An efficient architecture for scheduling flexible job-shop with machine availability constraints

This paper investigates flexible job-shop problem (FJSP) with preventive maintenance activities. Makespan is considered as an optimal criterion for this problem. The contribution of this study could be divided into two parts: (1) since the preventive maintenance policies are not pondered in most prior FJSPs, we apply an enforceable and easily extendable preventive maintenance policy on FJSP; (2) in order to figure the problem out, an effective learnable genetic architecture, LEGA, which is established in previous articles for FJSPs, is expanded to integrated FJSP and preventive maintenance which is named preventive maintenance and LEGA. All operators and parameters are then calibrated by means of the Taguchi experimental design. For the evaluation of the proposed architecture, a benchmark with conscientious care is employed.