Selection of Optimal Redundancy Strategies for a Supply Network

Supply chains have become more vulnerable because of increased globalization and interdependency between network actors and risks. It is, therefore, extremely important to model interdependency between risks taking into account the supply network configuration. In this paper, we have followed the existing concept of modelling supply network as a Bayesian belief network capturing network configuration, probabilistic interdependency between risks and losses resulting from the realization of risks. We propose a new method for evaluating efficacy of different combinations of redundancy strategies. For each node of the network, we incorporated a strategy of adding redundant resource as a risk management approach that would disengage the particular node from its parent node(s). The model was solved against the bicriteria objective of maximizing normalized expected utility for loss exposure and normalized utility for cost of redundancy strategies keeping into account different risk attitudes and criteria weights. The model also helped in identifying optimal combination of strategies for specific allocation of budget as there could be different combinations of allocating redundant resources across the network resulting in supoptimal values of risk measures.