A Joint Optimization of Strategic Workforce Planning and Preventive Maintenance Scheduling: A Simulation–Optimization Approach

Abstract

This paper proposes a Simulation–Optimization framework, where a maintenance system of high-value assets is modelled via a novel large-scale Discrete Event Simulation model. In contrast to the existing simulation models in the maintenance domain, the developed simulation model provides an integrated view of the different aspects of the maintenance system including asset acquisitions, maintenance workforce planning, and scheduling of preventive maintenance activities. Further, the workforce planning studies the technicians’ progression based on their allocated maintenance activities, and the trade-off between their progression and the recruitment of additional technicians, which is infrequently studied in this domain. The developed simulation model is coupled with a Differential Evolution (DE) algorithm, that is further enhanced with a K-means clustering machine learning method. By using this coupling (i.e., simulation-based optimization), we jointly optimize the scheduling frequency of preventive maintenance activities and workforce planning decisions (e.g., recruitment and career progression). The proposed framework proved its superiority in terms of representing the dynamics of the overall system, and providing optimal decisions about the workforce and maintenance schedules. A comprehensive study is proposed and showed the ability of the proposed framework to reduce the total maintenance cost on average of 5.6%, that is around three million currency units.