questionnaire is translated to the target language directly from its source (Usunier, 1998). In back-translation, the translated form is translated back to the source language. The translated form is compared with the source language, and is assessed and corrected appropriately (Harkness, 2003). In parallel translation, two or more versions of the source language translations are compared to create a final version, and in mixed technique back translation is conducted by two or more translators and the resulting versions are compared to create a final version (Usunier, 1998).
Direct translation is undoubtedly the easiest approach; however, slight modifications may exist due to syntax differences between the source and translated language. On similar lines, back translation can showcase slight difference between the source and translated language; however, it will highlight lexical equivalency rather than content equivalency (Usunier, 1998). Additionally, back translation requires efficient translator skills and expert to assess the extent of likeness between the two versions (Harkness, 2003). Parallel translation can assure equanimity in questionnaire working; however, it is lengthy, costly and time consuming. Similarly, while mixed technique guarantees accuracy, it is costly, since it demands two or more translators (Usunier, 1998).
Given its simplicity, ease of application and inexpensiveness, direct translation approach is the most commonly used translating approach. While each approach has its benefits, the current study employs direct translation as the translating approach, given its simplicity and cost-effectiveness (Green and White, 1976). Additionally, decentring procedure is used in the current study while conducting direct translation. In decentring, the goal is to obtain a final version, which focuses on meaning or content rather than the syntax of the language. The resulting version is equivalent to the source language (Prieto, 1992). Considering the cultural differences present between English and Arabic languages, any ambiguity in the questionnaire is explained and simplicity is maintained across the questionnaire. Additionally, direct translation assesses the equivalence of the questionnaire and ensures the reliability of the study’s questionnaire (Punnett & Shenkar, 2004).

4.7.3 Pilot Study and Purification of Measurement Items
Pilot study also referred as feasible studies, can be defined as small versions of the actual study. An important construct of research methodology, pilot study is crucial to the researcher and fulfils range of functions (van Teijlingen & Hundley, 2002). The purpose of pilot study revolves around quantitative data collection instrument. Pilot studies assesses: a) the clarity of questionnaire instructions, b) the presence of biased questions c) aptness of questionnaire layout and, d) length of time required to complete the questionnaire. Pilot studies also assess the reliability and appropriateness of the translated questionnaire (Punnett and Shenkar, 2004) and eventually assesses the equivalence of the study’s questionnaire.
Considering the importance of pilot study, a pilot study is carried in this study on a sample of 50 subcontractors, Hajj campaigns, pilgrimages’ institutions and food suppliers. With the sample size of 50, the required guideline for pilot study sample size is met; given that, the minimum number in the sample size for any pilot study is 10, and between 100 and 200 for large sample-sized surveys (Saunders et al., 2012). The pilot study respondents appreciated the non-ambiguous nature of the survey questionnaire. However, most participants expressed confusion regarding demand uncertainty mitigation questions, given the role of demand uncertainty in the questions. Thus, the study questions under ‘demand uncertainty mitigation’ construct are respectively modified from: our customers’ place orders consistent with their nominated delivery lead time, we can provide products to our customer consistent with their nominated product specification and our customers provide us reliable forecasts on their demands, to we mitigate demand uncertainty when our customers place orders consistent with their nominated delivery lead time, we mitigate demand uncertainty by providing products to our customer consistent with their nominated product specification, and we mitigate demand uncertainty when our customers provide us reliable forecasts on their demands.
Apart from pilot study, purification of measurement items is essential to ensure the operability and clarity of the questionnaire (Bryman and Bell, 2011). Purification of measurement items can be defined as the process of assessing the face validity, content validity and reliability of the measurement items in the study (Saunders et al., 2012). Given that, the measurement tool in the current study is questionnaire, face and content validity and reliability is tested on the survey questionnaire, which is used to collect quantitative data. Prior to the pilot study, face and content validity of the study’s questionnaire is established: a) using a group of academic experts and b) via the opinions of 12 CEOs through semi-structured interviews. Face validity defines the appropriateness of the instrument’s measurement values and confirms whether the scale used in the questionnaire measures the relevant qualities which it is intended to measure. Face validity decides the actual completion of the questionnaire by the intended participants, and clarifies whether the used measurements items are relevant for the questionnaire or not (Todd and Bradley, 2013). Content validity can be defined as comprehensiveness of the measurement tool in the study, and it can be assessed using a panel of experts. The panel of experts assess the appropriateness of study’s questions to its corresponding constructs (Mitchell, 1996; Saunders et al., 2012). In the current study, the group of academic experts, validated the survey questionnaire’s questions and 12 CEOs validated the relativity of the study’s constructs, the structure and wording of the questionnaire; post which, the pilot study is conducted.
While the validity of the questionnaire is established before the pilot study, reliability of the measurement tool is established after the pilot study (Churchill, 1979). Reliability refers to the consistency of the measurement tool and repeatability of the measures (Trochim, 2006). Reliability in the current study is measured using Cronbach’s alpha test, inter-item correlations, and item-to-total correlation. The consistency of the whole questionnaire scale is assessed using Cronbach’s alpha test and correlation amongst items are measured using inter-item correlations. Item-to-total correlation assesses the correlation of questionnaire’s items to the overall summated score (Hair et al., 2010). A combination of the three tests is a great procedure for establishing reliability. Establishing correlation amongst items is crucial for partial least square analysis, which is not dependent on a single measure. Thus, the three tests are used in this study to purify the items of the questionnaire. Additionally, in Cronbach’s alpha test, the resulting value is dependent on the number of items in the questionnaire scale, which can produce misleading results (Field, 2005). Resultantly, correlations are applied to the results of the pilot study. If the total correlation value is >0.3, then the items in the questionnaire scale are considered reliable as per the inter-item correlation and item to total correlation tests (Field, 2005), and if the Cronbach’s alpha test value exceeds 0.7 (Hair et al., 2010; Kline, 2005), or 0.5 or 0.6 in some cases, items in the questionnaire scale are considered reliable (Churchill, 1979; Nunnally, 1978). Post purification, some items from each scale is removed to increase reliability. The corresponding questions under each study construct accompanied with their respective codes are provided in the following table.

Customer Integration
CI1: We are in frequent, close contact with our customers.
CI2: Our customers are actively involved in our product design process.
CI3: The customers involve us in their quality improvement efforts.
Supplier Integration
SI1: We maintain cooperative relationships with food suppliers.
SI2: We maintain close communications with food suppliers about quality considerations and design changes.
SI3: Our firm key food suppliers provide input into our product development projects.
Internal Integration
II1: The functions in our plant are well integrated.
II2: Our plant’s functions coordinate their activities.
II3: Our top management emphasizes the importance of good inter-functional relationships.
Postponement Practice
PP1: Our firm postpones final product assembly activities until receives customer orders
PP2: Our firm postpones final product labelling activities until receives customer orders.
PP3: Our firm postpones final packaging activities until receives customer orders.

Mass Customisation Capability
MCC1: We can are highly capable of large-scale product customisation.
MCC2: We can easily add significant food product variety without increasing costs.
MCC3: We can easily add product variety without sacrificing quality.
Demand Uncertainty Mitigation
DUM1: We mitigate demand uncertainty by providing products to our customer consistent with their nominated product specification.
DUM2: We mitigate demand uncertainty when our customers place orders consistent with their nominated delivery lead time.
DUM3: We mitigate demand uncertainty when our customers provide us reliable forecasts on their demands.
Competitive Intensity
CPI1: We are in a highly competitive industry.
CPI2: Our competitive pressures are extremely high.
CPI3: We do not pay much attention to our competitors.
Table 6: Corresponding questions under each study construct accompanied with their respective codes

4.7.4 Data Analysis
Quantitative data analysis in comparison with qualitative data is highly comprehensive and complex. Given the array of methods and procedures, and software for conducting various tests, quantitative data analysis in the current study follows a two-step process. In the first step, data is cleaned and prepared for analysis and in the second step, actual analysis is conducted. In the current study, data cleaning involves checking for any blank data and outliers using SPSS software. Descriptive statistics measures are used by numerically describing the variables, followed by mean and standard deviations (Saunders et al., 2012). The actual analysis conducted in the second step involves partial least square (PLS) analysis. The analysis strategy, the measurement model and structural model of the current study are covered in the following sections.

4.7.4.1 Analysis Strategy:
Partial least square is one of the important structural equations modelling (SEM) technique. It is considered as second generation modelling technique and performs dual function. The function as measurement model and assesses the quality of the research constructs. It also functions as structural model and assesses the relationship between the outlined constructs (Fornell & Brookstein, 1982). Applied in numerous information systems (Gefen & Straub, 2005), management (Cording, Christmann & King, 2008), marketing (McFarland, Bloodgood & Payan, 2008) and operational management studies (Braunscheidel & Suresh, 2009; Cheung, Myers & Mentzer, 2010). SEM is an integral representation of large number of statistical models that use empirical data to validate the substantiality of theoretical constructs. An extension of general liner modelling (GLM), SEM is used to study the relationships between latent constructs indicated by multiple measures. Applied on experiential and non-experiential data and used in cross-sectional and longitudinal studies, SEM is a confirmatory, hypothesis-testing approach that stipulates casual relations for multiple variables in multivariate analysis. It involves two models, measurement model and structural model, whose evaluation determines whether the theoretical model is consistent with the collected data. While the measurement model is evaluated through Confirmatory factor analysis (CFA), structural model models the structural relationships between observed variables (vs latent variables) are modelled (Lei and Wu, 2007). The evaluation of each of these models in the current study is discussed in the following sections.

Measurement Model:
CFA is conducted using PLS software to assess the reliability and validity of the multiple-item scale. Reliability, convergent validity and discriminant validity tests are conducted in lieu with the 1981 guidelines of Fornell & Larcker. Item reliability and composite reliability tests, which are superior to Cronbach’s Alpha, are conducted in quantitative analysis since these tests consider actual factor loadings rather than assigning assumed equal weight for each item. Average Variance Extracted (AVE) is used to assess convergent validity, and discriminant validity is assessed in the following process. The square roots of the AVE of each construct are compared with correlations between the focal construct and each construct. Discriminant validity thus is established when a square root is higher than the correlation with other constructs (Fornell & Larcker, 1981).
Structural Model:
A hierarchical procedure is applied to test the hypotheses after assessing the validity of the measurement model. Specific relationships amongst constructs are indicated. The analysis thus shifts itself from CFA to SEM to test the hypotheses. The analysis stage moves from the mere specification of a relationship between the latent constructs and measured variables to advanced level, where the nature and strength of relationships between constructs are established and determined (Hair et al., 2010).

4.7.4.2 Elaborating PLS and Justifying the Use of PLS in the Current Study:
The PLS technique is slowly integrating itself into several research contexts and studies. PLS is also considered a variance-based SEM method, where its complementary approach, covariance-based SEM (CBSEM), as the name suggests, is based on covariance-based SEM method (Henseler, Ringle & Sinkovics, 2009). Numerous operational management studies have adopted covariance-based SEM (CBSEM) methods. However, the advantages of PLS or PLS software SMARTPLS may be understood by understanding the functionalities of its popular counter-part CBSEM (Peng & Lai, 2012), and software, such as LISREL and AMOS, which are the popular statistical software packages for CBSEM. Comparison functionalities are applied in the current study to justify the use of PLS.
Developed by Wold (1975), PLS is statistical technique based on dual processes. In the first process, the latent variable scores are computed using PLS algorithm, which is followed by the second process of applying OLS regressions on the computed latent variable scores to estimate structural equations. The first process is accomplished in two steps: outside approximation and inside approximation. In the case of outside approximation, the latent scores for the variables are calculated based on the weight of indicators using simple or multiple regressions. In inside approximation, the obtained latent scores are combined with neighbouring latent variables to obtain a proxy estimate. A conventional stopping technique stops the variables from converging, and the minimisation principle is applied to the residual variance with respect to the subset of estimated parameters. The process thus is partial in the least squares sense (Chin, 1998).
The PLS approach has its own advantages: for example, it relies on the data and is exploratory in nature (Chin & Newsted, 1999), meaning the focus of PLS approach lies in summarising and making predictions rather than explaining the covariance of measurement items (Chin, 2010). This attitude matches the study’s predetermined objectives. Furthermore, PLS does not mandate normal data and functions with small sample size (Chin & Newsted, 1999).
In the CBSEM approach, the covariance matrix is developed based on a specific set of structural equations, with focus on estimation. The central focus of estimation lies in minimising the difference between the two matrices: theoretical covariance and estimated covariance (Rigdon, 1998). CBSEM requires a specific set of assumptions without which the analysis cannot be accomplished. The multivariate normality of data and minimum sample size are two examples of such assumptions (Diamantopoulos & Siguaw, 2000). The research objective and outcome is confirmatory in nature rather than prediction-based in the CBSEM approach, which is the central objective of PLS. The PLS–SEM outlines the robust estimates of the structural model (Wold, 1982; Reinartz, Haenlein & Henseler, 2009).
Additionally, PLS is different from LISREL-type SEM since it is purely dependent on the predictive power of independent variables (Chin, 1998). This can be capitalised in such a way so as to explain complex relationships, and subsequently may be used to build theory. A component-based approach (Lohmoller, 1988), PLS problems associated with inadmissible solutions and factor indeterminacy can be avoided using PLS (Fornell & Brookstein, 1982). PLS also functions better than LISREL and AMOS since it does not require the assumption of normal data distribution (Gefen & Straub, 2005; Chin, 1998). Under the conditions of non-normality, PLS is capable of executing its functionalities. Whilst the LISREL-type SEM provides goodness-of-fit indices, PLS estimates path loadings and R2 values; they do not provide goodness-of-fit indices. Whilst path loadings identify the strength of the relationship between independent and dependent variables, R2 measures the predictive power of the variables. R2 values measure the degree of variance present in the independent variables (Gefen and Straub, 2005). Whilst the above comparisons provide a general overview between SEM-based method (PLS) and CBSEM, and the PLS method and LISREL-based SEM, the actual justification of using PLS can be obtained by drawing a comparison with AMOS. Given the use of CFA in the quantitative analysis, a comparison between SmartPLS and AMOS software for CFA highlights the benefits of using SmartPLS software.
AMOS is one of the popular statistical software for CBSEM (Hair, Ringle & Sarstedt, 2011). This is commonly used when the research objective is to test a theory, confirm a theory or compare alternate theories. If the formative measures in the measurement model are limited to specified rules and require additional specifications, such as co-variation, then AMOS based on CBSEM is used. If the structural model is non-recursive, then AMOS is used. If requirements with regard to model specification, non-convergence, data distribution assumptions and identification are met as per CBSEM, then AMOS software is used. Additionally, if the study requires global goodness-of-fit criteria and a test for measurement model invariance, then AMOS is the best software for conducting analysis (Hair et al., 2011).
In contrast, when the research objective is to predict the key driver amongst constructs, and if the structural model has many constructs and numerous indicators, then SmartPLS is used. If the data obtained is based on the sound approximation of the distributional assumptions and is considered non-normal, then SmartPLS is apt. Additionally, if latent variable scores are required to conduct subsequent analysis, then SmartPLS is the best software for conducting analysis (Hair et al., 2011).
Considering the several considerations associated with the data and research model, the PLS technique is used in this study. SmartPLS 2.0 M3 software is used for analysis to identify the measurement and structural model. Bootstrapping estimation procedure is used to identify the significance of scale factor loadings and path coefficients of measurement model and structural model, respectively (Gefen & Straub, 2005).

4.8 Ethical Considerations
Research ethics are ‘norms for conduct that distinguish between acceptable and unacceptable behaviour’ (Resnik, 2011). The ethical code of conduct of any research reflects the behavioural character of the personnel involved in the study, the researcher and research participants (Sekaran, 2003). Whilst ethics project the in-depth and perceptive values of each individual’s life, research ethics or code of ethics, as articulated by the American Educational Research Association (AERA), perform a multi-dimensional function. The code provides principles and guidelines for researchers to cover professional situations, educates the researchers and other personnel involved in the research about the benefits of conducting research in an ethical manner, and inculcates and percolates ethical behaviour from the academic perspective into the professional and personal life of the people involved in the study (AERA, 2011). Thus, numerous values, such as honesty, confidentiality, privacy and integrity, are crucial values that define the research ethics (Shamoo & Resnik, 2003).
Considering this, ‘honesty’ in the research study is elucidated across the various phases of the study; definition of research aims and objectives, primary and secondary data collection, data analysis and presentation and implicating conclusions. The current study’s aims and objectives are not plagiarised from any academic paper. A sense of freshness is present from the inceptive stages of the study. Secondary data, used for the literature review, is extracted from credible references—detailed information gained from the reference section. Primary data is qualitatively collected from the CEOs of SMEs, and quantitatively from 239 participants belonging to various SMEs across Saudi Arabia. Objectivity is maintained when defining the research methodology and conducting the analyses. Given the post-positivist philosophy and mixed-methodological nature of the study, the room for bias, unlike the qualitative approach, is minimal or altogether absent. The literature review provides a basis for extracting hypotheses, which are tested with valid primary data collection via a surveys and questionnaire tool. Content validity and reliability is established prior to and following the pilot study. The questionnaire, as a tool, enables unbiased data collection, where data analysis is conducted using appropriate descriptive statistics measures, details of which are presented in the findings and appendix section.
In order to maintain the semblance of research ethics, this study confines itself to the boundaries of confidentiality, privacy and integrity. The confidentiality, privacy and integrity of the research participants are maintained in this study. Informed consent is obtained via email or personally from participants before conducting the study, where the names or any personal information relating to the participants is not disclosed at any cost. The survey data will be maintained for a period of one year, after which the data will be deleted in an effort to ensure the confidentiality, privacy and integrity of the research participants are ensured. In this manner, the research study not only adheres to the ethical considerations from the report-writing perspective, but also from the research participants’ perspective. Issues pertaining to copyright and plagiarism are tested prior to submission, with the research content free from any contamination.

4.9 Role of Theory
Research studies are founded on the basis of a self-generated idea/s or on the basis of idea/s proclaimed by other researchers or philosophers, depending on the research area and context. Whilst research studies based on the ideas developed by others can be synonymously viewed as a theory-based research, research studies on self-generated idea/s are also based on certain theories developed by the individual mind. Thus, the role of theory, in any research study, is phenomenal, and research studies are conducted either to develop new knowledge and add to the existing literature or to test or verify an existing theory. A systematic process which involves data collection, research studies are directly or indirectly built on theoretical constructs (Blaikie, 2007). An overview of the ERBV of the firm and contingency theory is provided in this section to highlight the role of theories in the current study. ERBV is built on the resource-based view (RBV) of the firm. Additionally, contingency theory stresses the importance of possessing practices and strategies adhering to the business environment (Lawrence & Lorsch, 1986).
The current study aims at validating a theoretical model based upon the extended resource-based view (ERBV) of the firm and contingency theory that establishes the effect of Supply Chain Integration (SCI), postponement (PP), Mass Customisation Capability (MCC), on Demand Uncertainty (DUM) under high Competitive Intensity (CI). The study attempts to expand the study and further add to the existing literature on Demand Uncertainty under high Competitive Intensity using the case of Saudi SME food suppliers during the Hajj season. The study also aims at testing 17 hypotheses based on the study’s variables, and follows theory-building and theory-testing phases (De Vaus, 2007).
Whilst an overall relationship between constructs is established in the literature review, the relationship between supplier integration, internal integration and customer integration, and postponement, mass customisation capability and demand uncertainty mitigation, and the relationship between postponement, and mass customisation capability and competitive intensity requires more focus in an effort to understand the theoretical concept of the current study. The following sections focus on these relationships.
4.10 Conclusion
This chapter summarises the research methodology applied in the current study. The current study addresses the phenomenon of demand uncertainty mitigation through management practices in an effort to satisfy all customers performing the Hajj at Mecca, which, in turn, leads to leveraging SMEs’ performance in the context of Saudi Arabia. A theoretical model is used to establish the effects of Supply Chain Integration (SCI), postponement (PP), Mass Customisation Capability (MCC), on Demand Uncertainty (DUM) under high Competitive Intensity (CI).
The research follows the post-positivism philosophy since reality in this philosophy is critical in positivism research, and is established through social actors. Research findings are probable, and are based on modified objectivism and research methodology. Based on this philosophy, the research study usually is conducted to explain the research phenomenon, and the overriding logic and purpose of the actions of the elements in the social setting. Therefore, this philosophy stresses the importance of qualitative and quantitative methods.
Accordingly, the current study follows a mixed-methodology approach. Various theoretical aspects of the mixed-methodology are assessed, whilst a fixed-methods approach and across-stage mixed method model is applied in this study. A typology-based approach is used and partially-mixed sequential dominant status design is applied in this study. The current study is QUAN+qual in nature, and given the application of mixed-methodology, the current study uses abductive reasoning. Abductive reasoning utilises both qualitative and quantitative approaches based on inductive and deductive reasoning in an effort to test and accordingly validate the research objectives in line with the information available at the time. In this study, data comprises both primary and secondary, and will be collected using both qualitative and quantitative methods.
Resultantly, the research design is divided into two phases: an exploratory design is used in the first phase by reviewing the literature and conducting interviews to clarify concepts regarding aligning sources of uncertainty with supply chain strategies in order to improve supply chain performance; a descriptive-explanatory design is applied in the second phase, obtaining in-depth information on the impact of Supply Chain Integration (SCI) on manufacturing strategies, such as postponement practice (PP) to mitigate Demand Uncertainty (DUM) through a cross-sectional sample survey. A total of 17 hypotheses are tested in an effort to understand how demand uncertainty can be mitigated in a high season of customer demand.
Direct translation, accompanied with the decentring process, is used in this study, and a pilot study is carried out across a sample of 50 subcontractors, Hajj campaigns, pilgrimage institutions and food suppliers in an effort test the validity and reliability of the study’s questionnaire.
Primary and secondary data are collected in this study. Focus groups serve as a means for supplying qualitative primary data. A total of 12 CEOs belonging to various SMEs across Saudi Arabia constitute the participants of the interviews. A questionnaire tool is used in the current study, serving a dual purpose: it provides primary data and also validates the content of the questionnaire. It aims to establish content validity.
Surveys are used to collect quantitative primary data for the study. Employees belonging to SMEs across Saudi Arabia that supply food to Hajj pilgrims constitute the population for the current study. Given the non-availability of the sampling frame, the current study applies convenience sampling so as to obtain a relevant sample size for quantitative data collection. Secondary data used in the study are detailed in the ‘Reference’ section of the study. Qualitative data analysis in the current study is accomplished using the direct content analysis method. Quantitative data analysis in the current study follows a two-step process. In the first step, data is cleaned and prepared for analysis; in the second step, actual analysis is conducted. In the current study, data cleaning involves checking for any blank data and outliers. The actual analysis carried out in the second step involves Partial Least Square (PLS) analysis. Structural equation modelling is applied in an effort to validate the measurement and structural model. The analysed data is presented for logic connections, and validation between the study’s constructs and relevant conclusions can be extracted.

CHAPTER 5: ANALYSIS AND FINDINGS

5.1 Introduction
This chapter provides the findings garnered through the completion of semi-structured interviews, document review and online surveys. The data present is qualitative in nature, following the completion of interviews with 12 CEOs of various SMEs operating in Saudi’s food industry. The quantitative data, on the other hand, has been gathered following the completion of an online survey, carried out across a total of 239 respondents from the SMEs in the food section of the KSA. The primary data were gathered in order to provide support on the way in which relevant supply chain approaches could be applied in order to mitigate the demand uncertainty of the kingdom’s small and medium enterprises (SMEs) in the food sector throughout the period of Hajj. Accordingly, this chapter is broken down into seven sections.
The first section centres on the data analysis and data screening, which explains the way in which data are analysed and applied in order to test the structural links hypothesised between the different constructs. This needs to include an explanation of how missing values underwent examination and the way in which outliers were identified. The second section provides the subjects’ demographic information; in other words, details of their company location, education levels, occupation and work experiences, whilst the subsequent section provides the findings of the control variables, including firm age, firm production type and firm size. The fourth section provides insight into descriptive statistics, in addition to the approaches applied in testing the study constructs’ normality. In the following section, in consideration to the structural model evaluation processes and measures and the testing of the hypotheses through the application of the PLS-SEM, a discussion is provided. The sixth section provides a representation of the nomological validity, discriminant validity, multicollinearity, CV communality and CV redundancy measures, Goodness of Fit, Common Method Variance (CMV), Heterogeneity, path coefficients and predictive relevance. The following section provides the findings on the hypothesised link between the various constructs, in line with the path coefficients, and their significance level. Finally, a conclusion is provided.

5.2 Data Analysis
5.2.1 Data Analysis and Screening
The study of Hair et al. (2010) provides the suggestion that all researchers carry out a process of examining the data prior to completing analysis. Moreover, the study of Tabachnick & Fiddell (2007) considers the examination of data as being concerned with the approach of identifying and overcoming outliers and values, and accordingly testing the normality assumption of the data. Accordingly, this section considers the processes adopted in examining and screening the data, which includes missing data analysis, the various methods applied in order to identify outliers, and the techniques adopted in testing the normality of the data.

5.2.1.1 Missing Data Analysis:
As outlined in the work by Hair et al. (2010), there is the occurrence of missing values with the research subjects fail to provide answers to some of the questions in the survey. Missing data are further broken down into other categories, namely ignorable and non-ignorable missing data (Hair et al., 2010). Although ignorable missing data are not required to undergo changes, non-ignorable data, on the other hand, necessitate the making of adjustments by the researcher. In this instance, when there is data missing below 10%, the suggestion is made that it cannot be ignored (Hair et al., 2010). The present study adopts the SPSS software in an effort to clarify the occurrence of missing data and to identify outliers and the missing data for all variables, hence those that are ignorable. In the view of Hair et al. (2010), when missing data is seen to be between 10% and 15%, the researcher is required to delete various variables as the modification approach for decreasing biasness. Nonetheless, the findings of the SPSS analysis for the missing data did not reach such a percentage, meaning deletions do not need to be made.
Furthermore, upon analysing missing data, it is essential for the researcher to establisher whether or not the data is missing at random (MCA) or missing completely at random (MCAR) (Graham, 2009). In an effort to identify systematic errors or establish missing data patterns, the MCAR test, as devised by Little, was applied through the use of SPSS software, utilising as test variables the standard deviation, chi-square, and significance level and p values. The null analysis of the test showed that the missing data are missing completely at random, with the test results through the application of Little’s MCAR (Standard deviation = 8.626; Variance = 74.410; sg. 1; P > 0.05) emphasising that the present study comprises no systematic errors. Accordingly, the missing data can be treated by the researcher, with the mean used as a substitute for values that were missing in line with the suggestions made by Tabachnick & Fidell (2007), centring on the fact that substituting the mean is the most common and appropriate approach to imputing for missing values.

5.2.1.2 Detecting Outliers:
Outliers are recognised as observation points, which are distance from the remaining observations that stem from measurement variability or experimental errors (Hair et al., 2010, p. 64). Although it may be that outliers arise accidentally, it remains that their presence suggests either the existence of a measurement error or otherwise that the distribution of the study population is heavy-tailed (Cousineau & Chartier, 2010). In the instance that outliers arise following measurement error, it is suggested that researchers implement statistical tools centred on identifying outliers or otherwise achieve their complete removal (Cousineau & Chartier, 2010). In contrast, those outliers that arise following a heavy-tailed distribution of the study population provide some insight into the distribution with high kurtosis needing careful handling, particularly when adopting statistical tools that assume that there is normal data distribution (Cousineau & Chartier, 2010).
Owing to the fact that faulty values can be indicated by outliers, incorrect study approaches or invalid theories need to be removed, corrected or retained depending on their levels of magnitude (Leys et al., 2013). Nonetheless, a researcher cannot make decisions pertaining to how outliers can be managed and handled without identifying their presence (Leys et al., 2013). Moreover, failure to establish and correct outliers could mean the statistical testing approach could be distorted, meaning all data analysis, along with its results, are compromised (Larson-Hall, 2009). Accordingly, in an effort to improve reliability and validity in the present work, the researcher needs to test for both the univariate and multivariate outliers. In the view of Leys et al. (2013), there is the identification of univariate outliers through changing all data into standardised scores with the use of a standard deviation of either 2 or 2.5 around the mean, depending on the research situation and stance adopted by the researcher (Leys et al., 2013). In the present study, owing to the fact that the sample size is larger than 80 and would warrant a standard deviation exceeding 2.5, a standard deviation of 3 was used (Hair et al., 2010).
In contrast, multivariate outliers were identified with the use of using Mahalanobis D2 analysis, which was applied with the use of SPSS regression analysis. Cases were considered as multivariate outliers if the results of the D2 probability were found to lower than or otherwise equal to 0.001 (D2 ≤ 0.001). In the view of Tabachnick & Fidell (2007), data were gathered from a sample comprising a wide range of characteristics, meaning it is able to yield multivariate outliers. In the present work, it was expected that multivariate outliers would be identifiable owing to the fact that the data were gathered from subjects of varying firm age, firm production types and firm size.
As mentioned earlier, there are various approaches to managing outliers, including correcting, removing or retaining (Leys et al., 2013). Moreover, Larson-Hall (2009) makes the suggestion that outliers need to be retained for the duration for which they represent the population or otherwise or otherwise provided they do not show significant divergence from the normal distribution. The present research utilised the PLS-SEM data analysis approach, which is recognised for its data normality insensitivity. Accordingly, outliers have been retained in this instance owing to the statistical data analysis approach implemented, which means data are restricted from significant deviation from normal distribution.

5.2.1.3 Examining Data Normality
The majority of statistical approaches, such as correlation, regression, t-tests and variance analysis are applicable to Gaussian distribution theorem, where there is the normal distribution of data (Ghasemi & Zahediasl, 2012). In the context of the statistical data analysis process, the assumption of normality is fundamental, particularly when devising the reference intervals of variables. In the view of Ghasemi & Zahediasl (2012), data that have an invalid normality assumption are unable to deliver accurate, reliable and valid inferences pertaining to the research phenomenon; therefore, there is a need to ensure serious consideration in this regard. Although larger sample sizes, such as those exceeding 30, are not likely to violate distribution assumption normality, very small or large sample sizes of less than 30 could have significant distribution implications that could impact the overall validity and reliability of the work (Pallant, 2013). Moreover, the central limit theory postulates that the distribution of the sampling and their means should be normal when the sample data are almost normal when there is a sample size exceeding 30 (Ghasemi & Zahediasl, 2012).
Nonetheless, irrespective of the central limit and Gaussian theorems, the recommendation is made that a distribution normality test should be carried out in mind of establishing the seriousness of the data deviation from normality (Ghasemi & Zahediasl, 2012). The normality of data can be carried out either graphically or numerically through the use of SPSS software (Pallant, 2013). The present work applies skewness and kurtosis tests in an effort to establish the normality of the data’s distribution, as highlighted in the following table:

Table 7: Kurtosis and Skewness Tests-Normality

The distribution’s balance and disproportionateness is displayed through skewness, whereas the peakedness or flatness of the distribution, relative to the normal distribution, is measured by kurtosis (Hair et al., 2010). Upon evaluating the normality of distribution through the use of kurtosis, normally distributed data should show a zero level of kurtosis statistic. Upon a level of kurtosis static exceeding zero, i.e. a positive value, the data is described as having peaked, whereas a lower than zero kurtosis level, i.e. a negative value, is considered suggestive of a flatter distribution (Hair et al., 2010). Comparably, it is stated that there is the normal distribution of data when there is balance and symmetry in the skewness, with a statistical value of zero for skewness. Should a value greater than zero be seen for skewness, i.e. a positive value, then data then is seen to be unbalanced or distributed towards the left, whereas a negative value of lower than zero is seen when unbalanced distribution is present towards the right.
In the view of Hair et al. (2010), the critical value for establishing the skewness statistical and kurtosis values can be identified through the Z distribution, which is seen to show dependence towards the significance level of the study. Accordingly, the present research has adopted a cut-off value equating to ± 2.58, with a corresponding significance level of 0.01, as suggested through the work of Hair et al. (2010). Such statistics’ particulars can be seen in the descriptive statistics section of this chapter.

5.2.1.4 Basic Demographic Variables:
The online survey, in addition to the semi-structured interview approach, queried the subjects in providing basic data pertaining to both themselves and their business. Of the total 239 subjects to have completed the survey, 3.3% were seen to have acquired high school education, whilst almost two-thirds (65.7%) had been educated up to Bachelor’s degree, a quarter (25.55) had a Master’s degree, and the remaining 2.5% had a PhD, as detailed in the following table.
FrequencyPercentValid PercentCumulative PercentValidHigh School83.33.33.3Bachelor Degree15765.765.769.0Master Degree6828.528.597.5PhD62.52.5100.0Total239100.0100.0Table 8: Basic Demographic Variables

Moreover, the subjects were asked to emphasise their present job positions, with the findings recognising that the vast majority of the participants, notably equating to 91.2%, were top management whilst a small portion (7.1%) were from middle management, with the remainder (1.7%) in junior management. When required to explain their role within the business, a large majority (87.9%) were highlighted as CEO, 4.2% were Vice Presidents, and 7.9% were Operations Managers. The subjects were further asked to detail the duration for which they had been in the role, with the participants emphasising most (40.6%) had worked in their current business for 10–15 years, 18% for 15–20 years, 14.2% for 5–10 years and another 14.2% for 20–25 years. Importantly, only seven of the subject participants (2.9%) had worked for 1–5 years, and only one (0.4%) had been employed by their current firm for in excess of 30 years. The duration for which the participants have been employed by the firm was also used to establish the organisation’s age, which was recognised as a fundamental control variable.
In consideration to the firms’ locations, the majority (89.1%) stated that their offices were based in Makkah, with the remaining 10.9% stating they were based in Jeddah. The tables below display all of this data.
Your current job positionFrequencyPercentValid PercentCumulative PercentValidTop management21891.291.291.2Middle management177.17.198.3Junior management41.71.7100.0Total239100.0100.0
Which of the following best describes your role in your organisation?FrequencyPercentValid PercentCumulative PercentValidCEO21087.987.987.9Vice president104.24.292.1Operation manager197.97.9100.0Total239100.0100.0
How long have you been working with your current employer?FrequencyPercentValid PercentCumulative PercentValid1-572.92.92.95-103414.214.217.210-159740.640.657.715-204318.018.075.720-253414.214.290.025-30239.69.699.6Over 301.4.4100.0Total239100.0100.0
Where is your office located?FrequencyPercentValid PercentCumulative PercentValidMakkah21389.189.189.1Jeddah2610.910.9100.0Total239100.0100.0
Tables 9: Firm Location Basis

5.2.1.5 Control Variables:
In an effort to better establish the hypothesised links, three different control variables were applied for utilisation, which were firm age, firm production type and firm size. The size of the firm was established in line with the number of individuals employed by the entity. Moreover, the scores recorded by the organisations were examined in mind of establishing the number of employees of a firm. Moreover, the age of the firm was established through consideration to the year of registration, where the subjects were also asked to state when the business began its operations. Finally, firm production category was established in consideration to the type of food made by the establishment. Accordingly, there was the incorporation of three different categories, namely fresh meals, pre-cooked, and raw material wholesale.
The participants were asked, through the online survey, to detail the number of employees in their respective organisations. The table below provides the results of this.

Number of employees? FrequencyPercentValid PercentCumulative PercentValid30-50197.97.97.9210-250177.17.115.150-703715.515.530.52562.52.533.150-903414.214.247.3352.8.848.1382.8.849.090-110229.29.258.2401.4.458.6421.4.459.0441.4.459.4110-1302410.010.069.5531.4.469.95531.31.371.1130-1503815.915.987.0601.4.487.4150-1703012.612.6100.0Total239100.0100.0
Table 10: Online Survey Participants

The online surveys generated data, which subsequently was grouped and analysed, with the results detailed in the following figure. In line with these results, it was found that 3% of the subjects were employed by firms with 25 staff members; 11% were from businesses with 30–50 employees; 15% were from organisations with 50–70 employees; 16% were employed by firms with 90–110 employees; 16% also were found to be hired by organisations with 130–150 employees; whilst the remaining 13%, 10%, 9% and 7% came from businesses with 150–170 employees, 110–130 employees, 90–110 employees and 210–250 employees, respectively.
Figure 21: Online Survey Participants

The subjects were further questioned on the sectors in which their employing business operated. As can be seen displayed in Figure 2 below, approximately half (49%) of the online survey participants were employed by organisations operating in the food manufacturing domain, whereas 22% were from the food providing sector. A total of 14% were employed by those in the subcontracting sector, whilst 12% were from the Hajj campaign field, with the final 3% hired in the SC management arena.

Figure 22: Online Survey Participants

As discussed earlier, the information pertaining to the duration for which the subjects had been hired by an organisation was utilised in order to establish the age of the organisation. Accordingly, in line with the results garnered, the figure below provides insight into the results obtained.

Figure 23: Online Survey Participants

5.2.1.6 Descriptive Statistics:
In an effort to secure insight into the key elements that may be applied in order to eradicate DMU, the measurement framework highlights five important considerations, namely internal integration (II), supply integration (SI), customer integration (CI), postponement practice (PP) and Mass Customisation Capability (MCC). As discussed earlier on, the theoretical framework has been concerned with identifying the effects of Supply Chain Integration (SCI), postponement (PP), mass customisation capability (MCC) on DMU in a situation of high competitive intensity in an effort to authenticate the link of the effects of Supply Chain Integration (SCI), postponement (PP), and Mass Customisation Capability (MCC) (independent variables) on eradicating DUM (dependent variable) in a situation of high competitive intensity (CPI). For all of these aspects, a number of indicators were devised in mind of measuring their dimensions and links, as detailed in Appendix II. As can be seen displayed in Table VI, the hypothesised variables’ descriptive statistics were used to measure the ways in which internal integration, supply integration, Customer integration with their relationship with postponement practice and Mass Customisation Capability could be applied in an effort to mitigate demand uncertainty. These include maximum and minimum values, the mean, kurtosis, standard deviation and skewness.
The key aspects and their corresponding values were found to score above average, as shown through the descriptive statistics below. SI2 was found to be the lowest variable (we ensure close communications with food suppliers is maintained in regard to various quality considerations and design changes), which was found to have derived a score of a mean of 4.26 out of 7 (0.6086). CPI2 (competitive pressures are extremely high) was found to be the second least variable, scoring a mean of 4.29 out of 7 (0.6129). The third least was CPI3 (little attention is directed towards our competitors), the score of which was found to be 4.56 out of 7 (0.6514), with CPI1 (we operate in an industry that is highly competitive) closely following, whose score was 4.75 out of 7 (0.6786). It can be seen through these findings that all competitive intensity-related items were seen to be variables with lower scores.
The remaining variables were seen to be applicable in the following order: SI3 (our business key food suppliers provide input in relation to our projects of product development), establishing a score of a mean of 4.82 out of 7 (0.6886); II2 (the functions adopted by our plant are aligned with its activities) scored a mean of 5.15 out of 7 (0.7357); PP3 (our organisations postpones final packaging activities until the customer is in receipt of their order 5.18 out of 7 (0.7); SI (cooperative relationships with food suppliers are maintained) scored a mean of 5.21 out of 7 (0.74429); and II (top management highlights the importance of good inter-functional relationships) scored a mean of 5.23 out of 7 (0.74714).
When considering the descriptive statistics, we can see that demand mitigation uncertainty was found to have an average score with DMU (we mitigate demand uncertainty when our customers place orders consistent with their nominated delivery lead time), scoring 5.36 out of 7 (0.7657), DMU (we mitigate demand uncertainty by providing products to our customer consistent with their nominated product specification) scoring 5.48 out of 7 (0.7829) and DMU (we mitigate demand uncertainty when our customers provide us reliable forecasts on their demands) scoring 5.48 out of 7 (0.7829).
Although the numerous other elements of mass customisation capacity were carried out much better, MCC2 (we can easily incorporate significant food product variety without the need to increase costs) scored 5.52 out of 7 (0.7886). In contrast, the internal integration variable II1 (there is sound alignment between the functions in our plant) were seen to perform much better when compared with the remainder, scoring a mean of 5.59 out of 7 (0.7985). The postponement practice variables PP1 (our firm postpones final product assembly activities until receives customer orders) and PP2 (our firm postpones final product labelling activities until receives customer orders) were found to score relative higher than PP3, demonstrating a mean score of 5.67 out of 7 (0.81) and 4.86 out of 6 (0.81), respectively.
In a comparable vein, the variables of customer integration demonstrated sound performance with three of its variables, positioning it at the top five variables with the highest scores. CI2 (our customers are actively involved in our product design process) achieved a score of a mean of 4.91 out of 6 (0.8183), MCC3 (we can easily add product variety without sacrificing quality) scored a mean of 5.91 out of 7 (0.8443); CI3 (the customers involve us in their quality improvement efforts) scoring 5.93 out of 7 (0.8471); and CI1 (we are in frequent, close contact with our customers) scoring a mean of 6 out of 7 (0.8571). When considering the highest of the score, the variable MCC1 (we can are highly capable of large-scale product customisation) was found to have a mean of 6.1 out of 7 (0.8714). The table below details these statistics.

Descriptive StatisticsNMinimumMaximumMeanStd. DeviationSkewnessKurtosisStatisticStatisticStatisticStatisticStatisticStatisticStd. ErrorStatisticStd. ErrorAge239142.54.982-.087.157-.994.314How long have you been working with your current employer?2391253.641.8716.357.15770.668.314We are in frequent, close contact with our customers.239376.001.051-.823.157-.091.314Our customers are actively involved in our product design process.239264.911.177-.863.157-.051.314The customers involve us in their quality improvement efforts.239475.931.103-.558.157-1.074.314The functions in our plant are well integrated239375.591.306-.724.157-.735.314Our plant’s functions coordinate their activities.239275.151.310-.700.157-.345.314Our top management emphasizes the importance of good inter-functional relationships.239275.231.251-.327.157-.783.314We maintain cooperative relationships with food suppliers.239175.211.448-.296.157-.893.314We maintain close communications with food suppliers about quality considerations and design changes239174.261.453-.020.157-1.402.314Our firm key food suppliers provide input into our product development projects.239174.821.529-.283.157-.936.314Our firm postpones final product assembly activities until receives customer orders239375.671.242-.854.157-.325.314Our firm postpones final product labelling activities until receives customer orders239164.861.082-1.008.157.629.314Our firm postpones final packaging activities until receives customer orders239275.181.275-.739.157-.119.314We can are highly capable of large-scale product customisation.239476.101.034-.832.157-.567.314We can easily add significant food product variety without increasing costs.239475.52.703-.543.157-.164.314We can easily add product variety without sacrificing quality.239475.911.129-.533.157-1.160.314We mitigate demand uncertainty by providing products to our customer consistent with their nominated product specification.239275.481.371-.670.157-.796.314We mitigate demand uncertainty when our customers place orders consistent with their nominated delivery lead time.239275.361.242-.578.157-.551.314We mitigate demand uncertainty when our customers provide us reliable forecasts on their demands.239275.481.371-.670.157-.796.314We are in a highly competitive industry.239174.751.568-.013.157-1.082.314Our competitive pressures are extremely high.239174.291.779.117.157-1.510.314We do not pay much attention to our competitors.239174.561.436.152.157-.748.314Valid N (list wise)239
Table 11: Descriptive Statistics

In an effort to analyse data normality distribution, there was the examination of skewness and kurtosis. As considered earlier, a kurtosis and skewness level recognised as in the range of ± 2.58 should be seen amongst normally distributed data (Hair et al., 2010). Upon more in-depth examination of the statistics, it was found that all of the variables were seen to be in the acceptable ± 2.58 range, thus suggesting that all of the variables are identified as within normal distribution, but are not entirely distributed normally.

5.3 Structural Model Assessment and Testing
5.3.1 Structural Model
The present work is centred on the SEM, with the application of the PLS-SEM method in mind of validating the link between the constructs of the study. Viewed a second-generation modelling, SEM is known to carry out a dual function: it operates both as a measurement model and as a structural model. In the case of the former (outer model), its functions are concerned with evaluating the overall quality of the individual study constructs. In the case of the latter (inner model), the SEM evaluates the link between the various constructs (Fornell & Brookstein, 1982). The sample size is established in line with SEM and is dependent on various critical aspects. The assumptions of the study are tested with the adoption of multivariate analysis. An estimation approach is identified in order to predict the sample size. Model complexity is examined, with the data screened in mind of missing data and corresponding outliers. The estimation of average variance error is performed (Hair et al., 2010).
In the view of Chin (1998), more conventional significance testing approaches are not considered suitable for PLS-SEM, as they are seen to take on a variance that is distribution-free. Accordingly, the PLS-SEM models need to be evaluated in line with the use of measures that are prediction-centred and non-parametric as opposed to with the use of measures of fit (Chin, 1998). Emphasising the arguments of Chin (1998), the work of Hair et al. (2011) suggests the adoption of the Stone-Geisser test, path coefficient and coefficient of determination (R2) as the most suitable approaches to evaluating the PLS structural model. They further recognise that the resampling approaches, including bootstrapping and jackknifing, may be applied in evaluating not only the significance but also the overall stability of path coefficient evaluations (Hair et al., 2011).
In the case of PLS, the bootstrapping approach is for 500 samples, whilst the goodness-of-fit index is not applied in mind of testing assessment of the inner model (Henseler & Sarstedt, 2013), with Cronbach’s alpha for inner consistency not applied in regard to outer model assessment (Bagozzi & Yi, 1998), respectively. On the other hand, in the present work, which adopts the PLS approach, there is the estimation of path loadings and R2 values. Although path loadings are recognised as establishing the overall strengths of the links between dependent and independent variables, the predictive power of variables is measured with the use of R2. In this vein, R2 values measure the degree of the independent variables’ variance. The software introduced by Ringle et al. (2005), namely SmartPLS 2.0 M3, is adopted in an effort to establish the structural and measurement model. There is the adoption of the bootstrapping estimation procedure in mind of establishing the measurement model’s overall significance of the scale factor loadings and the structural model’s path coefficients (Gefen & Straub, 2005). Furthermore, when evaluating the latter model, it is essential to consider the possibility of the classification of data when assessing not only the observed but also the unobserved heterogeneous variables.

5.4 Exploratory Factor Analysis
This approach is applied for establishing variables and accordingly recommending dimensions (Churchill, 1979; Field, 2005: 234), and identifies the inter-correlation amongst the measurement items, which then are accordingly grouped into sets known as factors. Subsequently, through the use of theory, these factors will correspond to a concept (Hair et al., 2010). Hair et al. (2010) details two key objectives associated with factor analysis: to identify the structure of both respondents and variables; and to complete data reduction/summarisation and the selection of variables. The data are summarised by FA, through defining variables’ structure through positioning them in groups, and accordingly providing variables identification for subsequent analysis and data reduction. The key objective associated with completing exploratory FA for this work is data reduction and summarisation. Through the completion of three steps, the exploratory factor is carried out (Pallant, 2010): the first evaluates the overall data suitability for FA in line with the size of the sample and the inter-correlations between items. In regard to the size of the sample, the ratio between the cases and the number of items is larger than 5:1 (Hair et al., 2010: 87). Moreover, the size of the sample in the present work is recognised as adhering to the threshold outlined by Tabachnick & Fidell (2006), who make the suggestion that ‘it is comforting to have at least 300 cases for factor analysis’. In regard to the present work’s variables, the inter-correlation amongst items is greater than 0.3, as detailed through the correlation matrix. The MSA (Measure of Sampling Adequacy) provides quantification of the inter-correlations between the variables, with values ranging from 0 to 1. Those variables found to have values of 0.5 or more are sound variables, estimated through other variables without errors; whereas those falling below 0.5, on the other hand, need to be disregarded (Hair et al., 2010). In the present work, all of the MSA values for all items are seen to be greater than 0.5, suggesting a good inter-relation between items. Moreover, there has been the adoption of two statistical approaches in order to evaluate data factorability: Bartlett’s Test of Sphericity and Kaiser-Meyer-Olkin (KMO) Measure of Sampling Adequacy (Pallant, 2010, 183). The Test of Sphericity is concerned with establishing the significance of the correlation matrix, where the presence of sufficient correlations between variables is signified by a significance level of <0.5. KMO can be described as the ratio between the sum squared of correlations and the summation of sum squared correlations and the sum of squared partial correlations. FA appropriateness requires a minimum value of 0.6 (Tabachnick & Fidell, 2006). The KMO findings, as well as those derived from the Barlett test, are detailed in Table VII, highlighting the significance of the Bartlett test (p< 0.05) and the exceeding of KMO index above the minimum value of 0.6; thus suggesting the factorability of data.
Table VII
Commonality should be > 5.5
Exploratory Factor Analysis (EFA)

5.5 CFA Analysis
With the use of PLS software, CFA analysis was carried out with the objective to evaluate the overall validity and reliability of the multiple-item scale. Considering the overall reflective nature of the measurement scale, there was the examination of the composite reliability, AVE, outer loadings and square roots. Discriminant validity, convergent validity and reliability were all carried out in consideration to the guidelines provided by Fornell & Larcker. Moreover, composite reliability and indicator reliability tests, which are recognised as superior to Cronbach’s Alpha, are carried out in quantitative analysis owing to the fact such tests consider actual loadings as opposed to assigning assumed equal weight to all items (Fornell & Larcker, 1981). Indicator reliability is concerned with the square root of outer loadings, with a value of 0.70 or greater validating the indicator reliability (Hulland, 1999). Accordingly, there is a need to establish internal consistency reliability if the composite reliability is seen to be 0.70 or greater (Bagozzi & Yi, 1988). When considering that the composite reliability ranges 0.–317 0.9754, which is recognised as larger than the suggested 0.70 value, there was the establishment of internal consistency reliability.
The study questionnaire’s overall reliability was determined following the completion of the pilot study, with content validity identified also before the pilot study (Churchill, 1979). Although face and content validity of the study’s questionnaire was determined in various ways—through a) consideration to academic experts, and b) through the views of 12 CEOs in the completion of semi-structured interviews—construct validity subtypes, including convergent and discriminant validity, can be identified through PLS analysis. Average Variance Extracted (AVE) was also applied in mind of evaluating convergent validity, with the evaluation of discriminant validity in the subsequent process. The presence or absence of convergent validity was established through AVE values. Should the AVE values be equal to or greater than 0.5, then the convergent validity would be determined (Baggozzi & Yi, 1988). When considering that the AVE values are greater than the suggested value of 0.5 (spanning 8206–0.9295), as detailed in Table VIII, and that the outer model loadings, as detailed in Table VII, are larger than the 0.70 values, there is the establishment of convergent validity. Moreover, as a result of the t-statistics, factor loadings were found to exhibit significance at p<0.01 and communalities >0.500, which clearly established convergent validity (Hair et al., 2010).
Note: Outer model loadings or factor loadings are extracted to conduct the CFA analysis
Table 13: Outer Model Loadings (Factor Loadings)
NOTE: CR- composite reliability; AVE-average variance extracted; * all item loadings are significant at the P<0.01 level; n.a: not applicable for single-item construct. ** Please refer to the appendix for all description for the item
Table IX: CFA Analysis

5.5.1 Discriminant Validity
Discriminant validity undergoes evaluation through the contrasting of the square roots of each construct’s AVE alongside the links between the focal construct and all other constructs. Therefore, the establishment of discriminant validity is achieved upon there being a greater square root than the correlation with other constructs (Fornell & Larcker, 1981). The table below provides clear indication as to the inter-construct correlation values of the diagonal of the matrix. On the diagonal, a contrast between the AVE square roots and the correlation values suggests discriminant validity.
Table X: Inter-Construct Correlation Values
Forner-Larcker’s Criterion
*Ssquare root of AVE is written in bold on the diagonal of the table.
Table XI: Cross Loading

T-Statistics of Path Coefficients (Inner Model)
Critical t-values for a two-tailed test is 1.65 (significance level = 10 percent), 1.96 (significance level = 5 percent), and 2.58 (significance level = 1 percent).
Table XII: Checking Structural Path Significance in Bootstrapping

5.5.2 Nomological Validity
In order to ensure constructs’ nomological validity, a theoretical support garnered through prior works for the suggested links between the constructs have been detailed in the theoretical framework chapter in mind of ensuring the soundness of the links between constructs in the measurement theory. Moreover, there has been the analysis of the nomological validity in line with the correlation matrix, as highlighted by Hair et al. (2010).
As shown in the tables, the correlation matrix is detailed, with the P values of correlations detailed in Table 20 providing support for the prediction that such constructs are linked with one another, where these relationships are recognised as sound:

Table 20: Correlations between Constructs
Table 21: P Value of Correlation between Constructs

5.5.3 Multicollinearity
Multicollinearity stems from the links identified between at least two predictors comprised in the tested framework. The most valuable situation for a research is to garner high correlations between the dependent and independent variables, but a minor degree of correlation between the independent variables amongst the independent variables (Hair et al., 2010). A high degree of multicollinearity is recognised as threatening to the overall validity of the results garnered as a result of utilising the tested model as this can result in incorrect predictions pertaining to the regression coefficients, and also its sign (Hair et al., 2010). As has been highlighted by Hair et al. (2010, p. 201), ‘As multicollinearity increases, the total variance explained decreases (estimation). Moreover, the amount of unique variance of independent variable is reduced to levels that make estimation of their individual effects quite problematic (explanation)’.
One approach to evaluating multicollinearity is to determine the correlation maxtrix between the independent variables. The presence of high correlations between independent variables (recognised as 0.90 or more, overall) provides some insight into an issue in multicollinearity (Hair et al., 2010).
In the present work, the correlation matrix between independent variables (as detailed in Table XVI) has been analysed, which suggests no high correlation between the independent variables (where the correlation maxtrix maximum is seen to be 0.6017).
Table 22: Correlation Between Independent Variables
Table 23: FIV’s

In the present work, the independent variables’ VIFs (see Tables XVII) have undergone clarification, where it is highlighted that the largest VIF value is 1.382. Such findings of the correlation matrix and VIFs give overall assurance that there are no concerns in regard to the study’s overall mulitcollinearity.
VIF is determined as ‘1/Tolerance’. Generally speaking, there is a need to garner a VIF equating to 5 or lower (i.e., Tolerance level of 0.2 or higher) in order to ensure the collinearity issue is circumvented (Hair et al., 2011).
Table 24: Stone-Geisser’s (Q2) Value

Q2 values of 0.02, 0.15 and 0.35 provide indication as to an exogenous construct with a small, medium and large predictive relevance for an endogenous latent variable, respectively. Although the measurement model’s quality can be measured through CV-communality, CV-redundancy, on the other hand, is measured through consideration to the structural model’s quality (Tenenhaus et al., 2005). Overall, there is the recommendation of CV-redundancy in evaluating the overall predictive relevance of PLS-SEM, as this is known to utilise both the structure model’s and measure model’s estimates in mind of data prediction (Hair et al., 2011a). Overall, a predictive relevance of a model is indicated by Q2 > 0, whilst Q2 < 0 suggests a complete lack of predictive relevance in a model (Chin, 1998). In specific regard to the present work, both CV-redundancy and CV-communality statistics have been calculated with the use of the blindfolding approach in the application of the Smart-PLS software. The findings of the CV-redundancy and CV-communality statistics (detailed IN Table XV11) suggest a positive Q2 value across all constructs, thus implying an estimated relevance for the framework tested in this work. These findings suggest that the tested model’s proposed structural relationships are not only restricted to the present data set, but also may be applied in mind of estimating the endogenous latent variables through the use of other sets of data.

5.6 Assessing the Structural Model
This study has utilised the R2 change in order to assess the competing framework, as per the recommendation made by Chin (1998). Such an approach, overall, may be applied in order to analyse whether a specific independent latent variable has a notable impact on a dependent variable across two stages (Chin, 1998). In the initial stage, R2 change is calculated by drawing a comparison of R2 both prior to and following the addition of a specific independent latent variable related to the framework. The second stage is concerned with relating the R2 change to (1- R2), achieving the effect size f 2, which could have an effect on the structural model spanning small, medium or large if f2 is found to be equal to 0.02, 0.15 or 0.35, respectively (Chin, 1998). Owing to the fact that the PLS-SEM’s key goals are estimating and maximising on the described variance in the latent endogenous variable, the key criteria of assessing the structural framework needs to be R2, path coefficients and the significance levels of path coefficients (Hair et al., 2011a). Importantly, path coefficient significance may be evaluated with the application of resample approaches, including jackknifing and bootstrapping, owing to the fact that PLS-SEM makes no distribution assumption for the data applied in the analysis. Importantly, bootstrapping is adopted in this work as it is recognised as more efficient than the jackknifing approach (Chin, 1998). Moreover, evaluating the structural framework requires the evaluation of its overall ability to estimate the endogenous latent variables (Hair et al., 2011a). One of the key measures applied in evaluating the predictive relevance in a framework is the Stone-Geisser Q2 value (Geisser, 1974; Stone, 1974). Such a value may be determined with the adoption of the blindfolding technique, which is seen to systematically disregard some of the data and instead apply the resulting predictions in estimating the omitted part (Hair et al., 2011a).

Model’s f2 Effect Size

Furthermore, evaluating the observations’ heterogeneity is a fundamental stage in assessing the structural mode. Not completing an evaluation of data heterogeneity could mean the validity of the PLS-SEM results are potentially damaged owing to the fact that different parameter estimations could be secured for different subpopulations (Hair et al., 2011a). A number of instruments have been devised in PLS-SEM in mind of evaluated the unobserved heterogeneity, including FIMIX-PLS (Ringle et al., 2010). Dissimilar to CB-SEM, in this case, there is no generally goodness of fit measure, with the key goals of PLUS-SEM recognised as different to those of the CB-SEM (Hulland, 1999; Hair et al., 2011b). A number of academics have presented global measures of fit (e.g. Tenenhaus et al., 2004). Importantly, however, such measures are not widely acknowledged, with some holding the view that it is inconsistent with PLS-SEM objectives and assumptions (Hulland, 1999; Hair et al., 2011b).
5.6.1 PLS-SEM Software
Although the underlying algorithms applied in the case of PLS-SEM were devised in the 1970S, the first software packages, namely LVPLS (Lohmoller, 1984) and PLS Path (Sellin, 1989), were not made available to the public prior to the 1980s (Temme et al., 2010). The somewhat restricted application of PLS-SEM throughout the course of recent years may be partly attributed to the lack of progress in regard to the development of the software in regard to availability, user-friendliness and methodological options (Temme et al., 2010).
Nonetheless, this particular situation has undergone much change, with a number of alternative software solutions now available for selection, including PLS-GUI, Visual-PLS, PLS-Graph, Smart PLS and SPAD-PLS (Temme et al., 2010). Furthermore, all software packages are known to comprise different features in regard to ease of use, methodology, options and requirements (Temme et al., 2010). This research utilises both Warp-PLS and Smart-PLS, with all of the software packages each having their own distinctive characteristics. The most recently available of these is Warp-PLS (Kock, 2011), which is able to provide various features that are, in the main, lacking across the majority of the other, if not all other, PLS-based SEM packages available at the current time (Kock, 2011):
1. It provides the automatic prediction of the p values for path coefficients rather than providing merely standard errors or t values, and allows the user to determine the p values.
2. It predicts a number of model fit indices, which have been designed in a way to be valuable in the context of PLS-based SEM analyses.
3. There is the automatic creation of the indictors’ product structure underlying moderating relationships, which provides further expansion in this regard. It emphasises moderating relationships, associated path coefficients and associated p values in a model graph.
4. It enables the use of scatter plots pertaining to all of the links amongst latent variables, together with the regression curves that are most suitable for such relationships, where such plots are saved as .jpg files for inclusion in reports.
5. Variance Inflation Factor (VIF) coefficients are calculated in mind of the latent variable predictors linked with each latent variable criterion. This enables users to determine whether various predictors need to be removed as a result of multicolinearity (where such a feature is recognised as notably valuable with latent variables that are measured in line with only one or several indicators. Notably, however, the inclusion of tools centred on evaluating predictive relevance, such as blindfolding approaches, or otherwise unobserved heterogeneity (including FIMIX-PLS) are not included in Warp-PLS. Such instruments are incorporated within the Smart-PLS software. Accordingly, this was applied in the present work in mind of cross-checking the findings garnered by Warp-PLS and the application of both the FIMIX-PLS and blindfolding methods.

5.6.2 Goodness of Fit
In contrast to CB-SEM, there is no suitable overall criterion for goodness of fit in specific regard to PLS-SEM (Hulland, 1999; Hair et al., 2011b). Although CB-SEM comprises parametric estimation approaches that are seeking to reproduce, as closely as possible, the observed covariance matrix, the key goals of PLS-SEM are centred on reducing the error or otherwise improving the variance described in endogenous variables measured by R2 (Hulland, 1999). A number of academics in the field provide a goodness of fit criteria, including the Bentler-Bonett fit index (Bentler & Bonett, 1980), as well as the global criterion for goodness of fit (Tenenhaus et al., 2004). Nonetheless, such criteria have faced much criticism for lacking value and meaning owing to their assumption that the predicted model parameters have been chosen in mind of decreasing the differrnt between the reproduced and the observed covariance matrices (Hulland, 1999). Moreover, the goodness of fit global criterion is centred on the model average R2 and the average communality of reflective models, which do not apply in this instance of single indicator constructs or formative models (Hair et al., 2011b). Furthemore, establishing a onset for an acceptable goodness of fit measure level is not a simple task when considering there is no common acceptable threshold of R2 (Hair et al., 2011b). This work does not deliver goodness of fit measures owing to the lack of appropriateness of the goodness of fit measures to PLS-SEM, particularly where some constructs are measured with the use of single measures.

5.6.3 Heterogeneity
Furthermore, evaluating the observations’ heterogeneity is a fundamental stage in assessing the structural mode. Not completing an evaluation of data heterogeneity could mean the validity of the PLS-SEM results are potentially damaged owing to the fact that different parameter estimations could be secured for different subpopulations (Hair et al., 2011a). A number of instruments have been devised in PLS-SEM in mind of evaluated the unobserved heterogeneity, including FIMIX-PLS (Ringle et al., 2010). Dissimilar to CB-SEM, in this case, there is no generally goodness of fit measure, with the key goals of PLUS-SEM recognised as different to those of the CB-SEM (Hulland, 1999; Hair et al., 2011b). A number of academics have presented global measures of fit (e.g. Tenenhaus et al., 2004). Importantly, however, such measures are not widely acknowledged, with some holding the view that it is inconsistent with PLS-SEM objectives and assumptions (Hulland, 1999; Hair et al., 2011b).

5.7 Common Method Variance (CMV)
Common Method Variance (CMV) is an essential consideration necessitating fundamental attention in surveys that are carried out with the use of multiple respondents. A number of analyses have been carried out in mind of establishing the presence of CMV in line with the guidelines devised and presented by Podasakoff et al. (2003). Harmon’s single-factor test centred on the analytical method devised by Liang et al. (2007) is carried out in mind of establishing CMV. Throughout the course of the obtained data, should one covariance be considered by one factor, then CMV will be found to be present. Moreover, the correlation matrix will be checked in order to establish the presence of excessively high correlations (<0.9). The test results provide clear indication of the potential of CMV influence on the results of the study.
Despite the fact that CMV has been managed with the application of various respondents, it remains a consideration requiring much more focus in the context of survey-based works. As such, a series of analyses have been conducted in order to identify the presence of CMV in this study, with adherence to the guidelines devised by Podsakoff et al.(2011). As a primary step, Harmon’s single-factor test was carried out through adopting the analytical procedure suggested by Podsakoff et al. (2003), where CMV is present in the data if one factor is seen to account for the majority of the covariance. Secondly, the correlation matrix was checked. It is noted that CMV is not likely to be identified if correlations are not excessively high (>0.9). The findings garnered through the completion of these tests imply that CMV was not likely to have non-important influence on the results.

5.8 Coefficient of Determination (R2)
Coefficient of determination (R2) is the leading criterion applied in the assessment of the inner framework, and is known to represent the amount of variance described of the endogenous latent variable (Hair et al., 2011b). Nonetheless, establishing which R2 level is high differs across all disciplines, as highlighted by Hair et al. (2011a).
Moreover, there is a difference between PLS and LISREL-type SEM owing to the fact that the former has much dependent on independent variables’ predictive power (Chin, 1998). This may be capitalised in such a way so as to describe the links and accordingly applied in order to build theory. A component-based approach, as introduced by Lohmoller (1988), is PLS, where problems linked to inadmissible solutions and factor indeterminacy may be circumvented through the use of PLS (Fornell & Brookstein, 1982). Moreover, PLS further works better than AMOS and LISREL owing to the fact there is no need to make the assumption pertaining to normal data distribution (Gefen & Straub, 2005; Chin, 1998). In consideration to the various non-normality conditions, PLS has the ability to execute functionalities. Although LISREL-type SEM delivers sound goodness of fit indices, PLS predicts R2 values and path loadings. Importantly, they do not provide goodness-of-fit indices. Although path loadings are able to establish a relationship’s strength in regard to dependent and independent variables, the predictive power of the variables is measured with the use of R2, which measure the degree of variance present in the independent variables (Gefen & Straub, 2005). Although the aforementioned contrasts deliver an overview between SEM-based approaches (PLS), and the CBSEM and PLS approaches and LISREL-based SEM, it remains that the use of PLS can be justified through drawing a contrast between it and AMOS. When considering CFA adoption in the quantitative analysis, a contrast between AMOS and SmartPLS for CFA, the advantages of utilising SmartPLS software may be highlighted.
It is clear that one of the most commonly implemented statistical software for CBSEM is AMOS (Hair, Ringle & Sarstedt, 2011). This approach is commonly implemented when the aim of a study is centred on theory testing, theory confirming or otherwise drawing a comparison between theories. Should the measurement model’s formative measures be restricted to the specific rules and additional criterion, including co-variation, then AMOS centred on CBSEM is applied. Should the structural framework be non-recursive, then the application of AMOS would be opted for. Should requriements with regard to framework specifications, non-convergence and data distribution identification and assumptions be fulfilled as per CBSEM, then AMOS software is applied. Moreover, if a global goodness of fit criteria is required in the study, with measurement model invariance to be tested, then AMOS would be most valuable for completing such analyses (Hair et al., 2011).

5.9 Path Coefficients
The examination of the structural model is carried out through PLS, with the various outlined hypotheses subjected to testing. A basic model with key effects is primarily created and tested, the findings of which are detailed in Figure 4 below. As can be seen from this depiction, the DUM shows a 28% variance (R2) and 33% in PP, as explained.

Figure 24: Structural Model with Path Coefficient Estimates
5.10 Relationship between Various Constructs
The present work has provided the hypothesis that SCI not only has a notable indirect and direct effect on MMC and PP, but also adopts a key role in the implementation of PP as a fundamental strategy, empowering MCC to mitigate demand uncertainty. In an effort to establish the link between the numerous hypothesised constructs, the study subjects were asked a number of different questions centred on gathering data that would enable the link between the constructs to be established.
These centre on the link between II and external integration (SI and CI); the link between the various forms of SCI (II, CI and SI) and postponement (PP); the link between SCI and Mass customisation capability (MCC); the link between II and PP; the link between II and MCC; the link between CI and PP; link between CI and MCC; the link between SI and PP; the link between SI and MCC; the link between MCC and PP; the contingent effects of Demand Uncertainty and Competitive Intensity; and the link between II, SI, CI, PP, and MCC with Demand Uncertainty Mitigation (DUM).

5.10.1 Direct Effects
These models’ derived values may be applied in order to establish the indirect and direct effects of the constructs of the study on one another. Direct effects are garnered in mind of validating the link between each PP and SCI type. In line with the path coefficients, internal integration (0.199, p < 0.01), customer integration (0.318, p < 0.001) and supplier integration (0.253, p < 0.001) are seen to have a direct effect on postponement. This provides H2, H3 and H4 with validation. Furthermore, the direct impact of PP on DUM is also validated by the path coefficient (0.534, p > 0.001).

5.10.2 Mediating Effect (Indirect Effect):
In order to establish whether or not external integration causes the effect of internal integration on postponement practice, the indirect effects were calculated through the multiplication of path coefficients from internal integration through to external integration (a), as well as from external integration through to postponement (b). The indirect effect associated with customer integration is 0.397* 0.318 = 0.126; the indirect effect of supplier integration is 0.453* 0.253= 0.114. Importantly, as detailed in Appendix I, H10 and H11 are supported.
In order to establish whether the effect of internal integration to mass customisation capability is carried out by external integration, the indirect effects were calculated through multiplying the path coefficients from integration through to external integration (a), as well as from external integration through to mass customisation capability (b), where the customer integration indirect effect was found to be 0.397* 0.196= 0.077, and the indirect effect of supplier integration was 0.453* 0.094= 0.042. Accordingly, as shown in Appendix I, H12 and H13 are supported.
In order to establish whether the effect of postponement practice to demand uncertainty is carried out by mass customisation, the indirect effects were calculated through multiplying the path coefficients from postponement practice to mass customisation capability (a) and from mass customisation capability to demand uncertainty (b), where the indirect effect of postponement practice was found to be 0.383* 0.305= 0.116. Appendix I provides details pertaining to the support of H14.
Subsequently, there was the application of the Sbel A-Test in mind of evaluating the significance of these indirect effects. Through the resultant Z values, it was indicated that there is an indirect effect stemming from customer integration, which his significant at the p < 0.05 level, whilst supplier integration, on the other hand, is significant at the p < 0.01 level (on both postponement practice and mass customisation capability). Furthermore, the mass customisation capability has an indirect level at the p < 0.05 level in regard to demand uncertainty (James et al., 2014: 71–71, 72).

5.10.3 Moderating Effect
In order to complete the testing of the moderating effects relating to competitive intensity in regard to the internal integration’s indirect effects through external integration on both mass customisation capability and postponement practice, there was the building of conditional indirect models, adhering to the process detailed by Iacobucci, as shown in Figure 5.

Figure 25: Conditional Indirect (Moderated Mediation) Models

In an effort to create the conditional indirect models, moderator and four interaction terms were incorporated as follows:
Two interaction terms for postponement practice side (Internal integration * moderator and supplier integration * moderators), and (Internal integration* moderators and customer integration* moderators), two interaction terms of mass customisation side (Internal integration * moderator and supplier integration * moderators), and (Internal integration* moderators and customer integration* moderators) to the basic model. The interaction terms were computed by cross multiplying the standardized items of each construct. If path a’ is significant, then path a (in the mediation model) is significantly moderated, but the moderator, and the same rule is applied to b’ and c’. If path a’*b’ is significant, then the indirect effect of a*b is significantly moderated by the moderator. It should be noted that when interpreting the conditional indirect effect, the bold path (i.e., a’, b’, and c’) alone should be examined. Other paths were also added to the model, but they used as controls and should not be interpreted from any perspective, as suggested by Iacobucci (2008). In adherence to this particular approach, the moderated mediation results, as detailed in figures 26–29, were obtained.
As can be seen displayed in Figure 26, the moderating effects of competitive intensity on the path from internal integration to supplier integration is negatively significant (β = – 0.171; p < 0.01). Moreover, the moderating effects of competitive intensity on the path from supplier integration to postponement practice is significant (β = 0.129; p < 0.01). The products a’∗b’ (0.171∗0.129 = 0.022) is significant at the p < 0.05 level. As shown in Appendix I, H15 is validated.
Figure 26: Conditional Indirect Effect

As can be seen in Figure 27, the moderating effects of competitive intensity on the path from internal integration to customer integration is significant (β = 0.185; p < 0.01). Importantly, however, the moderating effects of competitive intensity on the path from customer integration to postponement practice is not deemed significant (β = 0.05). The products a’∗b’ (-0.185∗0.05 = – 0.009) is negatively not significant. Accordingly, as shown in Appendix I, H16 is rejected.
Figure 27: Conditional Indirect Effect.

As detailed in Figure 28, the moderating effects of competitive intensity on the path from internal integration to supplier integration is significant (β = 0.171; p < 0.01). The moderating effects of competitive intensity on the path from supplier integration to mass customisation capability is negatively significant (β =- 0.166; p < 0.01). The products a’∗b’ (-0.171∗0.166 = – 0.028) is negatively significant at the p < 0.05 level. Accordingly, as shown in Appendix I, H17 is accepted.

Figure 28: Conditional Indirect Effect

As can be seen in Figure 29, the moderating effects of competitive intensity on the path from internal integration to customer integration is significant (β = 0.185; p < 0.01). The moderating effects of competitive intensity on the path from customer integration to mass customisation capability is negatively significant (β = -0.40; p < 0.01). The products a’∗b’ (0.185∗-0.40 =- 0.074) is negatively significant at the p < 0.05 level. H18 is supported. Accordingly, these results provide some degree of validation for H15, H17 and H18, despite the fact that H16 is unsupported. Details can be found in Appendix I.

Figure 29: Conditional Indirect Effect

As detailed in Figure 28, the moderating effects of competitive intensity on the path from internal integration to mass customisation capability is negatively significant (β = – 0.165; p < 0.01). The moderating effects of competitive intensity on the path from mass customisation capability to demand uncertainty mitigation is negatively significant (β = – 0.582; p < 0.01). The products a’∗b’ (- 0.165∗-0.582 = – 0.096) is significant at the p < 0.05 level. In a comparable vein, as detailed in Figure 29, the moderating effects of competitive intensity on the path from internal integration to mass customisation capability is negatively significant (β = – 0.134; p < 0.01). The moderating effects of competitive intensity on the path from mass customisation capability to demand uncertainty mitigation is negatively significant (β = – 0.229; p < 0.01). The products a’∗b’ (- 0.134∗-0.229 = – 0.030) is significant at the p < 0.05 level. As shown in Appendix I, H19 is validated.

5.11 Control Variables
The analysis comprised various control variables, through which plant size was measured by querying respondents to indicate the number of employees in a firm. In line with previous studies, logged scores were computed and utilised as the size variables (Swamidass & Kotha, 1998: 87). Furthermore, dummy variables were also included in mind of controlling for firm age effects. Moreover, firm production type (fresh meals, pre-cooked, raw material wholesale) was evaluated.
As shown in Figure 6 and Figure 7, it was found that firm size has a positive significant effect on postponement practice (β= 0.497; p < 0.01). There are negative effects stemming from firm age on postponement practice β= -0.422; p < 0.01). The product type has no significant effect on postponement practice β= 0.009. On the other hand, it was found that firm size, firm age, and firm product type have no significant effect in regard to demand uncertainty mitigation, β= -0.003, β= -0.13, β= -0.021, respectively.
As shown in Figure 6 and Figure 7, firm size was found to have positive significant effects on postponement practice (β= 0.172; p < 0.01). Moreover, a significant effect is seen to stem from firm age on postponement practice (β= -0.198; p < 0.01) at p. Product type has no significant effect on postponement practice β= 0.022. Furthermore, firm size and firm product type have no significant effect on demand uncertainty mitigation, β= -0.123, β= -0.28, respectively, whilst a negative effect is apparent between firm age and demand uncertainty mitigation (β= -0.198; p < 0.01).

5.12 Managerial Implications
On an annual basis, the Hajj is carried out by millions of Muslims in the Kingdom of Saudi Arabia (KSA), as recognised by many scholars, including (Long, 2014). It is recognised that, as a result of Hajj in 2012 alone, the KSA secured revenues equating to US$16.5 billion, 3% of which was acknowledged as Gross Domestic Product (GDP). Considering the global involvement of this celebration and the size of the event, it is imperative that food supply is consistent, which is one of the most pressing requirements throughout the season (Rashid, 2012). Moreover, during this time, food is supplied through service offices, self-cooking and missions (Asem Arab Study, 2010). Furthermore, more than half (58%) of the pilgrims are known to visit from international regions (Kaysi et al., 2010); in 2014, pilgrims from 163 countries visited the KSA and were participants in Hajj (Blake, 2014). Throughout the Hajj season, meals were made up of Arab local meals, such as grilled foods, Saudi local meals, Asian local meals, light meals, such as sandwiches, and ready meals, as established by the Asem Arab Study (2010), all of which and are provided through a multitude of entities, including charitable organisations, catering services, fast meal services by car, restaurants, meals by cars equipeed with chillers, and other foods carried by workers (Asem Arab Study, 2010). The countless number of pilgrims in the Kingdom of Saudi Arabia therefore are recognised as having a diverse range of demands, expectations, needs and tastes, especially in regard to food (Kaysi et al., 2010). Various cuisines and different international chefs in the KSA have been identified throughout the period of Hajj (Al Arabiya, 2013), with the congregation recognised as representative of countless different cultures. Moroever, there is a need for food suppliers to ensure the deliverance of particular cuisines in order to fulfil the needs of pilgrims (Ebdon, 2013). Pilgrims and tour operators all have a significant degree of dependence on catering companies and restaurants when seeking food during this season (Al Arabiya, 2013). Through the use of both domestic and foreign channels, food is sourced via well-connected markets. Although fruits and vegetables are sourced through the use of domestic suppliers (Mousa, 2013), meats and poultry, on the other hand, are imported from foreign countries (AgroChart, 2013). Exclusive agents or private label owners in the country ensure the direct provision of food to customers through key service outlets, and also indirect food supply to consumers through smaller food service outlets (Mousa, 2013).
Accordingly, the present research has centred on investigating and examining the link between external and internal integration, and its interaction with the postponement practice to mitigate the uncertainty of demand amongst Small and Medium Enterprises throughout the period of the Hajj season. The findings from the present work show that SCI has a notable effect on MCC and PP, where the interlink between the SCI types are seen to mitigate demand uncertainty. Firstly, a direct and positive effect on postponement is seen as a result of internal integration, which is a finding seen to be aligned with various works, all of which highlight the value of internal integration to postponement (Waller et al., 2000; Towill et al., 2000). Through breaking down and dividing conventional silos and accordingly inspiring and sanctioning learning across a wealth of various firm functionalities and asset combination, the activities pertaining to internal integration encourage and facilitate the use of inside firm assets and proficient in such a way that they can efficiently ensure postponement. This finding is recognised as in support of the RBV theory. In the KSA, it is therefore recognised that SMEs are required to coordinate, collaborate and cooperate with their varied processes, resources, equipment and people in the preparation, packing, processing, preserving and storing of foods throughout the season of Hajj. All SMEs need to examine a number of different aspects of postponing activities, and accordingly make the decision as to when, where and how they can integrate postponement within their internal functionalities.
Secondly, a direct and positive effect on postponement is recognised as a result of customer integrate, which further supports the results from crucial studies that validate the effects of CI on postponement (Towill et al., 2000). A number of pivotal competencies and skills required in the business domain can be picked up by producers through ensuring the coordination of both outer and inner assets from production network accomplices, particularly from customers, thus providing the ERBV with experimental validation. Such a finding is recognised as reliable in line with the administration predominant rationale that suggests there is a need for the producer to assume network partners adopt the role of asset integrator and concrete esteem (Vargo & Lusch, 2008). When considering that the supply chain aim is concerned with the management of customer requirements in terms of fulfilling their needs and accordingly assessing market characteristics, it then may be seen that customer integration adopts a pivotal role in business success (Stevens, 1989). When considering that there is much dependence of postponement practice on the input of customers in the fabrication of an organisation’s processes and products, Saudi-based SMEs should carry out market research centred on reviewing and analysing historical data concerning the Hajj pilgrims’ food preferences and nationalities. This can be achieved through the use of cross-sectional data collection from travel agencies or otherwise through KSA’s Department of Tourism. For example, when considering the cross-sectional data, SMEs will be well positioned to choose to assemble raw materials that are based on the market input and built to stock; these are then accordingly assembled by the firm in order to create the final product (Ogawa & Piller, 2006) in line with the needs and preferences of the customer.
Third, a positive and direct effect on postponement is witnessed as a result of supplier integration, which again is seen to be in line with past works carried out in the context of supply chain (Feitzinger & Lee, 1997; Van Hoek, 1998) and also in alignment with a number of works by Fine & Freund (1990), Vesanen (2007), Kumar (2007), Mikkola & Skjott-Larsen (2004), Prasad et al. (2005) and Yang et al. (2005) in the domain of supplier integration, which emphasise the advantages and role adopted by supplier integration in the field of supply chain networks. A supplier network is considered fundamental to effective postponement application when it is able to supply administrations and parts. Accordingly, suppliers could also be included in item improvement as a sound, strong information supply unit in the arena of modules and parts. Combining together the three tiers of the supply chain, the conclusion may be drawn that SCI has a direct and positive impact on postponement. In the Kingdom of Saudi Arabia, it is necessary for SMEs to establish strong links with their individual suppliers in an effort to ensure a consistent supply of raw materials and/or food products is maintained, which can be achieved through trust and loyalty. Through gaining insight into the relationships with suppliers, SMEs also can position their varied requests in relation to their postponements and also may outline their needs across different suppliers, without affecting their own and their supplier’s performance, and accordingly managing customers and the bullwhip at one time.
Fourth, internal integration is recognised as having a positive effect on supplier and customer integration, which also is seen to be in support of past evidence and literature (Zhao et al., 2011; Braunscheidel & Suresh, 2009; Koufteros et al., 2005). Moreover, the findings garnered by the study suggest that there is a significant, indirect effect on postponement as a result of internal integration, which is seen through the improvement of supplier and customer integration. This provides clear insight to the fact that the SCI foundation is initiated through internal integration. This further ensures that effective SCI, without significant efforts directed towards establishing strong buyer–supplier relationships, are restricted and of very limited value, or otherwise may be wasted.
Fifth, in the present study, the incorporation of the contingent factor competitive intensity has proven that, through supplier integration, there is a significant and indirect effect of competitive intensity on internal integration on postponement practice. These results show that, through supplier integration, competitive intensity improves organisations’ operational levels and further allows the internal functions of the business to be integrated, thus helping to create the necessary effects in regard to the postponement practice. Nonetheless, the enhancement effects of competitive intensity on the link between customer integration, postponement practice and internal are disregarded and contradicted through the lack of indirect and significant effect of competitive intensity on enhancing internal integration on postponement practice. In the supply chain literature, an exceptional addition emphasises that SMEs located in Saudi Arabia are better able to focus on supplier integration as opposed to customer integration; postponement practices, on the other hand, depend on competition intensity. Such aspects highlight the conclusive role adopted by competitive intensity in PP and SCI, as illustrated during the stage of hypotheses development in the present work.
Sixth, a direct and positive effect on demand uncertainty is apparent as a result of postponement. This is in line with past works, which have highlighted postponement practices; applicability in regard to management uncertainties (Koh et al., 2007), and eventually removes uncertainties (Taylor, 2004). Pivotal in insulating upstream supply chain aspects from final customer demand, fundamental interventions in the types and tiers of supply chain are induced through postponement, which mitigate or otherwise completely eradicate uncertainties (Waters, 2007).
Finally, the firm’s age and size are seen to be key factors in postponement practice, with the results emphasising the positive significant effect of the size of the firm on postponement practice, and the negative effect on postponement practice as a result of the age of the firm. This suggests that, in businesses, postponement practice is directly affected by the number of staff in the firm, with an inverse link also demonstrated in the case of the age of the firm. The identification of this link means organisations now need to consider their age and size when implementing postponement practice. Nonetheless, the absence of the significant link between the age and size of the firm, and the type of product the firm provides, in line with the DUM, facilitates the study’s generalisation across a number of KSA food industries in regard to the DUM aspect.
Essentially, therefore, the present work closes the gap in regard to the effects of Supply Chain Integration on postponement practice, and seeking to mitigate demand uncertainty throughout the period of the Hajj season. SMEs therefore need to consider and prioritise the coordination of their supply chain strategies in line with postponement practice to their business environment, especially in line with high competitive intensity, in order to improve demand uncertainty mitigation throughout peak periods; this will help in dealing with the uncertainty of demand.
The findings of this research are recognised as pivotal to those SMEs operating in the food sector overall, as well as those specifically affected by seasonal events such as Hajj, which are distinguished by their large number of options, short lead times and uncertain levels of demand.

REFERENCES
Abdelkafi, N. (2008), Variety Induced Complexity in Mass Customisation Concepts and Management. Berlin, Erich Schmidt.
Alarape, A. A. (2007), ‘Entrepreneurship Programmes, Operational Efficiency and Growth of Small Business, Journal of Enterprising Communities, 1 (3), p. 222
Al-Awwaad, A. (2007), Eliminating Minimum Capital Requirement and Facilitating Business Start-up in Saudi Arabia, European Business Review Jeddah, SmartLesson,com, cf. Doing Buisness in Saudi Arabia.
Aldenderfer, M. S. & Blashfield, R. K. (1984), Cluster Analysis, Sage, London.
Ambrosini, V., Bowman, C. & Collier, N. (2009), ‘Dynamic Capabilities: An Exploration of How Firms Renew their Resource Base’, British Journal of Management, 20: S9-S24
Amit, S. et al., (2005), ‘Supply-Chain Networks: A complex adaptive systems perspective’, International Journal of Production Research, 43 (20), pp. 4235–4265.
Amosalam, S. (2008), ‘Factors Influencing the Entrepreneurial Development of Small to Medium Sized Enterprises (SMEs) in Saudi Arabia’, A Management Project Presented in Part Consideration for the Degree of ‘MBA in Entrepreneurship, pp. 1–94.
Angerhofer, B. J. & Angelides, M. C. (2000), System Dynamics Modelling in Supply Chain Management: Research review, In Simulation Conference, 2000, Proceedings, winter (Vol. 1, pp. 342-351). IEEE.
Antonio K. W., Lau, E. T. & Richard, C. M. (2010), Effects of Supplier and Customer Integration on Product Innovation and Performance: Empirical Evidence in Hong Kong Manufacturers. 27, p. 761–777
Arabie, P. & Huber, L. (1994), ‘Cluster Analysis in Marketing Research’. In: Bagozzi, R. P. (Ed.), Advanced Methods in Marketing Research, Blackwell, Oxford, pp. 160–189.
Ayers, J. B. (2001), Handbook of Supply Chain Management, CRC Press.
Bagozzi, R. P. & Yi, Y. (1988), ‘On the Evaluation of Structural Equation Models’, Journal of Academy of Marketing Science, Vol. 16 No. 1, pp. 74–94.
Bajpai, N. (2011), Business Research Methods. Delhi, Pearson. http://proquest.safaribooksonline.com/?fpi=9789332510449.
Baofeng, H. (2012), The Impact of Supply Chain Integration on Company Performance: An organisational capability perspective.’ Supply Chain Management: An International Journal, Vol. 17 Iss: 6 pp. 596–610.
Barney, J. B. (2013), ‘Purchasing, Supply Chain Management and Sustained Competitive Advantage: The Relevance of Resource-based Theory,’ Journal of Supply Chain Management, 48(2), pp. 3–6.
Battezzati, L. & Magnani, R. (2000), Supply Chains for FMCG and Industrial Products in Italy: Practices and the advantages of postponement. International Journal of Physical Distribution & Logistics Management, 30 (5), pp. 413–424.
Bhatnagar, R. & Sohal, A. S. (2005), Supply Chain Competitiveness: Measuring the impact of location factors, uncertainty and manufacturing practices, Technovation, 25 (5), 443–456.
Blecker, T., Abdelkafi, N., Friedrich, G., Kaluza, B. & Kreutler, G. (2005), Information and Management Systems for Product Customisation. Boston, MA, Springer Science + Business Media, Inc.
Boër, C. R. (2013), Mass Customisation and Sustainability an Assessment Framework and Industrial Implementation. London, Springer.
Bohnenkamp, T. (2013), ‘The Effect of the Resource-Based View on Decisions in Supply Management,’ 1stIBA Bachelor Thesis Conference, June 27th, 2013, Enschede, The Netherlanda, pp. 1–14
Boyle, E., Humphreys, P. & McIvor, R. (2008), ‘Reducing Supply Chain Environmental Uncertainty through E-intermediation: An organisation theory perspective’, International Journal Production Economics, 114 (1), 347–362.
Braunscheidel, M. J., & Suresh, N. C. (2009), The Organisational Antecedents of a Firm’s Supply Chain Agility for Risk Mitigation and Response, Journal of Operations Management, 27 (2), 119–140.
Brun, A. & Zorzini, M. (2009), ‘Evaluation of Product Customisation Strategies through Modularization and Postponement. International Journal of Production Economics. 120, pp. 205–220.
Bundagji, F. (2005), Small Business and Market Growth in Saudi Arabia. Jeddah, Jeddah Chamber of Commerce and Industry.
Burgess, K., Singh, P. J. & Koroglu, R. (2006), ‘Supply Chain Management: A structured literature review and implications for future research’, International Journal of Operations and Production Management, 26 (7), 703–730.
Can, K. C. (2012), Postponement, Mass Customisation, Modularization and Customer Order Decoupling Point: Building the Model of Relationships. Master Thesis Department of Management and Engineering Linkoping University, pp. 1–74
Can, C. K. (2008), ‘Postponement, Mass Customisation, Modularization and Customer Order Decoupling Point: Building the Model of Relationship, pp. 1–52.
Capitas Group International (n.d.), Importance of SME Authority for Saudi Arabia. Pop. 1-4
Carneiro, L. M. (2012). Intelligent non-hierarchical manufacturing networks. London, ISTE Ltd. and John Wiley & Sons.
Carvalho, H, Cruz-Machado, V. & Tavares, J. G. (2012), A Mapping Framework for Assessing Supply Chain Resilience, International Journal of Logistics Systems and Management, 12 (3), pp. 354–373.
Cavusoglu. H, Cavusoglu, H. & Raghunathan, H. (2012), ‘Value of and Interaction between Production Postponement and Information-sharing Strategies for Supply Chain Firms.21 (3), pp. 470–488.
Chandra, C. & Kamrani, A. K. (2004), Mass Customisation: A supply chain approach. New York, Kluwer Academic/Plenum Publishers.
Chandra, C. & Kumar, S. (2001), Enterprise Architectural Framework for Supply Chain Integration. Industrial Management & Data Systems, 101 (6), pp. 290–304.
Chavez, R., Gimenez, C., Pieter van Donk, D. & Taco van der Varrt (n.d.), ‘Customer Integration, Information-Sharing and Operational Performance,’ A relational view, pp. 1–10.
Chen, I. J. & Paulraj, A. (2004), ‘Towards a Theory of Supply Chain Management: The constructs and measurements’, Journal of Operations Management, Vol. 22 No. 2, pp. 119–150.
Chen, J. J. & Paulraj, A. (2008), ‘Towards a Theory of Supply Chain Management: The constructs and measurements,’ Journal of Operations Management, 22 (2): 119–150.
Chin, W. W. (1998), ‘The Partial Least Squares Approach to Structural Equation Modelling’. In G. A. Marcoulides (Ed.), Modern Methods for Business Research (pp. 295–336). Mahwah, N. J.: Lawrence Erlbaum Associates, Publisher.
Choi, T & Cheng, T, C, (2011), Supply Chain Coordination under Uncertainty. Berlin, Springer Berlin.
Choi, T. & Krause, D. R. (2006), ‘The Supply Base and its Complexity: Implications for transaction costs, risks, responsiveness, and innovation’, Journal of Operations Management, Vol. 24 No. 5, pp. 637–652.
Chung, W. W. C., Yam, A. Y. K. & Chan, M. F. S. (2004), ‘Networked Enterprise: A new business model for global sourcing’, International Journal of Production Economics, Vol. 87 No. 3, pp. 267–280.
Churchill, G.A. Jr. (1979), ‘A Paradigm for Developing Better Measures of Marketing Constructs’, Journal of Marketing Research, vol. 16, no. 1, pp. 64–73.
Christopher, M. (2005), Logistics and Supply Chain Management: Creating value-adding networks, Pearson Education
Christopher, M. (2005), Logistics and Supply Chain Management: Creating value-adding networks, Pearson Education.
Clarke, E. (2007), ‘Chain Reaction’, Financial Management, April, pp. 10–14.
Cohen, S. & Roussel, J. (2005), Strategic Supply Chain Management: The five disciplines for top performance, McGraw-Hill, Boston, MA, pp. 139–167.
Copacino, W. (1997), Supply Chain Management: The Basics and Beyond. Florida, CRC Press.
Cousineau, D. & Chartier, S. (2010), ‘Outliers Detection and Treatment: A review’. International Journal of Psychological Research, 3(1), 58–67.
Coyle, J. J. & Coyle, J. J. (2009), Supply Chain Management: A logistics perspective. Mason, OH, South-Western Cengage Learning.
Craig, S. (2005), International Marketing Research. Chichester, John Wiley & Sons.
Currie, G. & Shalaby, A. (n.d.), ‘Synthesis of Transport Planning Approaches for the World’s Largest Events, Transport Review’, A Transnational Transdisciplinary Journal, 32:1, pp. 113–136.
Da Silveira, G., Borenstein, D. & Fogliatto, F. S. (2001), ‘Mass Customisation: Literature review and research directions. International Journal of Production Economics, 72 (1), pp. 1–13.
Dam Jespersen, B. & Skjøtt-Larsen, T. (2006), Supply Chain Management: in theory and practice. Frederiksberg, Copenhagen Business School Press.
Darling, J. M. et al, (2007), ‘Enhancing Contemporary Entrepreneurship’ A Focus on Management Leadership’, 19 (1), p. 4.
Das, A., Narasimhan, R. & Talluri, S. (2006), ‘Supplier Integration—Finding an Optimal Configuration;, Journal of Operations Management, Vol. 24 No. 5, pp. 563–582.
Das, S. K. & Abdel-Malek, L. (2003), ‘Modelling the Flexibility of Order Quantities and Lead-Times in Supply Chains’, International Journal of Production Economics, Vol. 85 No. 2, pp. 171–181.
Davis, T. (1993), ‘Effective Supply Chain Management’, Sloan management review, 34, pp. 35–35.
Dillon, W. R. & Goldstein, M. (1984), Multivariate Analysis: Methods and applications, John Wiley & Sons, New York.
Dixon, H. D. (2004), The New Macroeconomics: Imperfect markets and policy effectiveness. Cambridge, Cambridge University Press.
Donaldson, L. (2006), ‘The Contingency Theory of Organisational Design: Challenges and Opportunities,’ Information and Organisation Design Series, 6, pp. 19–40.
Donaldson, L. (2001), The Contingency Theory of Organisations, Sage, Thousand Oaks.
Doran, D. (2004), ‘Rethinking the Supply Chain: An automotive perspective,’ Emerald, p. 9.
Drury, C. (2005), Management Accounting for Business. London, Thomson Learning.
Eliot, S., Hendry, L. & Stevenson, M. (2011), ‘Supply Chain Uncertainty: A Review and Theoretical Foundation for Future Research’, International Journal of Production Research, 50 (6), pp. 4493–4523.
Feitzinger, E. & Lee, H. L. (1997), ‘Mass Customisation at Hewlett-Packard: The power of postponement’, Harvard Business Review, 75, pp. 116–123.
Field, J. M., Ritzman, L. P., Safizadeh, M. H. & Downing, C. E. (2006), ‘Uncertainty Reduction Approaches, Uuncertainty Coping Approaches, and Process Performance in Financial Services. Decision Sciences, 37 (2), pp. 149–175.
Fisher, M. L. (1997), What is the Right Supply Chain for Your Product? Harvard business review, 75, pp. 105–117.
Flynn, B. B., Huo, B. & Zhao, X. (2010), The Impact of Supply Chain Integration on Performance: A contingency and configuration approach. Journal of Operations Management, 28, pp. 58–71.
Flynn, B. B., Huo, B. & Zhao, X. (2010), ‘The Impact of Supply Chain Integration on performance: A contingency and configuration approach’, Journal of Operations Management, 28 (1), 58–71.
Fogliatto, F., Silveira, G. & Borenstein, D. (2012), ‘The Mass Customisation Decade: An updated review of the literature,’ International Journal of Production Economics, 138, pp. 14–25.
Fornell, C. & Larcker, D. F. (1981), ‘Evaluating Structural Equation Models with Unobservable Variables and Measurement Error’, Journal of Marketing Research, vol. 18, no. 1.
Fujun, L., Min, Z., Denis, M. S. & Xiande, Z (2012), The Impact of Supply Chain Integration on Mass Customisation Capability: An Extended Resource-Based View. IEEE Transactions on Engineering Management, Vol. 59, NO. 3, August 2012, p. 443.
Ganeshan, R., Magazine, M. J. & Tayur, S. R. (1999), Quantitative Models for Supply Chain Management. Boston, Mass. [u.a.], Kluwer Acad.
Gattorna, J. (2009), ‘Dynamic Supply Chain Alignment: A new business model for peak performance in enterprise supply chains across all geographies, Farnham, England, Gower Pub.
Gershenson, J. K., Prasad, G. J. & Allamneni, S. (1999), ‘Modular Product Design: A Life-Cycle View, Journal of Integrated Design and Process Science, 3, p.. 3–26.
Ghasemi, A. & Zahediasl, S. (2012), ‘Normality Tests for Statistical Analysis: A guide for non-statisticians.’ International journal of endocrinology and metabolism, 10(2), p. 486.
Gimenez, C. & Ventura, E. (2005), ‘Logistics-Production, Logistics-Marketing and Internal
Integration’, International Journal of Operations and Production Management, Vol. 25 No. 1, pp. 20–38.
Gliner, J.. & Morgan, G. (2000), Research Methods in Applied Settings: An Integrated Approach to Design and Analysis. New Jersey, Lawrence Erlbaum Associates, Inc.
Govindan, K, Diabat, A & Panicker, V. (2011), ‘Supply Chain Risk Management and its Mitigation in a Food Industry’, International Journal of Production Research, 50 (1), pp. 3039–3050.
Graham, J. W. (2009)), ‘Missing Data Analysis: Making it work in the real world’. Annual review of psychology, 60, pp. 549–576.
Harbour, J. L. (2009), The Basics of Performance Measurement, Second Edition. New York, Taylor & Francis Group.
Hair, J. F., Black, W. C., Babin, B. J. & Anderson, R. E. (2010), Multivariate Data Analysis, 7th Edn. Upper Saddle River, New Jersey: Prentice Hall.
Hair, J. F., Ringle, C. M. & Sarstedt, M. (2011), ‘PLS-SEM: Indeed a silver bullet’. Journal of Marketing Theory and Practice, vol. 19, no. 2, pp. 139–151.
Hair, J. F., Ringle, C. M. & Sars, M. (2011a), PLS-SEM: Indeed a silver bullet. The Journal of Marketing Theory and Practice, 19(2), 139–152.
Henry, A. (2007), Understanding Strategic Management. Oxford, Oxford University Press.
Henseler, J. & Sarstedt, M. (2013), ‘Goodness-of-Fit Indices for Partial Least Squares Path Modelling’, Computational Statistics, vol. 28, no. 2, pp. 565–580.
Hillson, D. (2006), Integrated Risk Management as a Framework for Organisational Success. In: PMI Global Congress proceedings, 2006, Seattle, WA.
Hines, P. (1996), ‘Purchasing for Lean Production: The new strategic agenda’, International
Journal of Purchasing and Materials Management, Vol. 32 No. 1, pp. 2–12.
Hochberg, Y. & Tamhane, A. C. (1987), Multiple Comparison Procedures, John Wiley & Sons, New York.
Hoek, R. I. (1999), ‘Postponement and the Reconfiguration Challenge for Food Supply Chains. Supply Chain Management, 4 (1), pp. 18–24.
Howard, M., Miemczyk, J. & Graves, A. (2006), ‘Automotive Supplier Parks: An imperative
for build-to-order?’, Journal of Purchasing and Supply Management, Vol. 12 No. 2, pp. 91–104.
Huang, X., Kristal, M. & Schroeder, R. (2010), ‘The Impact of Organisational Structure on Mass Customisation Capability: A Contingency View’, Production and Operations Management, 19 (5), pp. 515–530.
Hulland, J. (1999), ‘Use of Partial Least Squares (PLS) in Strategic Management Research: A review of four recent studies’, Strategic Management Journal, vol. 20, no. 2, pp. 195–204.
Hult, G. T. M., Craighead, C. W. & Ketchen, D. J. (2010), ‘Risk Uncertainty and Supply Chain Decisions: A real options perspective’, Decision Sciences, 41 (3), 435–458.
Hult, G. T. M., Ketchen, D. J., & Arrfelt, M. (2007), Strategic Supply Chain Management: Improving performance through a culture of competitiveness and knowledge development, Strategic Management Journal, 28 (10), pp. 1035¬–1052.
Huo, B. (2012), The Impact of Supply Chain Integration on Company Performance: An organisational capability perspective. Supply Chain Management: An International Journal, 17 (6), pp. 596–610.
Iacobucci, D. (2008), Mediation Analysis, Sage, Los Angeles.
Jacobs, D. (2006), ‘The Promise of Demand Chain Management in Fashion’, Journal of Fashion Marketing and Management, 10 (1), pp. 84–96.
Kotcharin, S., Eldridge, S. & Freeman, J (n.d.), ‘Investigating the Relationships between Internal Integration and External Integration and their Impact on Combinative Competitive Capabilities, pp. 1–12.
Kahn, K. B. & Mentzer, J. T. (1996), ‘Logistics and Interdepartmental Integration’, International Journal of Physical Distribution & Logistics Management, 26 (8), 6–14.
Kazmi, S. H. H. (2007), Marketing Management: Text and cases. New Delhi, India, Excel Books.
Kew, J. & Stredwick, J. (2005), Business Environment: Managing in a strategic context. London, Chartered Institute of Personnel and Development.
Khosrowpour, M. (2006), Emerging Information Resources Management and Technologies. Hershey, Pa, Idea. http://site.ebrary.com/id/10158004
Koçog˘lu, I., Imamoglu, Z., Ince, H. & Keskin, H. (2011), ‘The Effect of Supply Chain Integration on Information-Sharing,’ Enhancing the supply chain performance, 24, pp. 1630–1649
Koçog˘lu, I., Imamoglu, S., Ince, H. & Keskin, H. (2011), The Effect of Supply Chain Integration on Information-Sharing: Enhancing the supply chain performance. Procedia Social and Behavioral Sciences, 24, pp. 1630–1649.
Kotcharin, S., Eldridge, S. & Freeman, J. (n.d.), ‘Investigating the Relationships between Internal Integration and External Integration and their Impact on Combinative Competitive Capabilities, pp. 1–12.
Kouvelis, P. & Su, P. (2007), The Structure of Global Supply Chains: The design and location of sourcing, production, and distribution facility networks for global markets. Boston, Now.
Krajewski, L. J., Ritzman, L. P. & Malhotra, M. K. (2007), Operations Management: Processes and value chains.
Krajewski, L., Wei, J. & Tang, L. L. (2005), ‘Responding to Schedule Changes in Build-to-Order Supply Chains’, Journal of Operations Management, Vol. 23, No. 5, pp. 452–469.
Kumar, S. & Krob, W. W. (2005), Managing Product Life Cycle in a Supply Chain Context A Prescription Based on Empirical Research. Boston, MA, Springer Science+Business Media, Inc.
Lai, F. Zhang, M., Lee, M. S. & Zhao, X. (2012), ‘The Impact of Supply Chain Integration on Mass Customisation Capability: An Extended Resource-Based View’, IEEE Transactions On Engineering Management, 59 (3), pp. 443–456.
Lambert, D. M. & Cooper, M. C. (2000), Issues in Supply Chain Management, Industrial marketing management, 29 (1), pp. 65–83.
Lambert, D. M. Stock, J. R. & Ellram, L. M. (1998), Fundamentals of Logistics Management, Irwin/McGraw-Hill, Chicago, IL.
Larson-Hall, J. (2009), A Guide to Doing Statistics in Second Language Research Using SPSS. Routledge.
Lau, H. C. W. & Lee, W. B. (2000), On a Responsive Supply Chain Information System, International Journal of Physical Distribution & Logistics Management, 30 (7/8), pp. 598–610.
Lee, C. W., Kwon, I. W. G. & Severance, D. (2007), Relationship between Supply Chain Performance and Degree of Linkage amongst Supplier, Internal Integration, and Customer. Supply chain management: an International journal, 12 (6), pp. 444–452.
Lee, H., Padmananabhan, V. & Whang, S. (2007), ‘The Bullwhip Effect in Supply Chains’, Sloan Management Review, 38 (3), pp. 93–102.
Lee, H. L. & Whang, S. (2005), ‘Higher Supply Chain Security with Lower Cost: Lessons from total quality management’, International Journal of production economics, 96 (3), pp. 289–300.
Lehmann, D. R. (1979), Market Research and Analysis, Irwin, Homewood, IL.
Leys, C., Ley, C., Klein, O., Bernard, P. & Licata, L. (2013), ‘Detecting Outliers: Do not use standard deviation around the mean, use absolute deviation around the median’. Journal of Experimental Social Psychology, 49(4), pp. 764–766.
Li, J. & Hong, S.-J. (2007), ‘Towards a New Model of Supply Chain Risk Management: The cross-functional process mapping approach’, International Journal of Electronic Customer Relationship Management, 1 (1), pp. 91–107.
Li, L. & Schulze, L. (2011), ‘Uncertainty in Logistics Network Design: A review’, Proceedings of the International MultiConference of Engineers and Computer Scientists. II
Li, L. (2008), Supply Chain Management: Concepts, techniques and practices: enhancing value through collaboration. Singapore [u.a.], World Scientific.
Li, S.. Rao, S., Ragu-Nathan, T. & Ragu-Nathan, B. (2005), Development and Validation of a Measurement Instrument for Studying Supply Chain Management Practices. Journal of Operations Management, 23, pp. 618–643.
Liang, H., Saraf, N., Hu, Q. & Xue, Y. (2007), ‘Assimilation of Enterprise Systems: The Effect of Institutional Pressures and the Mediating Role of Top Management’, MIS Quarterly, vol. 31, no. 1, pp. 59–87.
Liu, G. J., Shah, R. & Schroeder, R, G. (2010), ‘Managing Demand and Supply Uncertainties to Achieve Mass Customisation Ability’, Journal of Manufacturing Technology Management, 21 (8), pp. 990–1012.
Liu, G., Shah, R. & Schroeder, R. (2012), ‘The Relationships amongst Functional Integration, Mass Customisation and Firm Performance’, International Journal of Production Research, 50 (3), pp. 677–690.
Long, D. (2012), The Hajj and its Impact on Saudi Arabia and the Muslim World. Retrieved from <http://www.susris.co m/2012/10/25/the-hajj-and-its-impact-o n-saudi-arabia-and-the-muslim-world-david-lo ng-3/>
Luhmann, N. (2005), Risk: A sociological theory, Aldine de Gruyter.
Malhotra, N. K. (1996), Marketing Research: An applied orientation, 2nd edition, Englewood Cliffs, Prentice-Hall, New Jersey.
Markman, G. & Phan, P. H. (2011), The Competitive Dynamics of Entrepreneurial Market Entry. Cheltenham, Edward Elgar Pub.
Mason-Jones, R. & Towill, D. R. (1998), Shrinking the Supply Chain Uncertainty Circle, IOM Control, 24 (7), pp. 17–22.
Mentzer, J. T. (2001), Supply Chain Management. Thousand Oaks, Calif. [u.a.], Sage Publ.
Mentzer, J. T., Min, S. & Zacharia, Z. G. (2000), The Nature of Inter-Firm Partnering in Supply Chain Management, Journal of Retailing, 76 (4), 549–568.
Meydanoglu, E. (2009), ‘The Role of Supply Chain Event Management Systems for Supply Chain Risk Management’, European and Mediterranean Conference on Information Systems.
Mikkola, J. H. & Larsen, T. S. (2004), ‘Supply-chain Integration: Implications for mass customisation, modularization and postponement strategies, Production Planning & Control,’ The Management of Operations, 15(4), pp. 352–361.
Miller, J. G. & Roth, A. V. (1994), ‘A Taxonomy of Manufacturing Strategies’, Management
Science, Vol. 40 No. 3, pp. 285–304.
Mills, J, Schmitz, J. & Frizelle, G. (2004), ‘A Strategic Review of ‘Supply Networks’, International Journal of Operations & Production Management, 24 (10), pp. 1012–1036.
Min, S. & Mentzer, J. T. (2004), ‘Developing and Measuring Supply Chain Management Concepts, Journal of Business Logistics, 25 (1), pp. 63–99.
Moser, K. (2006), Mass Customisation Strategies: Development of a competence-based framework for identifying different mass customisation strategies.
Mourdoukoutas, P. (1999), Collective Entrepreneurship in a Globalising Economy. Westport, Conn. [u.a.], Quorum Books.
Narasimhan, R. & Das, J. (2001), ‘The Impact of Purchasing Integration and Practices on Manufacturing Performance’, Journal of Operations Management, Vol. 19 No. 5, pp. 593–609.
Narasimhan, R., Jayaram, J. & Carter, J. R. (2001), ‘An Empirical Examination of the Underlying Dimensions of Purchasing Competence’, Production and Operations Management, Vol. 10 No. 1, pp. 1–15.
Narasimhan, R., Swink, M. & Kim, S. W. (2006), ‘Disentangling Leanness and Agility: An empirical investigation’, Journal of Operations Management, Vol. 24 No. 5, pp. 440–457.
Natour, A., Kiridena, S. & Gibson, P. (2011), ‘Supply Chain Integration and Collaboration for Performance Improvement: An agency theory approach,’ In A. Alnatour, S. Kiridena & P. Gibson (Eds.), 9th ANZAM Operations, Supply Chain and Services Management Symposium (pp. 503–519), Geelong: Deakin University.
Neider, L. L. & Schriesheim, C. (2007), International perspectives. Charlotte, NC, Information Age Pub.
Neiger, D., Rotaru, K. & Churilov, L. (2009), ‘Supply Chain Risk Identification with Value-Focused Process Engineering’, Journal of Operations Management, 27 (2), pp. 154–168.
Nyaga, G. N., Whipple, J. M. & Lynch, D. F. (2010), ‘Examining Supply Chain Relationships: Do buyer and supplier perspectives on collaborative relationships differ?’ Journal of Operations Management, 28 (2), pp. 101–114.
Olhager, J. (2003, ‘Strategic Positioning of the Order Penetration Point’, International Journal of Production Economics, 85, pp. 319–329.
Olivella, J., Cuatrecasas, L. & Gavilan, N. (2008), ‘Work Organisation Practices for Lean Production’, Journal of Manufacturing Technology Management, 19 (7), pp. 798–811.
Otsuki, M. (2002), SMEs Supporting Systems in Saudi Arabia. October, pp. 1–9.
Pallant, J. (2013), SPSS Survival Manual. McGraw-Hill International.
Peck, H. (2006), ‘Reconciling Supply Chain Vulnerability, Risk and Supply Chain Management, International Journal of Logistics: Research and Applications, 9 (2), pp.127–142.
Pegels, C. C. (2005), Proven Solutions for Improving Supply Chain Performance. Greenwich, Conn, IAP – Information Age Pub.
Pertusa-Ortega, E. M., Molina-Azorín, J. F. M., Claver-Cortés, E. (2010), ‘Competitive Strategy, Structure and Firm Performance: A comparison of the resource-based view and the contingency approach’, Management Decision, 48 (8), pp. 1282–1303.
Plaggenhoef, W. V. (2007), Integration and Self Regulation of Quality Management in Dutch Agri-food Supply Chains: A cross-chain analysis of the poultry meat, the fruit and vegetable and the flower and potted plant chains. [Wageningen], [publisher not identified].
Podsakoff, P. M., MacKenzie, S. B., Lee, J.-Y. & Podsakoff, N. P. (2003), ‘Common Method Biases in Behavioural Research: A critical review of the literature and recommended remedies’, Journal of Applied Psychology, vol. 88, no. 5, pp. 879–903.
Ponce, E. & Prida, B. (2004), La Logística de Aprovisionamientos Para la integración de la
cadena de suministros, Pearson Educación, Madrid.
Prater, E., Biehl, M. & Smith, M. A. (2001), International Supply Chain Agility—Tradeoffs between flexibility and uncertainty, International Journal of Operations & Production Management, 21 (5/6), pp. 823–839.
Prasad, S. & Babbar, S. (2000), ‘International Operation Management Research’, Journal of
Operations Management, Vol. 18 No. 2, pp. 209–247.
Qi, Y., Zhao, X. & Sheu, C. (2011), The Impact of Competitive Strategy and Supply Chain Strategy on Business Performance: The Role of Environmental Uncertainty. Decision Sciences, 42 (2), pp. 371–389.
Ramanathan, U. (2013), ‘Aligning Supply Chain Collaboration using Analytic Hierarchy Process’, Omega-International Journal of Management Science, 41 (2), pp. 431–440.
Ringle, C. M., Wende, S. & Will, A. (2010), Finite Mixture Partial Least Squares Analysis: Methodology and numerical examples. In V. E. Viniz, W. W. Chin, J. pp. 39–50.
Ringle, C. M., Wende, S. & Will, A. 2005, SmartPLS 2.0. Hamburg [Online]. Available: www.smartpls.de [2014, Dec].
Ritchie, B. & Brindley, C. (2007), Supply Chain Risk Management and Performance: A guiding framework for future development, International Journal of Operations & Production Management, 27 (3), pp. 303–322.
Rudberg, M. & Wikner, J. (2004), ‘Mass Customisation in Terms of the Customer Order Decoupling Point’, Production Planning & Control, 15 (4), pp. 445–458.
Rungtusanatham, M., Salvador, F., Forza, C. & Choi, T. Y. (2003), Supply-Chain Linkages and Operational Performance: A resource-based-view perspective, International Journal of Operations & Production Management, 23 (9), pp. 1084–1099.
Rushton, A. & Walker, S. (2007), International Logistics and Supply Chain Outsourcing from Local to Global. London, Kogan Page.
Saeed, K. A., Malhotra, M. K. & Grover, V. (2005), ‘Examining the Impact of Interorganisational Systems on Process Efficiency and Sourcing Leverage in Buyer–Supplier Dyads’, Decision Sciences, Vol. 36 No. 3, pp. 365–396.
Salvador, F., Rungtusanatham, M. & Forza, C. (2004), ‘Supply-Chain Configurations for Mass Customisation’, Production Planning & Control, 15 (4), pp. 381–397.
Sanchez, R. (2008), A Focused Issue on Fundamental Issues in Competence Theory Development. Bingley, UK, Emerald/JAI.
Sanders, N. & Premus, R. (2005), ‘Modelling the Relationship between Firm IT Capability, Collaboration and Performance’, Journal of Business Logistics, Vol. 26 No. 1, pp. 1–23.
Sapsford, R. & Jupp, V. (2006), Data Collection and Analysis. London ; Thousand Oaks, Calif, Sage publications.
Sawhney, R. (2006), ‘Interplay between Uncertainty and Flexibility Across the Value Chain: Towards a transformation model of manufacturing flexibility’, Journal of Operations Management, Vol. 24 No. 5, pp. 476–493.
Schermerhorn, J. R. (2010), Management. Hoboken, N.J., Wiley.
Shalaby, N. (2012), SMEs’ Capabilities and Needs Assessment Eastern Province, Saudi Arabia. Eastern Province Chamber of Commerce and Industry.
Simangunsong, E., Hendry, L. C. & Stevenson, M. (2011), ‘Supply-Chain Uncertainty: A review and theoretical foundation for future research’, International Journal of Production Research, 50 (16), 4493–4523.
Simangunsong, E., Hendry, L. & Stevenson, M. (2012), Supply-Chain Uncertainty: A review and theoretical foundation for future research. International Journal of Production Research, 50 (16), pp. 4493–4523.
Skipworth, H. & Harrison, A. (2007), ‘Implications of Form Postponement to Manufacturing a Customised Product’, International Journal of Production Research, 44 (8), pp. 1627–1652.
Squire, B., Brown, S. & Cousins, P. D. (2003), Collaborating for Customisation: A resource-based perspective. Proceedings of the MCPC.
Stank, T. P., Keller, S. B. & Daugherty, P. J. (2001), ‘Supply Chain Collaboration and Logistical Service Performance’, Journal of Business Logistics, Vol. 22 No. 1, pp. 29–48.
Stock, J. R. & Lambert, D. M. (2001), Strategic Logistics Management.
Stratton, R. & Warburton, R. D. H. (2003), ‘The Strategic Integration of Agile and Lean Supply’, International Journal of Production Economics, Vol. 85 No. 2, pp. 183–198.
Swafford, P., Ghosh, S. & Murthy, N. (2006), ‘The Antecedents of Supply Chain Agility of a Firm: Scale development and model testing’, Journal of Operations Management, Vol. 24 No. 2, pp. 170–188.
Tabachnick, B. & Fidell, L. (2001), Using Multivariate Statistics, 4th Edn. Boston: Allyn and Bacon.
Tenenhaus, M., Amato, S. & Vinzi, V. (2004). A Global Goodness-of-Fit Index for PLS Structural Equation Modelling. Proceedings of the XLII SIS Scientific meeting (pp. 739–742). Padova: CLEUP.
Tomlin, B. (2006), ‘On the Value of Mitigation and Contingency Strategies for Managing Supply Chain Disruption Risks’, Management Science, 52 (5), pp. 639–657.
Towill, D. & Christopher, M. (2002), ‘The Supply Chain Strategy Conundrum: To be lean or agile or to be lean and agile’? International Journal of Logistics, 5 (3), pp. 299–309.
Ulrich, K. (1995), ‘The Role of Product Architecture in the Manufacturing Firm’, Research policy, 24 (3), pp. 419–440.
Van de Venn, A. H. & Ferry, D. I. (1980), Measuring and Assessing Organisations, John Wiley and Sons, New York, NY.
Van der Vaart, T. & Van Donk, D. P. (2004), ‘Buyer Focus: Evaluation of a new concept for
Supply Chain Integration’. International Journal of Production Economics, Vol. 92 No. 1, pp. 21–30.
Van Hoek, R. I. (2001), The Rediscovery of Postponement: A literature review and directions for research, Journal of Operations Management, 19 (2), pp. 161–184.
Van Hoek, R. I. (1999), ‘Postponement and the Reconfiguration Challenge for Food Supply Chains,’ An International Journal, 4 (1), pp. 18–34.
Van Hoek, R. I., Harrison, A. & Christopher, M. (2001), Measuring Agile Capabilities in the Supply Chain, International Journal of Operations & Production Management, 21 (1/2), pp. 126–148.
Van Hoek, R. I. & Weken, H. A. (1998), ‘The Impact of Modular Production on the Dynamics of Supply Chains’, International Journal of Logistics Management, The, 9 (2), pp. 35–50.
Van Weele, A. J. (2005), Purchasing & Supply Chain Management: Analysis, strategy, planning and practice. Hampshire, Cengage.
Vickery, S., Calantone, R. & Dröge, C. (1999), ‘Supply Chain Flexibility: An empirical study’, The Journal of Supply Chain Management, Vol. 35 No. 1, pp. 16–24.
Vickery, S., Jayaram, J., Dröge, C. & Calantone, R. (2003), ‘The Effects of an Integrative Supply Chain Strategy on Customer Service and Financial Performance: An analysis of direct versus indirect relationships’, Journal of Operations Management, Vol. 21 No. 5, pp. 523–539.
Vijayasarathy, L. R. (2010), ‘Supply Integration: An investigation of its multi-dimensionality and relational antecedents’, International Journal of Production Economics, 124: pp. 489–505.
Wagner, S. M. & Bode, C. (2009), Managing Risk and Security: The safeguard of long-term success for logistics service providers. Berne: Haupt.
Wang, W. Y. C., Heng, M. S. H. & Chau, P. Y. K. (2007), Supply Chain Management Issues in the New Era of Collaboration and Competition. Norwood Mass, Books24x7.com.
Wernerfelt, B. (1984), ‘A Resource-Based View of the Firm’, Strategic Management Journal, 5 (2), pp. 171–180.
Westbrook, R. & Williamson, P. (1993), ‘Mass Customisation: Japan’s new frontier’, European Management Journal, 11 (1), pp. 38–45.
Wisner, J. D., Tan, K.-C. & Leong, G. K. (2011), Principles of Supply Chain Management: A balanced approach. Mason, OH, South-Western.
Wolf, J. (2008), The Nature of Supply Chain Management Research Insights from a Content Analysis of International Supply Chain Management Literature from 1990 to 2006. Wiesbaden, Gabler.
Womack, J. P., Jones, D. T. & Roos, D. (1990), The Machine that Changed the World, Macmillan Publishing Company, New York, NY.
Wong, C. Y., Arlbjørn, J. S. & Johansen, J. (2005), ‘Supply Chain Management Practices in Toy Supply Chains’, Supply Chain Management: an International Journal, 10 (5), pp. 367–378.
Yang, B. & Yang, Y. (2009), ‘Postponement in Supply Chain Risk Management: A complexity perspective,’ International Journal of Production Research, 48 (7), pp. 1901–1912.
Yang, B. & Burns, N. (2003), ‘Implications of Postponement for the Supply Chain’, International Journal of Production Research, 41 (9), pp. 2075–2090.
Yao, D. D. & Song, J.-S. (2001), Supply Chain Structures: Coordination, information and optimization. London, Kluwer Academic Publishers.
Yeung, J., Selen, W. & Zhang, Z. (2007), ‘Postponement Strategy from a Supply Chain Perspective: Cases from China,’ International Journal of Physical Distribution & Logistics Management, 37 (4), pp. 331–356.
Zhang, X., Song, H. & Huang, G. Q. (2009), Tourism Supply Chain Management: A new research agenda, Tourism management, 30 (3), pp. 345–358.
Zhao, X., Huo, B., Flynn, B. B. & Yeung, J. (2008), ‘The Impact of Power and Relationship Commitment on the Integration between Manufacturers and Customers in a Supply Chain,’ Journal of Operations Management, 26 (3), pp. 368–388.
Zikmund, W. G. (2013), Business Research Methods. Mason, OH, South-Western.

APPENDICES
Appendix I: Hypothesis

The hypotheses are all tested how the demand uncertainty can be mitigated in high season of customer demand, by using managerial strategies.
H1: Customer integration positively associated with postponement.
H2: Customer integration positively associated with mass customisation capability positively.
H3: Internal Integration positively associated with postponement.
H4: Internal integration positively associated with mass customisation capability positively.
H5: Supplier Integration positively associated with postponement positively.
H6: Supplier integration positively associated with mass customisation capability positively. Not supported
H7: Postponement has a direct relationship with mass customisation capability positively.
H8: Postponement mitigates demand uncertainty positively.
H9: Mass customisation capability mitigates demand uncertainty positively.
H10: Internal integration has a positive indirect effect on postponement practice through customer integration.
H11: Internal integration has a positive indirect effect on postponement practice through supplier integration.
H12: Internal integration has a positive indirect effect on mass customisation capability through customer integration.
H13: Internal integration has positive indirect effect mass customisation capability through supplier integration.
H14: postponement practice has positive indirect effect on demand uncertainty mitigation through Mass Customisation Capability.
H15: Competitive Intensity enhances the indirect effect of internal integration on postponement practice through supplier integration.
H16: Competitive Intensity enhances the indirect effect of internal integration on postponement practice through customer integration. Not supported
H17: Competitive Intensity enhances the indirect effect of internal integration on mass customisation capability through supplier integration.
H18: Competitive Intensity enhances the indirect effect of internal integration on mass customisation capability through customer integration.
H19: Competitive Intensity enhances the indirect effect of internal integration on demand uncertainty mitigation through Mass Customisation Capability
Appendix II: Measurement Items

Mass Customisation Capability
MCC1: Competitive Intensity enhances the indirect effect of internal integration on mass customisation capability through supplier integration.
MCC2: Competitive Intensity enhances the indirect effect of internal integration on mass customisation capability through customer integration.
MCC3: Competitive Intensity enhances the indirect effect of internal integration on demand uncertainty mitigation through mass customisation capability.

Postponement Practice
PP1: Our firm postpones final product assembly activities until receives customer orders.
PP2: Our firm postpones final product labelling activities until receives customer orders.
PP3: Our firm postpones final packaging activities until receives customer orders.

Demand Uncertainty Mitigation
DUM1: We mitigate demand uncertainty by providing products to our customer consistent with their nominated product specification.
DUM2: We mitigate demand uncertainty when our customers place orders consistent with their nominated delivery lead time.
DUM3: We mitigate demand uncertainty when our customers provide us reliable forecasts on their demands.

Competitive Intensity
CPI1: We are in a highly competitive industry.
CPI2: Our competitive pressures are extremely high.
CPI3: We do not pay much attention to our competitors.
TO GET YOUR ASSIGNMENTS DONE AT A CHEAPER PRICE,PLACE YOUR ORDER WITH US NOW