Application of the HSM predictive method at four-leg signalized intersections. Case study: Lima, Peru

The Highway Safety Manual (HSM) recommends the predictive method to characterize intersections quantitatively in terms of the number of accidents that may occur. The time period, geometric design, type of traffic control and traffic volumes are used as variables. Considering that the predictive model was originally developed with information from the United States road network, a local calibration factor is needed to adjust the results to the reality of each country. In Peru, road traffic accidents are one of the most critical problems affecting society. More than 80,000 traffic accidents have been registered every year since 2010. Of these events, 70% occur in the Metropolitan area of Lima and 34.9% of these are reported at intersections. This considerable proportion is logical because this type of infrastructure generates the highest number of conflict points between vehicles and pedestrians. Hence, this article proposes the predictive method as a tool to improve road safety management and to make the best decisions in order to reduce the frequency and severity of traffic accidents. For this purpose, the predictive method was applied to 2 four-leg signalized intersections located in one of the main roads of Lima; both of them presents a high traffic volume of vehicles, passengers and pedestrians. Once applied the predictive method, a calibration factor of 0.841 was obtained. The result shows a dispersion of 15.89% compared to the accidents observed at the intersections within the study area. The calibration factor allowed us to adjust the HSM predictive method to a Peruvian reality. With this information, road safety strategies can be formulated to reduce the traffic accident rate at signalized intersections.