Solving the combined modular product scheduling and production cell reconfiguration problem: A genetic algorithm approach with parallel chromosome coding

Abstract

Manufacturers have to balance the desire for high productivity and the need for rapid responsiveness. These two interrelated problems can be solved by arranging the production of a family of modular products in Reconfigurable Manufacturing Systems (RMS). The potential benefits of RMS can be achieved by reconfiguring its structure in such a way that each configuration corresponds to one product variant in the same family. The successful implementation of this strategy lies in efficient scheduling of the system. However, little research has been done in addressing the scheduling issues in RMS. This study presents a modeling methodology to simultaneously determine the processing sequence of a family of modular products and select the optimal system configuration for producing each product variant. Due to the combinatorial nature of this problem, a genetic algorithm (GA) with parallel chromosome coding scheme is proposed to provide quick and near-optimal solutions to large problems. The efficiency of the proposed approach is demonstrated by comparing the computational results with that achieved using LINGO and by applying it to both small size and large size models.