A Mean-Variance robust model to minimize operational risk and supply chain cost under aleatory uncertainty: A real-life case application in petroleum supply chain

Abstract

The purpose of the research work is to consider a multi-echelon, multi-product, and multi-modal petroleum supply chain network design problem along with its various sources of uncertainties (demand, supply, production, etc.), and to minimize both total supply chain cost and risk simultaneously. The problem is articulated as a robust optimization problem and the results are derived under various risk attitudes (viz. risk-seeking, risk-neutral and risk-averse behaviors). The non-linear problem is proposed as a Mean-Variance robust optimization problem. Two-stage stochastic programming is extended to incorporate the robustness and capture the risk aversion behavior. The scenario-based planning method is used for the estimation of uncertain parameters. A real-life case study of petroleum supply chain is conducted in the Indian scenario, considering the requisite challenges and constraints. The results show that total supply chain cost and risk demonstrate conflicting behavior with each other. Total supply chain cost increases with an increase in risk aversion level. Significant amount of operational risks can be reduced with slight increase in total supply chain cost. Penalty cost contributes the maximum to the increase in total supply chain cost along with risk aversion level.