Proactive maintenance scheduling in consideration of imperfect repairs and production wait time

Abstract

Wait time due to the exhaustion of raw materials or lack of demand is common in manufacturing/production processes, which provides cost-effective maintenance opportunities. This paper designs a two-phase opportunistic maintenance framework based on defect information, which integrates properties of production waits into the decision-making process. At the first phase, a finite number of inspections are executed to reveal the defective state, followed by imperfect preventive repair. At the second phase, no maintenance action is taken until reaching a scheduled maintenance window (postponed maintenance) or arrivals of production waits (opportunistic maintenance). The integration of imperfect repair with postponed and opportunistic maintenance enables a sufficient utilization of remaining useful lifetime and a flexible resource allocation. Under the constraint of the steady-availability requirement, the optimal policy minimizing the system maintenance cost is obtained via the genetic algorithm. A case study from a steel convertor plant shows that our policy outperforms some classic maintenance policies.