Comparative evaluation of heuristic optimization methods in urban arterial network optimization

Heuristic optimization methods have been widely applied in the engineering applications that are known to be extremely difficult to find an optimal solution using traditional mathematical approaches. Examples of such optimizations in transportation problems includes: congestion pricing, dynamic traffic assignment, developing traffic signal timing plans, etc. Studies have shown mixed results on the performances of various heuristic optimization methods. Obviously, the performance depends largely on the nature of problems, complexity of solution space, etc. This paper presents an evaluation of a few selected heuristic optimization methods (genetic algorithm (GA), harmony search (HS), and OptQuest) applied to solving a transportation optimization problem of an urban arterial network. The study results showed that the control parameters in GA and HS significantly affected the performance, and with properly determined parameters, GA outperformed HS and OptQuest. In addition, the best control parameters in GA and HS are robust to increased traffic conditions.