Modelagem probabilística do tempo de resposta de motoristas em interseções semaforizadas no VISSIM

Abstract

The main objective of this work is to implement, in the microscopic simulator VISSIM, a probabilistic model for the driver response time in signalized intersections, and to evaluate the effects of this model on traffic performance. Based on data from a signalized intersection located in Fortaleza, statistical analyses of the response time were conducted by vehicle position in the queue, traffic lane, vehicle type and movement type. Theses analyses led to the modeling of the response time by a probability distribution, log-normal, which was implemented in VISSIM by an external algorithm programmed in Python. To assess the importance of this implementation, the results of delay, queue size and capacity were compared with the results of the simulations performed by the default modeling of VISSIM. The simulations conducted with the probabilistic response time model resulted in higher delays and queue lengths, and in lower capacities than those obtained by the default modeling of the software.