Impact of the Decision Horizon on Railway Systems Maintenance and Service Scheduling

Abstract

Prognostics and health management (PHM) has, recently, gained a lot of attention in the rail transportation context. However, most works in this field focus on the prognostic process. However, prognostics is not the end of PHM in itself, its results have to be used to better manage the life cycle of systems. PHM decision phase is applied to plan action in both operational and maintenance ways based on the current condition and the estimated RUL. One important parameter of this phase is the length of the scheduling horizon. In this paper, we propose a methodology to set the suitable values of this main parameter. In this paper, we also study the influence of some rolling stocks application parameters on the value of the decision horizon. As a result, we obtain a range of values for the decision horizon that optimizes the maintenance and service policies.