Validation and improvement of rear-end traffic conflict model for safety evaluation of coordinated signal systems

This study involved validation of a previous study (5) undertaken to study the effect of signal coordination on intersection safety. It was aimed to validate two statistical models developed in the previous study for prediction for rear-end conflicts. These models were function of number of stops at signalized intersection approaches. It was also aimed to study the effect of other promising roadway and traffic variables which could affect rear-end conflicts at signalized intersection approaches. The stops reduction curve and platoon dispersion factor values in TRANSYT-7F package were calibrated for the Traffic conditions of Eastern province of Saudi Arabia in two different previous studies (2, 5). In this study it was also aimed to test whether these calibrated TRANSYT-7F parameters could be used to obtain equivalent number of field stops which is the required input for the two statistical models developed previously for prediction of rear-end conflicts.; It was found from this study that the models developed previously for prediction of rear-end conflicts were valid for use in the Eastern Province of Saudi Arabia. It was also found that apart from number of stops, average approach speed is also a significant parameter affecting the rear-end conflicts. Thereby a new model was built with number of stops and average approach speed as independent variables. This newly built model when compared with the previously built models showed a better fit to an independent data set. It was also found from this study that the calibrated stops reduction curve and default platoon dispersion factor values in TRANSYT-7F were adequate in obtaining equivalent number of field stops.