Estimation of origin-destination matrices from link traffic counts and user equilibrium link information

This dissertation addresses the problem of estimating Origin-Destination (O-D) trip matrices from link volume counts and an assumption of user equilibrium flows in transportation networks. Two major improvements are achieved relative to existing approaches. First, an optimal trip table solution search method is derived. Second, this model includes a sub-algorithm that provides good estimates of unobserved link flows. The optimal trip table solution search replaces a heuristic trip table correction function used in previous formulations. The feasibility of the volume estimation algorithm is demonstrated using a simple transportation network. The new algorithm's performance is demonstrated on a larger, empirical urban network.; The optimal trip table solution search method and the algorithm for estimating unobserved link flows are integrated into a single model. This integrated model has two advantages relative to conventional approaches. First, the optimal trip table solution search method provides advantages with respect to both computation efficiency, and the quality of results. Second, the algorithm for estimating unobserved link flows provides volume estimates that are approximately consistent with both observed flows and an assumption of user equilibrium conditions. These estimated link volumes improve the constraints associated with model, providing improved starting conditions for solution.; The individual and combined effects of the optimal trip table solution search method and the algorithm for estimating unobserved link volumes are systematically tested. Model performance is tracked in terms of the root mean square errors (RMSE) in predicted travel demands, and where appropriate, predicted linked volumes. These results indicate that the new model substantially outperforms existing approaches to estimating synthetic O-D matrices.