An Improved Heuristic and NSGA-II with Q-Learning for Integrated Process Planning and Scheduling in Reconfigurable Manufacturing Systems

Integrated Process Planning and Scheduling (IPPS) with Reconfigurable Manufacturing Systems (RMS) is a complex planning problem. In this work, we address this problem using two solution approaches. The first approach is an extension of a heuristic from the literature, adapted to handle multi-objective optimization. The second approach is a Non-dominated Sorting Genetic Algorithm metaheuristic, enhanced with a Q-learning technique to dynamically update the crossover probability in each iteration (QNSGA-II). Our results demonstrate that the extended heuristic outperforms the mono-objective heuristic, while the Qlearning enhanced NSGA-II provides the best results compared to both heuristics.