A model and genetic algorithm for area-wide intersection signal optimization under user equilibrium traffic

Abstract

The use of signal control systems to reduce traffic movement conflicts at intersections often affects operational efficiency of an urban street network. This study introduces a new method for area-wide traffic signal timing optimization under user equilibrium traffic. The optimization model is formulated as a multi-dimensional search problem aimed to achieve minimized product of the total travel time associated with urban street network and the variance of travel time for unit distance of travel. A genetic algorithm is developed to derive the model solution. A simulation control protocol embedded in PARAMICS software tool capable of conducting area-wide microsimulation is adopted to design the logic frame and function module of the area-wide traffic signal control system. Computational experiment is performed for model application on the Central Business District of Nanjing, China for validation. The results show that mobility improvements are achieved after applying the proposed model along with the genetic algorithm for area-wide signal timing optimization, assessed by extended capacity ratio, and reductions in through and turning movement delays, as well as average and variance of travel time for unit distance of travel.