Link and effect model for performance improvement of railway infrastructure

Abstract: Railway traffic has increased over the last decade and it is believed to increase further with transportation shifting from road to rail, due to rising energy costs and the demand to reduce emissions. To manage railway infrastructure assets effectively against agreed-upon and set objectives, performance must be measured and monitored. Different systems are used to collect and store data of traffic, failures, inspections, track quality, etc., for subsequent analysis and data exchange. Performance indicators (PIs), e.g. for RAMS (reliability, availability, maintainability, safety), are continuously developed to support infrastructure managers (IMs) in identifying performance killers in order to make efficient and effective decisions. However, they are often ad hoc and seldom standardised. Moreover, the use of standards and the need for harmonisation of railway operations have grown with interoperability, e.g. building of a trans-European railway network. The~efficiency and effectiveness of railway infrastructure can be improved if an appropriate performance measurement (PM) system is identified and specifically developed. In traditional PM systems, PIs are given threshold values, indicating when an action needs to be taken, i.e. they can to some extent be reactive. Also, PIs are often aggregated measures, which can make them abstract. By this trend in transportation and shortcomings in performance measurement, there is a need to improve the strategic planning and measurement of performance for more proactive decision making and future standardisation. In this research, a link and effect model for performance improvement of railway infrastructure is developed. It provides a continuous methodology for breaking down objectives into operational requirements and linking them to results, using performance indicators, and algorithms for data analysis and simulation, for decision support. Keywords: railway infrastructure, performance, RAMS, maintenance, dependability, indicators, link and effect, decision support