Preventive Maintenance Policy On A Serial Manufacturing System Considering Costs And Availability

Preventive maintenance plays a key role in ensuring today's manufacturing system performance.Nevertheless,serial manufacturing systems are sensitive to any kind of stop.In fact,a failure,or a maintenance action leads to an avoidable downtime.During the downtime,machines are inactive and created a cost.The production is not done,which represents a loss for the company and it is traduced by a cost.If the downtime is caused by a failure,a minimal repair will be performed to make the system work again as soon as possible.However,models tend to predict system failures and maintenance actions are performed ahead to minimize unwanted downtime.This paper assumes machines degradations are caused by time and external factors.Failures follow a Weibull distribution.Costs and availability evaluation functions are observed to elaborate a global policy.From those factors a global-objective function is created in which the cost is minimized,and the availability maximized.The downtime caused by a maintenance action on one machine should be used to perform maintenance actions on other machines.This is an opportunistic model.One can be based on time allocate to perform an opportunistic action,another can be based on the location of the machine in the whole system.This paper introduces an opportunistic model based on the location of the machine in the system as regard to the bottleneck machine and the time-window allocated to perform the opportunistic maintenance action.The comparison of those models proves how complex it is to build a maintenance policy.This complexity results from the number of parameters handled and the computational cost experiences take to perform.Even though the combined model takes longer to be computed,the final total cost is lower than the other models.