Artificial Neural Networks and Logit Models for Traffic Safety Analysis of Toll Plazas
- 1 January 2002
- journal article
- conference paper
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 1784 (1), 115-125
- https://doi.org/10.3141/1784-15
Abstract
Little research has been conducted to evaluate the traffic safety of toll plazas and the impact of electronic toll collection (ETC) systems on highway safety, but analyses indicate that toll plazas do contribute to traffic accidents. Traffic safety issues related to toll plazas and ETC systems were studied using the 1999 and 2000 toll plaza traffic accident reports of the Central Florida expressway system. The analysis focused on accident location with respect to the plaza structure (before, at, after plaza) and driver injury severity (no injury, possible, evident, severe injuries). Two well-known artificial neural network (ANN) paradigms were investigated: the Multi-Layer Perceptron and Radial Basis Functions neural networks. The performance of ANN was compared with calibrated logit models. Modeling results showed that vehicles equipped with ETC devices, especially medium/heavy-duty trucks, have higher risk of being involved in accidents at the toll plaza structure. Also, main-line toll plazas have a higher percentage of accident occurrence upstream of the toll plaza. In terms of driver injury severity, ETC users have a higher chance of being injured when involved in an accident. Older drivers tend to have higher risk of experiencing more severe injuries than younger drivers. Female drivers have a higher chance of experiencing a severe injury than do male drivers.This publication has 13 references indexed in Scilit:
- The generalized nested logit modelTransportation Research Part B: Methodological, 2001
- Contribution to the Development of Guidelines for Toll Plaza DesignJournal of Transportation Engineering, 2001
- Predictions on surface finish in electrical discharge machining based upon neural network modelsInternational Journal of Machine Tools and Manufacture, 2001
- Development of Artificial Neural Network Models to Predict Driver Injury Severity in Traffic Accidents at Signalized IntersectionsTransportation Research Record: Journal of the Transportation Research Board, 2001
- New Model for Evaluation of Traffic Operations at Electronic Toll Collection PlazasTransportation Research Record: Journal of the Transportation Research Board, 2000
- Analysis of injury severity and vehicle occupancy in truck- and non-truck-involved accidentsAccident Analysis & Prevention, 1999
- Operational Benefits of Electronic Toll Collection: Case StudyJournal of Transportation Engineering, 1997
- Fast Learning in Networks of Locally-Tuned Processing UnitsNeural Computation, 1989
- Parallel Distributed ProcessingPublished by MIT Press ,1986
- Specification Tests for the Multinomial Logit ModelEconometrica, 1984