Procurement Optimization for Manufacturing Enterprises Considering Supply Chain Disruption Risks and Carbon Emissions

This study addresses the procurement problem in mechanical manufacturing enterprises, considering both supply chain disruption risks and carbon emissions. Based on a multi-product, multi-supplier procurement planning optimization problem, a high-dimensional multi-objective optimization model is developed with procurement cost, total loss, number of quality defects, and carbon emissions as objectives. The model is solved using an improved integer-coded NSGA-III algorithm, which includes four mechanisms: heuristic population initialization, infeasible solution optimization and repair, a weight-matrix-based crossover operator, a multi-column exchange mutation operator, and Pareto simulated annealing. Through numerical experiments, the performance of this algorithm is compared with NSGA-III and NSGA-II, demonstrating its superior ability to handle multi-objective, multi-constraint optimization problems. Ablation experiments further validate the effectiveness of the four improved mechanisms. Case study results show that the optimized procurement plan balances economic and environmental benefits while considering supply chain risks.