International Journal of Applied Science and Engineering 2010. 8, 1: 11-17 Int. J. Appl. Sci. Eng., 2010. 8, 1 11 Application of Prediction Techniques to Road Safety in De- veloping Countries Jamal Al-Matawah * , a and Khair Jadaan b a Department of Civil Engineering, College of Technological Studies in Kuwait b Department of Civil Engineering, University of Jordan Abstract: The dramatic increase in vehicle travel in developing countries calls for the effective introduction of features that reduce traffic accidents. An important piece of information for such an introduction lies in the prediction of accidents and their fatalities, which is addressed in this paper. Smeed’s model was originally developed for the prediction of traffic fatalities in both de- veloped and developing countries. More reliable prediction models are developed for a number of Arab countries, producing a much less absolute percentage error than those of Smeed’s model. Regression analysis was applied to time-series data in the studied countries, producing an abso- lute percentage error as low as 7.67 for Saudi Arabia and 12.17 for Kuwait. An accident predic- tion model that relates accident frequency in Kuwait to various contributory factors is developed using the Generalized Linear Modelling (GLM) technique. The final model shows that age, na- tionality, aggressive driver behaviour, dangerous offences, perception of effectiveness of en- forcement, marital status, speed, and experience are the main contributory factors that lead to accident involvement. Keywords: road safety; prediction models; developing countries. * Corresponding author; e-mail: jamaln1@hotmail.com Accepted for Publication: September 7, 2010 © 2010 Chaoyang University of Technology, ISSN 1727-2394 1. Introduction Road traffic accidents and their resulting fatalities may be regarded as a growing social and economic problem, especially in devel- oping countries where the resources are lim- ited. The World Health Organization has pre- dicted that traffic fatalities will be the third leading cause of death worldwide by 2020 [1]. The effects of some of the contributing factors to traffic fatalities have been studied and rela- tionships for predicting these fatalities have been developed by Haight [2-8]. Yet, these relationships produced somewhat large devia- tions between the expected and the observed fatalities. These deviations were greatest in developing countries and the need arises for a more realistic relationship to predict road traffic fatalities with greater accuracy. In ad- dition, these relationships failed to incorpo- rate many significant contributory factors. Attempts to produce prediction models for traffic fatalities avoiding the above-mentioned pitfalls are discussed in this paper. The study uses a regression analysis of time-series fatal- ity data for the development and testing of the model for the statistics available from the UAE, Jordan and Qatar. The Generalized lin- ear model (GLM) technique is also used to develop a model that incorporates various significant contributory factors.