Print Email Facebook Twitter Influencing factors for condition-based maintenance in railway tracks using knowledge-based approach Title Influencing factors for condition-based maintenance in railway tracks using knowledge-based approach Author Jamshidi, A. (TU Delft Railway Engineering) Hajizadeh, S. (TU Delft Railway Engineering) Naeimi, M. (TU Delft Railway Engineering) Nunez, Alfredo (TU Delft Railway Engineering) Li, Z. (TU Delft Railway Engineering) Date 2017 Abstract In this paper, we present a condition-based maintenance decision method usingknowledge-based approach for rail surface defects. A railway track may contain a considerable number of surface defects which influence track maintenance decisions. The proposed method is based on two sets of maintenance decision factors i.e. (1) defect detection data and (2) prior knowledge of the track. A defect detection model is proposed to monitor surface defects of the trackincluding squats. The detection model relies on track images and Axle Box Acceleration (ABA) signals to give both positions of severity and defects. To acquire the prior knowledge, a set of track monitoring data is selected. A fuzzy inference model is proposed relying on the maintenance factorsto give the track health condition in a case study of the Dutch railway network. The proposed condition-based maintenance model enables infrastructure manager to prioritize critical pieces of the track based on the health condition. Subject Condition-based maintenance decisionRail surface defectsBayesian model To reference this document use: http://resolver.tudelft.nl/uuid:754027da-0ca8-4064-b312-1cd3ce99da19 Source Proceedings of the First International Conference on Rail Transportation: ICRT2017 Event 1st International Conference on Rail Transportation, 2017-07-10 → 2017-07-12, Chengdu, China Part of collection Institutional Repository Document type conference paper Rights © 2017 A. Jamshidi, S. Hajizadeh, M. Naeimi, Alfredo Nunez, Z. Li Files PDF ICRT2017_ID_186.pdf 2.27 MB Close viewer /islandora/object/uuid:754027da-0ca8-4064-b312-1cd3ce99da19/datastream/OBJ/view