A Co-evolutionary Algorithm for Open-shop Scheduling with Disassembly Operations

In recent years, many used products are returned from customers through reverse supply chains in order to be reused and recycled. Therefore, it is necessary to provide suitable disassembly procedures from the view point of economic efficiency. This research proposes a scheduling method for open-shop scheduling problems containing both disassembly operations and post-processing operations. A co-evolutionary algorithm is proposed to modify not only the sequences of disassembly and post-processing operations of products but also the sequences of products loaded on disassembly and post-processing machines. Two different kinds of individuals are modeled for both the sequences of operations and the sequences of products in the co-evolutionary algorithm. These individuals alternately repeat evolution at short time intervals, and affect their fitness values each other. We developed a prototype of scheduling system and applied it to some computational experiments for open-shop type scheduling problems including disassembly and post-processing operations. The experimental results of the proposed method were compared with the one of another scheduling method with various heuristic rules in order to evaluate the effectiveness of the proposed method.