Train delay analysis and prediction based on big data fusion

Abstract

Despite the fact that punctuality is an advantage of rail travel compared with other long-distance transport, train delays often occur. For this study, a three-month dataset of weather, train delay and train schedule records was collected and analysed in order to understand the patterns of train delays and to predict train delay time. We found that in severe weather train delays are determined mainly by the type of bad weather, while in ordinary weather the delays are determined mainly by the historical delay time and delay frequency of trains. Identifying the factors closely correlated with train delays, we developed a machine-learning model to predict the delay time of each train at each station. The prediction model is useful not only for passengers wishing to plan their journeys more reliably, but also for railway operators developing more efficient train schedules and more reasonable pricing plans.