Modelling of Traffic Safety and Operation in Urban Networks Using DSTPN

Abstract

Abstract The objective of this work is to develop a simulation model for the analysis of the performance of urban traffic networks using Deterministic Stochastic Timed Petri Nets (DSTPN). Factors considered in the model include the physical characteristics of an intersection, such as the number of lanes in each direction, and the turning classification of each lane. The arrival process of vehicles at the intersection, the departure process from the conflict area, the traffic signal control strategy used in controlling traffic flow and the dynamic behaviour of traffic at the phase transition instant are the main parts of the model. Operational analysis is used to compute one or more traffic performance Measures of Effectiveness (MOE) such as queue length, traffic flow at a specific lane in certain period of time and system throughput. These MOE are used to evaluate the operation of signalized intersections and to compare the simulated results with the field ones. A Petri net simulator was used to prove the functionality and the applicability of the developed model. The developed urban traffic network model was also used to assess safety measures of a signalized urban area. Red light violations and its interaction with the signal timing plan, and traffic conflict technique were used to quantify the safety of signalized intersections,. The model was tested on a real network consisted of seven signalized intersections in the middle of the city of “Braunschweig”’ and the results are promising specially with the short computational time.