Accelerating Benders Decomposition via a Stronger Valid Inequality for Disruption-Resilient Supply Chain Design

Reliable supply chain network design has become an important issue in supply chain management, particularly in environments where disruptions may significantly affect system performance. In this study, a disruption-aware supply chain network design problem is investigated using a mixed-integer programming model. The model is developed based on an existing formulation from the literature, and Benders decomposition is employed to improve the solution process. The main objective of this research is to enhance the computational efficiency of the solution algorithm by strengthening the mathematical formulation. To achieve this goal, two valid inequalities from the base model are combined into a single stronger constraint that tightens the master problem in the Benders decomposition framework. To evaluate the effectiveness of the proposed enhancement, several test instances with different network sizes are generated and solved using GAMS. Computational results demonstrate that the enhanced formulation significantly reduces solution time and improves convergence compared with the original formulation and the standard Benders decomposition approach. Moreover, the proposed improvement does not affect the optimal solution but substantially enhances computational performance. The findings indicate that strengthening model formulations can play an important role in solving disruption-aware supply chain network design problems more efficiently.

کلیدواژه ها:

Supply Chain Network Design ، Mixed Integer Programming ، Benders Decomposition ، Valid Inequality ، Computational Performance