Traffic Collision Analysis Models: Review and Empirical Evaluation

Authors

  • Said M. Easa Ryerson University
  • Mohamad K. Hasan Kuwait University
  • Mohammad M. Hamed Jordan University of Science and Technology

DOI:

https://doi.org/10.34120/ajas.v12i3.599

Keywords:

Exponential models, Poisson model, Negative Binomial model, Multiple Linear models, Survival Models, Logit Model, Qualitative Data Analysis Models, Time Series Models, Traffic Collision Frequency

Abstract

Collision analysis models were classified in this paper into two categories: aggregate and disaggregate. A theoretical review of these models was presented, followed by an empirical (comparative) evaluation of multiple linear models, Poisson models, negative binomial models, and exponential models using three functional forms (additive, multiplicative, and quadratic). The evaluation results showed that the functional form of the model is very important for prediction accuracy. The Poisson model with a quadratic form provided the best accuracy, while multiple linear and exponential models provided the least accuracy.

Downloads

Download data is not yet available.

Author Biographies

Said M. Easa, Ryerson University

Ph.D. Transportation, University of California at Berkeley, California, U.S.A., 1982. Professor and Chair, Department of Civil Engineering, Ryerson University, Toronto, Ontario, Canada. Current research interests include Road safety, Highway geometric design, Human factors in transportation, and Traffic operations and management. He has authored and co-authored more than 300 publications, including 160 refereed-journal articles and book contributions. Dr. Easa's work received numerous national and international best-paper and lifetime achievement awards, including the 2001 Frank M. Masters Transportation Engineering Award from American Society of Civil Engineers in recognition of "out- standing contributions to ASCE and the transportation profession throughout his career."

Mohamad K. Hasan, Kuwait University

Ph.D. Transportation Systems, Texas A&M University, Texas, USA, 1991. Associate Professor, Department of Quantitative Methods and Information Systems, College of Business Administration, Kuwait University. His current research interests include: Network Optimization, Transportation Systems Analysis, Transportation Network Modeling, Freight Transportation Modeling, and Travel Demand Modeling, Logistics, Supply Chain Management, Production and Operations Management, Scheduling, Simulation, and Mathematical Programming.

Mohammad M. Hamed, Jordan University of Science and Technology

Ph.D. Transportation Planning. University of Washington, Seattle, USA, 1990. Professor of Transportation Planning, Department of Civil Engineering, Jordan University of Science and Technology, Irbid, Jordan and Director General of the Public Transport Regulatory Commission, Amman, Jordan. Current research interests include Travel Demand Modeling, Travel Activity-based analysis, Transportation Systems Analysis, Privatization of transport services, Public sector restructuring and reform, Public transportation demand modeling, Design of travelers information strategies, Monitoring public transportation services, and Public transportation tracking systems.

Downloads

Published

2005

How to Cite

Easa, S. M., Hasan, M. K., & Hamed, M. M. (2005). Traffic Collision Analysis Models: Review and Empirical Evaluation. Arab Journal of Administrative Sciences, 12(3), 473–499. https://doi.org/10.34120/ajas.v12i3.599

Issue

Vol. 12 No. 3 (2005)

Section

Operations Management