Using complex network theory to model supply chain network resilience: a review of current literature

Abstract

Traditionally, supply networks are modelled as multi-agent systems, in order to represent explicit communications between various entities involved. However, due to the increasingly complex and interconnected nature of the global supply networks, a recent trend of research work has focussed on modelling supply networks as complex adaptive systems. This approach has enabled researchers to investigate various topological properties which give rise to resilience characteristics in a given supply network. This paper presents a critical review of the published research work on this field. Key insights provided by this paper include; (1) the importance of defining the concepts of ‘resilience’ and ‘disruptions’ as measurable variables; (2) the limitations of existing network models to realistically represent supply networks; (3) potential improvements to the currently used growth mechanisms, which rely on node ‘degree’ to derive attachment probability instead of the more realistic and relevant node ‘fitness’; (4) importance of incorporating operational aspects, such as flows, costs, and capacities of connections between the nodes as well as topological aspects; and (5) derivation of a new set of resilience metrics capturing operational as well as topological aspects. Finally, a conceptual approach incorporating the above improvements to the existing supply network modelling approach is presented.