Performance Evaluation Of Hybrid Production-line Systems

Due to the specialization of work division and the development of modern market,it is obvious to us that the production system presents network characteristic.The relationships between different level and organization from the production department to the supplynetwork are complicated and variational.and what is more,they form the complex netwok which is response itself and randomization. Meanwhile,there is a lots of stochastic events which are controllable or uncontrollable in the manufacture system.Many academicians has studied the unreliable system for the inescapability of device failure.The performance evaluation of production plays an important role in the field of floor control.The paper was based upon the simplification of unreliable production network, with the studying the hybrid production such as merge system and the two-machines model.The main contributions of the dissertation are as follows:1.The dissertation studied background of the thesis, then a brief introduction on the basic notions of unreliable production lines was presented, reviewed the developments and research situation of unreliable production system and point out some deficiencies and future research directions.2.In accordance with a hybrid production line system consisting of two machiens subjected to independent failure modes which contains the so-called inadequacy farlure and unreliable buffer is considered, based on discrete model, the states transition probabilities of the system are studied analytically, with the reducing the difficulties of solving the problem, a corresponding algorithm are developed here and also the corresponding program is provided.3.Introduce a new system performance evaluation method to the two-machine line. After that, the extended system aggregation model is developed and corresponding aggregation formulations are deduced. Different from traditional production models, the production line features unreliable buffers and multiple stochastic failure modes of the machines. The method is applicable to analyzing the cases arising from two or more stochastic events or more complex production lines.4.To deal with hybrid nature of coexistence of continuity and discreteness in pmduction line, modeling and optimization control methods of hybrid system were proposed. The hybrid automata model of a manufacturing cell was studied.Based on above analysis, the optimal control problem of the hybrid system was studied,which maximized utilization rate of machines with certain production output. Numerical simulation based on Monte Carlo method was developed. The feasibility and efficacy of the model and control algorithm were proven via simulation example of a finite buffer capacity tandem line.5. A brief summary of this thesis and the follow-up work is viewed.