Abstract
Railway level crossings in urban areas can impose substantial delays on road traffic and can produce poor safety outcomes when road users ignore warnings to avoid delays. Grade separation is a solution but comes at a substantial cost. This paper focuses on predicting the closure times of urban railway level crossings and is part of a broader research program examining the potential for Intelligent Transport Systems to reduce delays at urban railway level crossings. Using data generated from a simulation model, regression and neural network models are developed relating a range of explanatory variables to crossing closure time. The results highlight that if closure times are to be predicted more accurately, there is a need for improved real time data including train speed data and more accurate data on whether a particular train is to stop at a station adjacent to the crossing or run express through the level crossing.
