Application Genetic Algorithm in Optimizing Traffic Control

摘要

Urban Traffic Control in developing countries is always a never ending problem due to rapid motorization. Optimization of traffic control is one way to reduce this problem. In this experiment, Genetic Algorithm was adopted to optimize a traffic light and offset between intersections. The objective functions are minimizing delay at an intersection and maximize traffic progressive flows along an arterial road. The experiment was conducted in real time on a stretch of arterial road in Bangi, Malaysia which consists of 5 intersections. Traffic data for the input such as traffic flows, queue lengths and traffic speed are collected using video detection system. For this purpose, a video camera is mounted facing every approach in the study area. The digital images from the camera were analyzed in real time. The experience shows that the traffic control performance is improved up to 56% during off-peak hours and 34% during peak-hours.