Optimizing Pull Production Systems with Cost Objectives by Modified Memetic Algorithm

This paper is concerned about the sequencing problems in pull production systems consisting of one mixed-model assembly line with limited intermediate buffers and one fabrication flow line with limited intermediate buffers. The optimization objective is to minimizing the total makespan cost in the whole production system. The mathematical models and, considering the safety inventory of the demanded parts, the method to determine the production sequences for the fabrication line according to the production sequences for the mixed-model assembly line are presented. Since the problem is NP-hard, a modified memetic algorithm is proposed for solving the models. In this algorithm, the method of generating the initial population, the local search procedures, the selection, crossover and mutation operators are newly designed. Optimization results are compared between the algorithm proposed in this paper and the traditional genetic algorithm. The feasibility and efficiency of the modified memetic algorithm is shown by the comparison results.