Train delay is one of the biggest challenges that is faced by the railway systems across the globe. The mechanism used to predict the train delay is based on a lot of factors that are directly or indirectly linked to delay of trains. They rely on the rules that are made by the experts of railway systems that are based on univariate statistics. There has been an increased interest in the application of advanced data analytics to solve train specific problems such as maintenance of railway assets, automatic visual inspection systems, network area estimation, energy efficient railway operations. Prediction of failures by most of the existing methods is done by comparing the current or computed values with a set of standard values. A few proposed methods are based on: Fuzzy rules and the univariate autoregressive integrated moving average (ARIMA) model, Graph topology, scheduling algorithms for PROFINET and MVB, the adaptive iterative learning control (AILC) mechanism and Markov model based methods. This paper presents a survey of the existing train delay detection techniques.
Volume 11 | 04-Special Issue
Pages: 917-920