Intelligent rail maintenance decision support system using KPIs

Key Performance Indicators (KPIs) enable the infrastructure manager to keep the performance quality of the infrastructure at an acceptable level. A KPI must include specific features of the infrastructure such as functionality and criticality. The KPIs can be classified into three performance levels: (1) technical level KPIs; (2) tactical level KPIs and (3) global level KPIs. For instance, some KPIs are related to individual rail components (technical level) and some correspond to a bigger picture of the rail including multiple components (tactical level). The global level also gives an overview indication of the full-length rail based on what the infrastructure manager requires. Hence, to use every KPI correctly, the infrastructure manager should be aware of the proper KPIs level.

In this dissertation, an intelligent rail maintenance decision support system using KPIs is proposed. The thesis is composed of three parts: design of KPIs, rail degradation model and condition-based maintenance decision system.