Origin-destination (OD) matrices are essential in-puts to dynamic traffic assignment and traffic simulation models and important tools for transportation planning. In this work we propose the use of detailed macroscopic traffic dynamics for the static OD matrix estimation problem. In contrast with most literature methodologies that consider average link flows over the time horizon under study, we utilise traffic counts in small intervals and employ a signalised path-based cell transmission model to accurately capture traffic dynamics and associate OD demands with link counts. In this setting the OD matrix estimation problem is formulated as a nonlinear mathematical program. The proposed methodology works well under both free-flow and congested conditions, as well as cases where loop detectors exist only on a subset of links in the network.
Estimating the origin-destination matrix using link count observations and traffic flow dynamics
2022-10-08
684416 byte
Conference paper
Electronic Resource
English
Novel Algorithm for Estimating Dynamic Origin-Destination Matrix in Traffic Networks
British Library Online Contents | 2003
|