Preventive Maintenance Decision Model for Multi-Station Manufacturing System Subject to System Deterioration

The qualities of products in a multi-station manufacturing system (MMS) shift from station to station. The rate of unqualified products increases rapidly with the increase of processing procedures. Repeated processing of non-conforming products is a great waste of productivity. In order to reduce the quality loss caused by quality shift, a preventive maintenance (PM) decision model is designed by balancing the reliability of MMS stations. Monte Carlo method is used to calculate the occurrence probability, defect products and expect running time of four maintenance scenarios. An optimization model is established to minimize the expect cost by using PSO algorithm. An ordinary PM model without considering quality shift is established as comparison.