Railway operators depend on accurate and reliable predictions on future arrival and departure times of trains. This information is also disseminated to passengers to facilitate their journeys. Railway operations are influenced by many variables and are thus highly stochastic and dynamic. Hence modeling the variability of arrival time prediction and its evolution over time and space is of great interest. We propose a set of non-stationary Markov chain approaches to model the stochastic evolution of train delays over a train run. The Markov chains base on a varying combination of arrival, departure, running, and dwell time events. The models are applied to a railway corridor in Switzerland, where we discuss the precision and accuracy of the models. The results indicate a way to improve stochastic railway predictions.
Modeling Evolutionary Dynamics of Railway Delays with Markov Chains
2021-06-16
2244109 byte
Conference paper
Electronic Resource
English
Modeling Bridge Deterioration with Markov Chains
British Library Online Contents | 1992
|Forecasting delays on railway sections
British Library Conference Proceedings | 2002
|Reliability-Based Processing of Markov Chains for Modeling Pavement Network Deterioration
Online Contents | 1996
|Reliability-Based Processing of Markov Chains for Modeling Pavement Network Deterioration
Transportation Research Record | 2019
|Reliability-Based Processing of Markov Chains for Modeling Pavement Network Deterioration
Transportation Research Record | 1996
|