Simulation modeling and variability assessment of delays at traffic signals

This paper presents a microscopic stochastic simulation model developed to emulate the traffic movement at signalized intersections and estimate vehicular delays including the acceleration/deceleration delay. The structure and overall logic of the model is discussed in this paper, and validation work using field data proved the model validity to represent real systems. The simulation model was applied to experiment 48 cases and evaluate the impact of the cycle length, approach speed, and degree of saturation on vehicular delays. The impact of each parameter was assessed and analyzed. It was found that the acceleration/deceleration delay component is directly proportional to-the cycle length and degree of saturation and inversely proportional to the approach speed. In addition, the impact of approach speed is not significant compared to that of the other two parameters. Furthermore, the ratio of total to stopped delays was also estimated and found to vary between 1.5 and 3.0. The relationship between simulated and the Highway Capacity Manual (HCM) 2000 model delays was analyzed for the 48 cases and represented by a linear regression model with R-2 of 94.6%. This relationship indicated that simulated delay is higher than the HCM delay by about 6%, and this may be attributed to the different methods applied in both models for calculating delay due to acceleration and deceleration at the signalized intersection.