Data-driven model for maintenance decision support: A case study of railway signalling systems. (January 2016)

• Record Type: Journal Article
• Authors: Morant, Amparo; Larsson-Kråik, Per-Olof; Kumar, Uday
• Is Part Of: Proceedings of the Institution of Mechanical Engineers. Volume 230:Number 1(2016)
• Page Start: 220
• Page End: 234
• Publication Date: 2016-01
• Subjects: Railway -- signalling systems -- decision support model -- data-driven model -- failure analysis -- maintenance -- corrective maintenance -- dependability -- case study -- reliability -- availability -- maintainability and safety -- data integration; Railroads -- Periodicals; Personal rapid transit -- Periodicals; 625.1
• DOI: 10.1177/0954409714533680
• Languages: English
• ISSNs: 0954-4097

Abstract

Signalling systems ensure the safe operation of the railway network. Their reliability and maintainability directly affect the capacity and availability of the railway network, in terms of both infrastructure and trains, as a line cannot be fully operative until a failure has been repaired. The purpose of this paper is to propose a data-driven decision support model that integrates the various parameters of corrective maintenance data and to study maintenance performance by considering different reliability, availability, maintainability and safety parameters. This model is based on failure analysis of historical events in the form of corrective maintenance actions. It has been validated in a case study of railway signalling systems and the results are summarised. The model allows the creation of maintenance policies based on failure characteristics, as it integrates the information recorded in the various parameters of the corrective maintenance work orders. The model shows how the different failures affect the dependability of the system: the critical failures indicate the reliability of the system, the corrective actions give information about the maintainability of the components, and the relationship between the corrective maintenance times measures the efficiency of the corrective maintenance actions. All this information can be used to plan new strategies of preventive maintenance and failure diagnostics, reduce the corrective maintenance and improve the maintenance performance.