A model to optimize the maintenance activities in job-shop manufacturing plants

Job shop processes are traditionally characterized by a large variability of the workload assigned to the different machines, due to the wide set of orders processed. To the contrary, such variability is not generally taken into account in maintenance plans generation. This can lead to a wrong definition and distribution of maintenance budget which, in turn, can prevent the achievement of the desired performance in terms of machine availability and operational costs. This paper provides a model that pursues the optimization of each machine maintenance using a master production schedule and availability historical data. The model, analyzing orders forecasts, production cycles and historical failure data, is able to evaluate the criticality of each production unit. An analysis of the kinds of failure and their frequency allows the identification of the actions to undertake to increase availability. Subsequently, the actions are ordered by cost and by the corresponding availability improvement obtainable. In this way, the set of maintenance activities which can lead either to a desired target of availability or to the minimum global cost (including both maintenance and downtime costs) can be determined