Modeling reliability considerations in the design and analysis of cellular manufacturing systems.

摘要

Reliability plays a vital role in the overall performance of cellular manufacturing systems (CMSs). Machine failures significantly impact the fulfillment of due dates and other performance criteria, despite the option of part rerouting to alternative workstations. These facts suggest a need for the consideration of machine reliability during the operation allocation process.;Attempting to improve a system's reliability invariably results in higher costs. It follows that the ideal strategy for achieving optimum balance lies in an approach that integrates both cost and reliability information. A mixed integer multi-objective mathematical programming model that incorporates machine reliability and cost considerations is developed for the design of CMSs. The model selects processing route for each part type which maximizes the overall system reliability of machines along the route, while minimizing the overall costs. The proposed approach provides flexible routing, ensuring high CMS performance by minimizing the impact of machine failure through the provision of alternative process routes. To account for the constant and increasing failure pattern of manufacturing machines, the CMS design model considers both the exponential and Weibull distribution approaches. A performance evaluation criterion in terms of system availability for the part-process plan assignment based on the exponential distribution is also developed. Applicability of the model is demonstrated by solving example problems by following the ∈-constraint approach.;Optimization techniques for solving such models for large practical-size problems require a substantial amount of time and memory space; therefore, a heuristic, based on the basic steps to simulated annealing and solution generation procedure of genetic algorithm is developed. The heuristic is evaluated by comparing the solutions generated by the heuristic with the LP relaxation solution for the large problems and optimal solution for the smaller-sized problems. The results reveal that the heuristic performs well in various problem instances for reliability and cost combinations. The sensitivity of the model outputs to key factors has also been investigated.;A reliability-based, preventive maintenance (PM) planning model is also incorporated, allowing CMS to restrict deterioration of machines due to usage and age and improve system reliability. A procedure for the integration of PM planning into the CMS design model is included for overall reliability and cost improvement of the CMS. Example problems are solved to illustrate the model's applicability.*.;*This dissertation is a compound document (contains both a paper copy and a CD as part of the dissertation).