| With the rapid development and transformation of modern industry,efficiency and quality have become the key to improve the competitiveness of manufacturing enterprises.Scientific and reasonable preventive maintenance of machine can effectively improve its availability and economy.However,maintenance will interfere with the operation of the production line.Many enterprises set up buffers between machines to reduce the impact of a single machine shutdown on the production line,but this affects the original maintenance system.In addition,there are many quality influencing factors in the production process,which will affect maintenance decisions.To meet the increasing quality management needs,quality factors need to be considered in maintenance planning.This thesis aims at the two-machine one-buffer production system.Preventive maintenance,buffer and quality factors are comprehensively considered to study the maintenance planning of the production system.(1)Based on the interaction mechanism between buffer and machine states,imperfect preventive maintenance and minor repairs that are more realistic in actual production are considered,and a state transition model for two-machine one-buffer system is established.Based on the Markov decision process,a long-term expected profit model of the system is established.The optimization goal is to maximize profit instead of cost minimization,and the optimal maintenance strategy under different system states is solved,in order to make full use of the regulating effect of buffer on maintenance and then the system efficiency can be improved.(2)Considering the random quality influencing factors during the production process in the maintenance plan of the two-machine one-buffer system,the random fluctuations in the quality of parts are described.The cost loss caused by the quality fluctuation is quantified by Taguchi’s quality loss function,and the cost structure and long-term expected profit model of the system are improved.Subsequently the corresponding optimal maintenance strategy is solved.Through the analysis of examples,the impact of considering random quality fluctuations on system maintenance decisions is discussed.(3)The degradation of machine state will also cause the degradation of the quality of parts.Combined with random quality influencing factors,the comprehensive quality degradation process is described.Then a quality loss cost model and a corresponding long-term expected profit model after considering the comprehensive quality degradation is established,and the optimal maintenance strategy of the system is solved.The necessity of considering comprehensive quality degradation in maintenance decisionmaking is discussed through example analysis.To sum up,this thesis proposes a maintenance planning method for twomachine one-buffer system that integrates quality factors,and provides a certain theoretical support for the maintenance decision and quality improvement of production lines with buffer that widely used in manufacturing enterprises.It is helpful for the enterprise to carry out reasonable maintenance planning and quality management in the production process to achieve efficiency and quality improvement. |