Joint Optimization Of Preventive Maintenance And Production Planning Of A System Under Time-Varying Operating Condition

Manufacturing systems upgrade with the increasing global competition.Modern production systems have gradually developed to precision,information,integration,and flexible manufacturing.An important feature of modern manufacturing systems is that machines work under time-varying operating condition(OC).OC studied in this research refer to the covariates,such as raw material,cutting depth,feed rate,rotation speed and production rate etc.What's more,short-term production plans are adopted by modern enterprises to adapt randomly arrived orders.Preventive maintenance(PM)maintains machines at reliable states,and the production plan(PP)ensures the completion of tasks.PM and PP are the two key elements of modern enterprises to reduce operating costs and improve competition ability.They interact with each other.It is of great significance to conduct joint optimization of PM and PP of a system under time-varying OC.It is crucial for manufacturing enterprises to maintain a safe,efficient and green production environment.An extended imperfect maintenance model of a machine working under piecewise constant OC is developed based on a classic model.A dynamic PM strategy based on a shortterm production plan is also proposed.For a batch production system,a machine-system layer interaction based opportunistic PM strategy is proposed considering OC variance between different batches.An integrated model of production speed and PM based on a game-theoretic framework is presented to accommodate with the unequal decision priority between a PM section and a PP department.At shop floor level,joint optimization of job schedule and PM considering OC is also conducted.This research can control the maintenance cost,improve the fluency of production,and reduce the gap between theoretical research and practical application.The main results of this research are as follows,(1)This research employs the accelerated failure time model(AFTM)to integrate the OC covariates with the traditional failure time function.A model is developed to convert the operating time under different OCs to the equivalent time under the baseline OC.The evolution models of reliability and failure rate of a machine when OC changes are constructed based on the model.An extended imperfect maintenance model is proposed based on a classic model.A short-term production plan based dynamic PM strategy is presented to minimize maintenance cost.It provides a good supplement to the PM for a single-machine system.(2)A dynamic opportunistic PM strategy based on a machine-system layer interaction for a multi-machine production system with a time-varying batch production pattern is proposed.It considers the variable degradation speed between batchs results from the time-varying OC,and the PM restriction during the processing of a batch.It avoids the potential influence on the product quality uniformity of a same batch.It can achieve the minimal total cost comparing with other strategies as the system downtime cost and maintenance cost have been jointly optimized.It extends the PM strategy for a batch production system.(3)A joint decision-making model based on a non-cooperative game framework is proposed to solve the problem of unequal decision-making priority in the joint optimization of lot size and PM.It meets the actual production decision-making process,where the decision maker of PP has decision priority over PM planner.The model can effectively handle the interaction effect between PP and PM,ensure the fulfilling of production requirements,minimize the maintenance cost,and avoid the waste of resources reallocation and the interruption of production process.(4)An integrated model of job scheduling and PM considering OC is developed.It takes the OC of each job into account,which affect the machines' failure rates,the corresponding PM plan,and the expected completion time of jobs.An effective GA-based heuristic algorithm is proposed for large scale problems.Numerical examples are conducted to validate the effectiveness of the integrated model and the necessity of considering OC.It enrichs the integrated optimization of job scheduling and PM.The model established in this research can accurately describe the reliability and failure rate evolution rules of the modern production system running under time-varying OC.The PM strategy proposed is adapted for the short-term PPs.The joint model of production rate and PM based on a non-cooperative game framework is close to the decision-making process of most enterprises.The integrated model of job scheduling and PM has a certain applicability.This research expands the traditional maintenance theory and production planning.It increases the applicability in the practical production process.It can provide enterprises with a safe and efficient production process,and enhances their competition ability.It can prospel enterprises to improve machine health management ability.Hope it can somehow promote the industrialization 4.0 of China.