Intelligent Signal Coordination on Congested Networks Using Parallel Micro Genetic Algorithms

This paper presents the application of Parallel Genetic Algorithms (GA) on signal coordination for networks with congested intersections. When a serial GA is applied to solve traffic control problems, its performance in terms of computation time diminishes as the size of signal networks increases, or the duration of congestion lengthens. This paper uses master-slave parallel micro GA (PMGA) to solve signal coordination, formulated as a dynamic optimization problem. The elapsed time per generation reduces as the number of processors involved increases. When the number of processors equals the number of individuals, for the network described in this paper, the master-slave PMGA reaches the highest speed-up. Although the master-slave PMGA provides a lower bound speed-up and its efficiency degrades as the number of processors increases, this study demonstrates the potential benefits of using parallel GA in solving signal coordination problems.