Mathematical Programming Approach to Optimize Tactical and Operational Supply Chain Decisions under Disruptions

Supply chain (SC) networks have become more prominent, complex, and challenging to manage, especially considering the multitude of risks and uncertainty that may manifest. Studies have shown two basic approaches to hedge against the negative impact of SC disruptions: proactive and reactive. While the former methods suggest different approaches to generating robust and resilient structures, the latter approach ensures that the SC recovers effectively. A general shortcoming of existing work is not considering SC dynamics. Consequently, disruptions are considered static events without including the durations and recovery policies. In this work, we develop a SC model that aids decision-making in addressing disruptions by considering proactive and reactive strategies. We adopted a discrete time-expanded model to solve the SC problem and consider the disruption dynamics using the rolling horizon framework. In the proposed SC model, a graph network represents the SC, where the nodes consisting of suppliers, manufacturing sites, warehouses, and customers interact using the arcs. The arcs determine the flow of materials between nodes. Independent disruptions can occur at the nodes and/or arcs, and the time of disruption is quantified using the geometric distribution. In the advent of disruption, we have adopted adjusting routing plans, inventory levels, capacity flexibility, and other tactical and operational decisions to hedge against disruption. To illustrate the proposed approach, we used a small problem to illustrate the effect of arcs and node disruption in decision-making and a realistic case study to demonstrate the proposed framework's computational complexity. The results suggested that the effect of node disruption is more predominant because the initial network configuration limits the flexibility at the nodes. Furthermore, it was shown that the SC operated efficiently, as the solution offers a balance between the service level and the total cost of operating the SC.