Process Planning Generation for Reconfigurable Manufacturing Systems Based on Improved Squirrel Search Algorithm

The concept of reconfigurable manufacturing has emerged in response to the need to adapt to the rapidly evolving market environment and meet the growing demands for personalization and customization in manufacturing. Reconfigurable Manufacturing Systems (RMS) are designed to combine the cost-efficiency features of rigid manufacturing systems with the agility to respond swiftly to market changes, similar to flexible manufacturing systems. Within the field of RMS research, process planning has consistently been recognized as a crucial aspect and a key technological focus. This paper addresses the process planning problem in reconfigurable manufacturing systems and proposes a multi-objective integer programming model. An improved squirrel search algorithm is also proposed to solve the problem. Firstly, the squirrel search algorithm is combined with the Pareto dominance theory to generate non-dominant solutions. Secondly, an adaptive predator presence probability strategy is proposed, which changes the predator probability from a fixed value to a decreasing value during the iteration process. This strategy aims to balance the global and local search of the algorithm. Finally, to enhance the algorithm's local search capabilities and prevent it from converging to local optima in the later stages, a random mutation strategy is employed. Additionally, the TOPSIS method is utilized to aid decision-makers in ranking and selecting the most optimal process plans. To validate the feasibility and practicality of the proposed model and new algorithm, numerical experiments were conducted. The results of the experiments demonstrate the superior performance of the algorithm presented in this paper.