Intelligent Predictive Maintenance of Electric Trains in South Africa

Mavhungu Mathalis (Central University of Technology, South Africa); Elisha Markus (Central University of Technology & Free State South Africa, South Africa)

This paper presents a critical review of using intelligent techniques to monitor the rolling stock equipment in order to predict the maintenance needs and activities before failure could actually occur. Conducting an investigation on this topic revealed that most of the on-going studies in the railway industry is majorly on the infrastructure, and very little studies covers the topic on maintenance or the rolling stock maintenance. The paper also explores the available current source and findings about three maintenance strategies, two which are commonly practiced in South Africa and one which is taking the focus on this research. The review also revealed that artificial intelligence techniques are rarely used in monitoring of railway equipment. Part of the discussion in this paper is about the use of AI techniques in maintenance. Different applications of AI are outlined, with the main focus being the ones most suitable and relevant with regards to the purposes of this research. A discussion of a proposed maintenance strategy that uses the application of AI techniques is included in the paper, and the whole review is close off by outlining the possible challenges and future considerations relating to the topic.

Conference: UKSim-AMSS 22nd International Conference on Computer Modelling and Simulation, UKSim2020

Published: Mar 25, 2020

DOI: 10.5013/IJSSST.a.21.02.32