Heuristic algorithms for simulation-based dynamic traffic assignment

Abstract

This study reviews and compares two heuristic algorithms: methods of successive averages and route swapping, which are commonly used in simulation-based dynamic traffic assignment models. A new hybrid algorithm that has the advantages of both algorithms is introduced. As finding a unique solution cannot be guaranteed, the concept of solution reliability is suggested for the formulation of convergence criterion. An example network is used to illustrate the method.

Keywords:

• dynamic traffic assignment
• method of successive averages
• route swapping
• convergence criterion

Acknowledgements

This research is supported by two research grants HKU 7132/03E and No. 717306 from the Hong Kong Research Grant Council of the Hong Kong Special Administrative Region. Two anonymous referees provided useful comments that resulted in improvements on an earlier draft of the article.