Risk adjusted multicriteria supplier selection models with applications

Most manufacturers are continuously seeking their supplier base around the world and look for an opportunity to significantly reduce supply chain costs. Singular emphasis on supply chain cost, however, can make the supply chain brittle and more susceptible to the risk of disruptions. The objective of this paper is to develop multicriteria supplier selection models incorporating supplier risk and apply them to a real company. We develop two different types of risk models, value-at-risk (VaR) and miss-the-target (MtT). We model the risk-adjusted supplier selection problem as a multicriteria optimisation problem and solve it in two phases. Phase 1 is the pre-qualification step, where a large set of initial suppliers is reduced to a smaller set of manageable suppliers using various multi-objective ranking methods. In Phase 2, order quantities are allocated among the short listed suppliers using a multi-objective optimisation model. In the multi-objective formulation, price, lead-time, VaR type risk of disruption due to natural event and MtT type risk of quality are explicitly considered as four conflicting objectives that have to be minimised simultaneously. We solve the multi-objective optimisation problem using four different variants of goal programming. The models are illustrated with an actual application to a global IT company.