Study Of Several Issues In Reliability Modeling And Inventory Management

Reliability is one of the important characteristics of an equipment or system. Maintenance decision-making activity is based on reliability analysis and modeling on the equipment (system), and proper maintenance decision-making is an important guarantee for maintaining or restoring the inherent reliability of the equipment. Maintenance decision-making also concerned with the spare parts in logistic level. Understanding the reliability of the equipment and mastering the failure law of the equipment is a key issue to gain an effective decision-making on spare parts inventory management. Actually, effective spare parts inventory management can provide accurate and timely supply of spare parts for the failure of equipment. Therefore, it's of great significance to research on reliability of equipment and spare parts inventory management. The major contributions are as follows based on the statements above.(1) The impact of different sampling time on the estimated reliability models and their reliability features in the process of reliability modeling is discussed. In industrial production some equipment may be newly acquired, so the information available from field data is incomplete for reliability modeling. In this way, study on the least data of samples for reliability modeling is of great importance. We discussed the following circumstances:·Data is witnessed while the product is undergoing failure ;·Data is observed before very moment of failure occurring, that is ,take failure data as censored;·Data is got between two failures.(2) Inventory classification. A huge number of inventory items are usually kept by the industrial sector or enterprise. Classifying and manage them well can enhancing production efficiency to some extent. Based on the traditional ABC classification approach and the idea of cluster analysis, we relax requirements of the ratio and number of groups of the spares, and then class the items with high similarity into one group. Thus, the boundary between each cluster is derived. The approach can always class two adjacent items into the same category.(3) Demand forecasting for the spare parts based on Monte Carlo simulation. Lacking of spare parts or excessive reserving can both impact on management decision-making. We perform demand forecasting simulation approach of spare parts under all potential case by using reliability model and maintenance policies of them. The output of the simulation is an experience distribution of the demand size which can be further fitting to an appropriate theoretical distribution so as to support other decision-making activities.