Supply chain complexity drivers

Framework for discovering supply chain complexity drivers

Supply chain complexity (SCC) arises when supply chains grow large in number of connection points, and the connectivity between these points is complicated. To ensure future operation, companies must eliminate non-value-adding complexities and streamline the business.

Framework for discovering supply chain complexity drivers

Written by Said Afandi from Implement Consulting Group and Zaza Nadja Lee Herbert-Hansen, Technical University of Denmark, DTU.

Based on a literature review and industry observations, this paper provides a thorough clarification of what is meant by SCC and interdependencies between complexity drivers. The result, the supply chain complexity canvas, is a framework that facilitates description and discussion between supply chain stakeholders. The framework functions as a dynamic platform that simplifies the identification of SCC drivers and the links between such.

Introduction

The network of business entities that is involved in an organisation's up- and downstream activities is what we define as a company’s supply chain (SC). The systematic and strategic co-ordination of this network has gained more attention during the last decades (1).

We refer to SC co-ordination as supply chain management (SCM). Historically, organisations found that managing their supply chain enabled an overall cost reduction, service level improvements, better quality insurance and customer satisfaction; hence, providing the organisation with a competitive advantage (2,3).

Companies attain wide distribution ranges and lowest price as they leverage global manufacturing and distribution capabilities. However, a side effect of this is increased complexity (1,3-9). Complexity is defined as “a condition of interconnectedness and interdependency across a network” (10).

If this is true, changes in one element can affect other elements, for better and worse, in unforeseen ways. As complexity itself is not bad, managers must ensure that they distinguish between good (value-adding) and bad (non-value-adding) complexity (11). They must accept that an SC will always be influenced by a certain degree of complexity (3,7,8) and thereby the associated uncertainty and costs (9). Christopher (8) argues that the organisation differentiate itself from competitors through complexity, i.e. if what they do is straightforward, it would be too easy for others to replicate.

Complexity management is an essential part of SCM. If left unchecked, complexities are potential cause of inefficient operations and higher costs. The organisation should manage this threat by first identifying complexity drivers and dynamics in order to understand the interconnectivity and consequences of these. Then, distinguish between value-adding and non-value-adding complexity (3,4,11). And lastly, develop action plans to reduce or remove non value-adding complexity. A literature review revealed that there are several methods to manage complexity, but no model or framework that takes into account the interconnectivities and consequences of such, even though several articles acknowledge the significance of this (2,3,10-13).

This paper addresses the research gap in relation to interdependencies of complexity drivers. The objective is to develop a framework that enables managers and researchers to identify relevant complexity drivers and their interdependencies within a company’s supply chain. Hence, the following two research questions have been formulated:

  1. How does literature define supply chain complexity (SCC)? 
  2. How can managers identify main drivers of SCC and their interdependencies?

Methodology

This research seeks to build a normative standard of how to identify complexity drivers. Due to the complex nature of the research aim, a qualitative approach is used to provide rich and in-depth data. The explorative nature of the study allows for in-depth understanding of the research area. Therefore, the case study approach is the most appropriate research methodology (14,15).

In total six different companies have been investigated, and data have been gathered through observations and semi-structured interviews. The chosen companies were selected based on several criteria, including that a) they were large organisations with extensive SCs, b) they had an international reach, c) they were struggling with both valueadding and non-value-adding complexity in their supply chain, and they came from a variety of industries. The interviewees were selected based on their experience with complexity management projects.

The research has five stages: (i) an extensive literature review of research papers and industry publications, (ii) semi-structured interviews and workshops in the case companies, (iii) structured comparison of the literature review and the empirical data, (iv) introduction of the system dynamics of supply chain complexity and (v) development of a novel framework that assists organisational leaders in identifying drivers of SCCs and interdependencies between such.

Literature review

Wilson and Perumal (11) introduces three sources of complexity, namely product, process and organisation, and states that the associated costs are located in the interaction between sources, e.g. product/process (11).

This is fully aligned with Ashkenas’ (2) findings that pinpoint the same three sources, but adds an extra dimension: managerial habits. Serdarasan (3) further categorises complexities into static, dynamic (16) and decision-making.

Serdarasan’s “static” complexity covers three of Ashkenas’ complexity types, specifically organisation, product/ service and processes. Ashkenas’ latter, the managerial habits, is equal to Serdarasan’s definition of decision-making. Lastly, Serdarasan then adds a third complexity type, dynamic complexity, as the “operational behaviour of the system and its [external] environment” (3) introducing the uncertainty of randomness and time (16). Table 1 below shows the complexity types derived from literature.

Table 1 – Overview of complexity types and their meaning in this paper

Table 1 – Overview of complexity types and their meaning in this paper

For the remainder of this paper, we refer to static, dynamic and decision-making based on the definitions in table. Table 2 presents a content overview of the most influential articles in the literature review. It illustrates the environments in scope (internal, external), whether the articles consider the supply chain as static or dynamic, and lastly whether interconnectivity is discussed. As illustrated, several articles specify the relevance of interconnectivity and that the revision of complexity is an iterate process. Still no model or framework incorporates the interconnectivity or associated feedback loops.

The increased complexity found in the organisation’s supply chain is not only costly. It also adds to a lower level of service, inefficient processes and bad work conditions as employees of the system need to interact with an increased amount of internal and external factors. The potential drivers of complexity are many and include variety, connectivity, opacity and dynamics as well as uncertainty (3,4,12,17-19).

Practitioners cannot easily pinpoint drivers of complexity as they are located in every stage of the supply stream and are influenced by the “parallel interactions” (19). Bozarth et al. (12) deliver an extended overview of possible complexity drivers at different stages of the supply chain and few interrelations across SC stages. Due to dynamic complexity such as system dynamic feedback loops, a complexity and its related cost can have a complexity relieving or enhancing effect further down the supply chain. For this reason, we must define the sum of a complexity cost as the sum of all interconnected costs. Full understanding of the entire supply chain, its interdependencies and complexities is therefore essential to minimise total complexity cost and exploit the SC’s full potential.

Table 2 – Review of the most influential articles’ coverage of supply chain complexity

Table 2 – Review of the most influential articles’ coverage of supply chain complexity

Kearney (13) argues that managers who optimise their supply chain design by reducing inventory, outsourcing, trimming the number of products and sourcing globally rarely gain the full potential. They neglect the dynamic complexity and optimise based on the assumption that the supply chain operates in a stable environment. Several recent articles present ways for designing resilient supply chains, but agree that no design fits all (20). Managers must therefore pick and choose in order to tailor the network design that fits their respective organisation. This approach often results in an iterative process (18).

The reviewed models provide different approaches to manage and/or reduce complexity. The levers must, though, be chosen wisely to fit the type of complexity that we seek to manage or reduce.

Serdarasan (3) distinguishes between external, supply/demand (S/D) interface and internal complexity drivers. As a company has little, if any, control over external complexity drivers, focus should be directed at internal and S/D interface drivers, those that can be managed or removed. Static (structural) complexity drivers can be reduced, whereas dynamic drivers may be more resistant. The aim is then to manage or adjust in order to cope with the dynamic complexity (3). Decisions directed towards dynamic complexity drivers may have positive or negative effects on various drivers in the system, as companies risk shifting the complexity from one driver to another, preferable to one which can be managed. Decision-making complexity drivers should be managed by centralising and automating decision-making (3).

Datta et. al. (20) illustrate that centralising ensures alignment across the business units, but can quickly become a burden, as gathering information can be costly and the responsiveness declines drastically. The organisation must therefore find the appropriate equilibrium where it decentralises decision-making on the basis of centralised guidelines in order to keep flexibility and quick responsiveness, while still obtaining alignment and benefits of a centralised structure. Perona and Miragliotta (21) argue that no model has been able to explain the relationships between all the relevant variables that a company must address in order to reduce or manage complexity in the supply network. The course of action will vary dependent on the situation and driver in question.

Traditional mathematical modelling approaches have showed ineffective in dealing with dynamic, complex supply chains (20) and the interaction between the up- and downstream needs of the system. Perona and Miragliotta (21) give an example of a conceptual framework. A framework with several limitations, but as argued earlier in this paper, the wide extent of complexity management and the broad variance in different cases make it nearly impossible to create a formula that suits all purposes. The proposed framework can help as a guide to understand the origin of SCCs and how these affect each other.

Winkler et al. (17) present the complexity strategy matrix that proposes one of four strategies to handle supply chain complexity based on effort and effect. The four strategies are: accepting, controlling, reducing and avoiding (17). Concrete guides on how to control, reduce or avoid complexities are not mentioned.

Today’s literature on SCC is limited. The literature review shows a broad consensus that no model fits all purposes. The way to tackle complexity in the supply chain heavily depends on the type of complexity, and where in the SC it is located.

Empirical findings

This section seeks to explore which of the theoretical drivers that is most common across a selection of companies from six different industries. Table 3 provides this overview. It illustrates which of the different drivers that have been observed at the case companies and in which industry the companies operate. Three complexity drivers are found across all six companies in six different industries and relate to uncertainty, organisational processes and IT/ERP systems. Complexity drivers related to suppliers and distribution network are located in four of the six companies. Product portfolio and product modularisation complexities are found in twothirds of the companies, indicating that most complexities originate across industries.

Table 3 – The types of complexity that case companies experience

Table 3 – The types of complexity that case companies experience

Table 4 shows three concrete examples of complexity-related difficulties that the organisations faced.

  • Four companies were struggling with stock-keeping of essential spare parts, i.e. procurement and supplier management, warehouse management and logistics.
  • Four companies struggled with product portfolio management in order to serve the increasing customer demand in a cost-efficient way.
  • All companies faced challenges managing the cost to serve customers.

Five companies had difficulties balancing service levels and costs. From the case observations, it became evident that, to some extent, all sectors are affected by complexity.

From the case observations, it became evident that, to some extent, all sectors are affected by complexity.

Table 4 – Concrete examples of complexity-related problems in the case organisations

Table 4 – Concrete examples of complexity-related problems in the case organisations

Discussion

By organising complexity drivers by the categories of table 2, an overview is created of the drivers’ proportions and position in the SC. Table 5 illustrates that the complexity drivers most often are internal drivers that are interrelated. In other words, several internal (or external) drivers affect each other in positive and/or negative ways. The table displays that complexity is found to be both static and dynamic. Practitioners need ways to manage these complexities with concern paid to the interrelatedness of the drivers. The four dynamic drivers from table 5 should be carefully addressed as they are all influenced by uncertainty of randomness and time. Consequently, the practitioner will not be in full control of the effects when addressing the complexity drivers and must therefore carefully manage the effects caused by the interrelatedness.

Based on the empirical evidence, complexity is found in all sectors, and many drivers impact more than one sector. The demand for a method or framework to target complexity is present in all industries, and because several drivers affect many industries, it points towards the possibility of constructing a generic framework that enables managers to identify SCC drivers.

Table 5 – Structured overview of the complexity drivers in the case companies

Table 5 – Structured overview of the complexity drivers in the case companies

Interrelation of SCC drivers

Combining the empirical evidence from the case companies with the results of the literature review enables us to structure a system dynamic model that describes the interrelations and interconnectivity of the complexity drivers. The model is built in Vensim PLE and illustrates how complexity drivers in different stages of the supply chain influence each other positively and negatively.

The model illustration below clarifies how the different SCC drivers influence the three business areas of the supply chain, i.e. up- and downstream activities and the internal processes of the company. The business areas used in the conceptual model (see figure 3) are marked with green text.

The arrows of figure 1 below illustrate how the different activities and drivers are interrelated. These relations create interdependencies and thereby feedback loops. This means that the output of one entity of the system is routed back as an input for the exact same entity, hereby creating a course-and-effect loop, known as a feedback loop. Figure 1 contains a total of seven feedback loops.

FIGURE 1 – Vensim PLE model that illustrates the interrelations of the SCC drivers

FIGURE 1 – Vensim PLE model that illustrates the interrelations of the SCC drivers

For illustrative purposes, figure 2 (page 8) exemplifies how procurement, inventory level and production plan are interrelated in feedback loops. Procurement/inventory and inventory/ production planning must be balanced at the same time as the company needs to ensure that the loop procurement - inventory - production plan supports the overall strategy of the company.

Illustration of the feedback loops

Illustration of the feedback loops

A conceptual framework for discovering SCC drivers

To accommodate all the areas in which SCC arises, the organisation should be addressed as a static entity as this eases identification of complexity drivers and in drawing links between the drivers to create a dynamic system later on. A dynamic model is ultimately required as it stresses how complexities affect each other across the organisation and the supply chain. The model is structured in such a way that it facilitates description and discussion between stakeholders of the supply chain. It is highly recommended to include external partners of the supply chain to create a shared language and set the direction across the supply chain.

The framework functions as a dynamic platform that practitioners can use to model their SCC drivers and the links between such. The model is illustrated in figure 3 and consists of 10 building blocks that cover three business areas: downstream, internal and upstream activities. In addition, it is possible to integrate partnerships with both down- and upstream partners.

  • Supplier capabilities: The processes or features that make the supplier involvement necessary to serve the customers.
  • Upstream logistics: Represents the logistics network from suppliers to the company.
  • Organisational structure and governance: Denotes the administrative and work-related processes and governance models the organisations follow.
  • Purchasing: The purchasing governance process.
  • Sales: The sales governance process.
  • IT systems: IT systems’ ability to promote or obstruct the company in fulfilling its value proposition.
  • Operation: The operation process’ ability to promote efficient delivery of the value proposition.
  • R&D: The innovative efforts that drive higher complexities.
  • Customer demands: The demands from the customers that the company seeks to fulfil via their value proposition.
  • Downstream logistics: Represents the logistics network from company to customer and all its challenges.
  • Partnerships: Partnerships made with external associates, e.g. suppliers, NGOs, 3PL, innovation labs and outsourcing partners.

FIGURE 3 – The supply chain complexity canvas

FIGURE 3 – The supply chain complexity canvas

Practitioners need to start by formulating their value proposition and writing it at the top of the model. The value proposition represents the value factor that the customers want to buy. This step is essential in order to distinguish between value-adding and non-value-adding complexities throughout the session.

Discussion

The model reaches its appearance and capabilities based on the empirical research, i.e. literature, case studies and input from complexity management experts in the industry. The framework intends to build the foundation of a company-wide complexity management process. A process that brings together qualified managers from all business units into one or more workshops facilitated around the canvas. The managers should have extensive knowledge of their own business area and its connectivity to internal and external business units.

The extensive stakeholder involvement produces valuable insight and data input, but it also demands time and effort from the stakeholders and their team. Without knowing the possible gain from a complexity relieving project, it is difficult to motivate stakeholders to fully prioritise the project, possibly on the expense of daily operational tasks. The issue is partly managed as the model enables the organisation to focus on either a part of the canvas or the full SC. An organisation might find more value in focusing on the internal and downstream activities at first, then upstream and internal activities before targeting the whole supply chain. This allows an organisation to focus their efforts according to the resources they are able to invest in the complexity relieving project.

Conclusion

Based on present literature, this paper highlights the contemporary understanding of the term SCC. This is found to cover three types of complexity: static, dynamic and decision-making. All of which are found in the internal and external environment. More importantly, research point to the importance of the interconnectivity of the complexity drivers. This means that choosing to relieve or manage one complexity driver within a company’s SC may have either positive or negative consequences on other areas of the SC. Contemporary literature is tested on six companies representing six different industries in order to acquire further insights into the complexity drivers that companies face.

We learnt that all industries are affected by complexities and that several drivers are found across more than one industry. Understanding the theory and practice enables us to design a framework that facilitates practitioners’ effort in understanding SCC. This paper introduces the supply chain complexity canvas, a model that provides practitioners with a new structure to retrieve an overview of the complexity drivers within their SC and the connectivity between such. As not all complexities are bad, i.e. non-valueadding, managers are encouraged to encounter the canvas with the company’s value proposition as the guiding parameter.

Bibliography

  • 1 Christopher, M. (2000). The Agile Supply Chain. Industrial Marketing Management, 29(1), pp 37 – 44.
  • 2 Ashkenas, R. (2007). Simplicity-Minded Management. Harvard Business Review, (December 2007), pp 1 – 10.
  • 3 Serdarasan, S. (2013). A review of supply chain complexity drivers. Computers & Industrial Engineering, 66(3), pp 533 – 540.
  • 4 Scheiter, S., Scheel, O. & Klink, G. (2007). How Much Does Complexity Really Cost? ATKearney.
  • 5 Frizelle, G.D.M. (2004). Complexity in the supply chain. 2004 IEEE International Engineering Management Conference (IEEE Cat. No.04CH37574), 3, pp 1181 – 1184.
  • 6 Gottfredson, M. & Aspinall, K. (2005). Innovation Versus Complexity: What Is Too Much of a Good Thing? Innovation Versus Complexity. Harvard Business Review, 83(11), pp 62 – 71.
  • 7 Blecker, T. & Kersten, W. (2006). Complexity Management in Supply Chains: Concepts, Tools and Methods, W. Torsten, ed., Erich Schmidt Verlag GmbH & Co KG.
  • 8 Christopher, M. (2015). Keeping a Lid on Supply Chain Complexity, p. 1. Available at: http://www.martin-christopher.info/news/keeping-lid-supply-chain-complexity [Accessed November 4, 2015].
  • 9 Serdarasan, S. & Tanyas, M. (2012). Dealing with Complexity in the Supply Chain: The Effect of Supply Chain Management Initiatives. Journal of Enterprise Information Management, Forthcoming, pp 1 – 25.
  • 10 Christopher, M. (2011). Logistics & Supply Chain Management, FT Press.
  • 11 Wilson, S.A. & Perumal, A. (2009). Waging War on Complexity Costs: Reshape Your Cost Structure, Free Up Cash Flows and Boost Productivity by Attacking Process, Product and Organizational Complexity. McGraw-Hill Education.
  • 12 Bozarth, C.C. et al. (2009). The Impact of Supply Chain Complexity on Manufacturing Plant Performance. Journal of Operations Management, 27(1), pp 78 – 93.
  • 13 Kearney, A.T. (2003). Supply chain in a vulnerable, volatile world. Executive Agenda, VI(3), pp 5 – 16.
  • 14 Yin (1989)
  • 15 Oakley (1999)
  • 16 Casti, J. (1979). Connectivity, Complexity and Catastrophe in Large-scale Systems. International Series on Applied Systems Analysis, 7, p. 203.
  • 17 Winkler, H., Kalu, za, B. & Bliem, H. (2006). Strategies and Metrics for Complexity Management in Supply Chains. In T. Kersten, Wolfgang; Blecker, ed. Complexity Management in Supply Chains. Erich Schmidt Verlag, pp 3 – 19.
  • 18 Wharton, K. (2006). Creating the Optimal Supply Chain. Available at: https:// www.bcg.com/documents/file14863.pdf [accessed December 12, 2015].
  • 19 Wilding, R.D. (1998). The Supply Chain Complexity Triangle: Uncertainty generation in the supply chain. International Journal of Physical Distribution & Logistics Management, 28(8), pp 599 – 616.
  • 20 Datta, P.P., Christopher, M. & Allen, P. (2007). Agent-based modelling of complex production/distribution systems to improve resilience. International Journal of Logistics Research and Applications, 10(3), pp 187 – 203.
  • 21 Perona, M. & Miragliotta, G. (2004). Complexity management and supply chain performance assessment. A field study and a conceptual framework. International Journal of Production Economics, 90(1), pp 103 – 115

Articles

  • 7 viewpoints on S&OP implementation

    In theory, S&OP is really simple to implement, but due to cross-functional processes, involvement of different parts of the organisation and the need for IT support, it often turns out to be a difficult task to implement S&OP.

  • Keep the balance with 8 inventory management viewpoints

    Many companies initiate projects on inventory optimisation, when inventories are too large and/or delivery performance is too low. All too often, companies face both challenges at the same time and, consequently, there is a need to find a balanced approach, where net working capital and delivery performance are equally in focus.

Implement Consulting Group