Predicting violations at gated active highway-railroad grade crossings

Abstract

Motorists deciding to drive around lowered gates at gated active highway-railroad grade crossings are likely one of the main contributors to many of the accidents that occur at these crossings. The objective of this study was to develop relationships between these violations and the geometric and operational characteristics that exist at gated active highway-railroad grade crossings. Data was collected using a remote video recording system. Train speed, warning time length, impedance, exposure, AADT per lane, number of highway lanes, false activations, number of tracks, warning time device type (constant versus fixed-distance), and adequacy of sight distance were the variables investigated in this study. The data was analyzed using both a site-specific approach and an arrival-specific approach. These approaches investigated the influence of average site characteristics and the specific conditions present during each train-vehicle arrival, respectively. It was found that the frequency of violations before the gates were fully lowered is almost exclusively a function of the exposure. Train speed and warning time length were found to be the variables that most influenced violations occurring after the gates were fully lowered, but before the train arrived. Variables that also significantly impacted the frequency and likelihood of this type of violation included false activations, number of tracks, and the variability in warning time (the presence of constant or fixed-distance warning time devices). Violations occurring after the train passed through the crossing occurred only on rare occasions. It appeared that violations of this type were influenced by longer than average train delays and delays due to the gate arms remaining down for an were developed to predict the number of violations expected during a 24hour period. It was recommended that a similar modeling procedure be used as a surrogate to accident prediction models to estimate the degree of hazard presented by a given highway-railroad grade crossing. It also was recommended that constant warning time devices be used whenever feasible with priority given to crossings with multiple tracks and/or average train speeds below 35 km/hr.