Breaking Local Optima: A Multi-Stage Strategy for Milk-Run Logistics Optimization

This paper explores the Milk-Run problem, a significant issue in modern manufacturing logistics concerning factory distribution and material collection. With the increasing globalization and complexity of manufacturing, the impact of logistics management on production efficiency and cost control has become increasingly prominent. Traditional Vehicle Routing Problem (VRP) algorithms face challenges such as local optima, high computational complexity, and difficulties in handling complex constraints. To address these challenges, this paper proposes a multi-stage optimization strategy. Initially, the algorithm generates a distribution route using the nearest neighbor algorithm, further optimizes it through the 2-opt local optimization algorithm, and finally employs a simulated annealing algorithm for global optimization to avoid local optima and improve route quality. The results indicate that the proposed algorithm effectively enhances resource utilization and reduces transportation costs, demonstrating broad applicability and practical value. This study lays the groundwork for further exploration and development in logistics and supply chain management.