Decision support tool based on multi-source data analysis for the tram wheel-rail interface
Wei Xie (Student TU Delft)
AA Nunez Vicencio (TU Delft - Railway Engineering)
Ron Maas (Sensornet)
Zili Li (TU Delft - Railway Engineering)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
This paper presents a methodology to support decision making based on the tram wheel-rail interface condition. The methodology relies on the following measurements: tram failure log-files regarding wheel-sliding events, monitored acoustics data and open source weather information. The proposed methodology consists of three stages: 1) data collection and pre-processing, 2) spatial analysis based on clustering, and 3) decision support based on the extracted information. For clustering, the Density-Based Algorithm (DBSCAN) is used for the analysis of wheel-sliding events. Self-organizing maps (SOMs) are employed for the analysis of acoustics data. A real-life case study is used to show how use of the methodology can find interesting hotspots that are candidates for further monitoring and maintenance actions. The measurements were obtained from the tram system in the city of Rotterdam, The Netherlands.