Selective Maintenance Decision Of Series-parallel System Based On Resource Constraints

With the development of modern manufacturing system towards complexity and intelligence,and the production methods tend to be flexible,the manufacturing system is highly integrated,the working load is variable,and the operating environment is complex.Under the constraints of limited resources,the guarantee system of high reliability is facing great challenges.In recent years,the problem of selective maintenance of complex systems has attracted the attention of experts in academia and industry,and achieved fruitful results.After reviewing and summarizing the domestic and foreign research progress in the selective maintenance decision,this paper finds that advanced intelligent algorithms are used in almost all the current researches on the optimization algorithm of selective maintenance in series-parallel system.However,when the selected algorithm is not applicable or the related parameter settings are unreasonable,the accuracy of decision-making results will be affected;in the aspect of maintenance mode,there are few literatures that consider the mixed imperfect maintenance model.When only considering the single factor imperfect maintenance model,the virtual high reliability of the system task will affect the decision-maker’s judgment,and then affect the successful execution of the task;In terms of task characteristics,there is little literature considering how to reasonably plan the number of task breaks in a limited plan period and take corresponding maintenance activities to achieve maximum reliability under limited resources in serial-parallel systems.However,in reality,many suppliers arrange task scheduling according to the period given by the demander.Therefore,this paper focuses on these issues,from the perspective of accurate decision-making results of selective maintenance of complex series-parallel system,improvement of maintenance methods and how to plan the number of task interruptions in a limited planning period to maximize the reliability of the system,and carries out the research on selective maintenance decision-making of series-parallel system based on resource constraints.The main contents of this paper are as follows:1.A decision-making method of selective maintenance for series-parallel system considering the priority of component maintenance is proposed.In this paper,the task reliability model of complex series-parallel system is established.Minimum repair,preventive replacement and corrective replacement are regarded as maintenance options for manufacturing components,and a selective maintenance decision model is built by taking reliability as the maximum goal.In order to ensure the accuracy of the decision-making results of the series-parallel system,this paper considers the benefits of maintenance activities,and a selective maintenance decision-making method considering the maintenance priority of components is proposed.2.A selective maintenance decision-making method for series-parallel system under imperfect maintenance is proposed.In this paper,the problem of selective maintenance of two-state system under imperfect maintenance is solved.The mixed imperfect maintenance model is used to improve the components.Through the effective combination of cost and imperfect maintenance effect,its service age reduction factor and hazard rate adjustment factor are in line with the trend of cost input change.Finally,the selective maintenance solution method considering the maintenance priority of components is used to solve the problem.3.A selective maintenance decision-making method for series-parallel system in a limited planning period is proposed.This paper comprehensively evaluates the impact of maintenance actions on current tasks and subsequent tasks,and establishes a maintenance scheduling model under the condition of imperfect maintenance for a given limited planning time.In order to maximize the reliability of the system,the optimal interruption times are found and the maintenance measures of the components in all tasks are solved.