| Summary: | 碩士 === 國立中正大學 === 企業管理所 === 94 === Queueing theory has already been widely applied to the research of the manufacturing system. However, the manufacturing system in the practical situation is often restricted under non-reliable factors, overly simple queueing system could not be suitable. Some scholars have presented various methods to evaluate the queueing system with non-reliable factors. This paper focused on the manufacturing system of Hi-Tech industry considering two non-reliable factors: machine breakdown and manufacturing processes, and then a non-reliable queueing model was constructed. To solve the model, first we depicted the state transition diagram and balance equations were derived, and then solved the probabilities of every possible state of the system. Secondly, the cost function and performance measures were defined. In the empirical analysis, we assumed that the capacity of the finite queue is known, and built the relations among working rate, yield rate and reworking rate in order to make the model more reasonable. Then, the cost function was transformed into the nonlinear function of the single variable with different parameter values. By solving the cost function for optimization, the optimal working rate can be determined. Finally, we proceeded the sensitivity analysis for this system. The model we constructed in this paper can offer not only another viewpoint in the research of queueing theory but also a practical tool in performance evaluation, control and optimization design of manufacturing system considering the non-reliable factors in the real world.
|
|---|
