The Research Of Preventive Maintenance Decision Model For CNC Lathe Subsystem Based On Markov Process

CNC machine tools is the workhorse of the equipment manufacturing industry,whose development level has become one of the key indicators to measure thedevelopment level of the national equipment manufacturing industry. Currently, thedevelopment of domestic CNC machine tools is rapid, especially the medium andhigh-grade CNC machine tools, however, when comparing the function, speed andprecision, there exists a huge gap between domestic and foreign CNC machine tools.Failure is inevitable for CNC machine tools in the production process, even whenserious endangers the operators’ safety, which leads to the requirement of not only thegood function, but also the good performance continuity for CNC machine tools, andthe continuity can be measured by reliability and maintainability (availability).Currently, the domestic users prefer the expensive imported CNC machine toolsbecause of their high reliability and maintainability, which requires us to strengthenthe study of reliability and maintainability for domestic CNC machine tools.According to the different preventive maintenance cycle, the paper established amaintenance decision model based on the Markov decision theory and method for thesubsystems of CNC lathe. Considering the maintenance expense as the objective ofoptimization, the model is optimized by genetic algorithm in order to get the optimummaintenance strategy corresponding to the preventive maintenance cycle. Combinedwith the conclusion above, the paper established the maintenance expense model forCNC lathe subsystems based on Imperfect Preventive maintenance theory and method.The model takes the virtual age as limiting condition and the average maintenanceexpense per unit time as the objective function for optimization. As a result, theoptimal preventive maintenance frequency, preventive maintenance cycle and the bestmaintenance cycle strategy could be obtained.The main research contents and methods in this paper are as follows:(1) The CNC lathe is divided into the subsystems as the electrical parts, the mechanical parts and the hydraulic parts.(2) The preventive maintenance decision model and optimization based onMarkov process for CNC lathe is built:Combined with the Markov decision method and periodic preventivemaintenance theory, the maintenance decision model for CNC lathe subsystems isestablished. The model considers the subsystems’ state, preventive maintenance cycleand preventive maintenance strategy’s influence, and it is optimized by geneticalgorithm in order to obtain the best maintenance strategy of different preventivemaintenance cycle.(3) The maintenance expense model and optimization based on Markov processfor CNC lathe subsystems is built:Considering the periodic preventive maintenance as the incomplete maintenanceand the corrective maintenance as the minimum repair, this paper analysed theformation of the maintenance expense, took the age improvement factor, maintenancetime, preventive maintenance cycle and the best corresponding maintenance strategyinto account, and established the maintenance expense model of CNC lathe subsystem.The model regards the virtual age as limiting condition, considering the averagemaintenance expense per unit time as the objective function and is optimized byenumeration method. The optimal preventive maintenance frequency, preventivemaintenance cycle and best maintenance cycle strategy have been obtained.(4) Examples analysis:The model is optimized with the given data and the optimal preventivemaintenance scheme is obtained.