The Effects of Train Passes on Dwell Time Delays in Sweden

Abstract

Railway traffic is growing, resulting in a highly interconnected train network. Due to the interdependence between trains’ activities, a better understanding of train passes and their effects can ensure dispatching decisions made have minimum risk of delays. The impacts of train pass on dwell time delays were investigated using historical Swedish railway operation data. Three scenarios were considered by combining the scheduled and actual operations: passes that happened as scheduled, unscheduled passes that happened in operation, and scheduled passes that were cancelled. A logistic regression model was used to explore the effects of these passes on delays. The findings show that train passes rarely occurred as scheduled, more frequently they are cancelled or unscheduled. This implies that some adjustments are required to assure the timetable’s feasibility. This study also found that the odds of delays for the cancelled pass was about 9.80 times lower than scheduled pass but 2.6 times more often for an unscheduled pass than a scheduled pass. The different types of train passes were quantified using an odds ratio to make comparisons easier for dispatching decision-making. The approach used in this study can be extended to other types of train movements, such as the meeting of trains, as well as other delay-influencing factors.