Multiobjective Optimization of Preventive and Corrective Maintenance Policies in Unreliable Manufacturing Systems
This paper considers the optimisation problem of controlling preventive and corrective maintenance policies of unreliable manufacturing systems by minimizing the surplus cost and by maximizing the system availably. The considered manufacturing systems consist of two machines, producing one part type and subject to random failures and repairs. A multiobjective optimisation approach, based on the desirability function, is used to determine the parameters of the production, preventive and corrective maintenance policies. The maintenance policy is based on the existing blockreplacement policy. Such a policy consists of replacing component at failure periods or preventively at scheduled intervals (T). The spare parts inventory management is based on (S, Q) model in which an order is placed when the replacement stock level decreases under a safety threshold level (S). A replacement part quantity (Q) is then ordered and received after a stochastic lead time (t). Compared to the most maintenance models developed in the scientific literature, based on a single response variable, the proposed model, allows optimizing simultaneously two response variables (cost and availability) and shows the best compromise parameters (T, S, Q) between the cost minimization and the system availability maximization.
Maintenance policies Desirability function Multiresponse system Simultaneous optimization Experimental
Jean-Francois BOULET Ali GHARBI Jean-Pierre KENNé
école de Technologie Supérieure,Automated Production Engineering Department,University of Québec Pro école de Technologie Supérieure,Mechanical Engineering Department
国际会议
北京
英文
2007-05-30（万方平台首次上网日期，不代表论文的发表时间）