A multi-agent system for integrated scheduling and maintenance planning of the flexible job shop

Abstract

This paper focuses on the problem of scheduling and maintenance planning of the Flexible Job Shop (FJS). Preventive maintenance is often being followed in the industry, which, if not considered while scheduling, may lead to unrealistic/sub-optimal schedules. Despite the importance of maintenance planning while scheduling, the problem has attracted very little attention in the literature. Further, the existing approaches assume centralized decision-making which not only suffers from low scalability but is not amenable to futuristic manufacturing systems such as industry 4.0. However, to the best of the authors’ knowledge, no decentralized system has been reported for integrated scheduling and maintenance planning of the FJS. This paper proposes a multi-agent system, a popular approach for decentralized decision-making, for integrated scheduling and maintenance planning of FJSP. The efficacy of our approach is compared with the existing approaches by solving 11 problem instances with fixed (to be performed at the predefined time) and flexible (to be performe any time within a time window) maintenance.