Simultaneous Calibration of Traffic Demand and Transport Network using Real-World Multi-Source Data

抄録

Transportation planning models play an important role in development of transportation systems. These models are calibrated before they are used. Usually, calibration cannot be done accurately in practice because, first, the collected data may be sparse and noisy, hence, insufficient for an accurate model calibration; second, there might be some inconsistency due to the disjoint calibration of the models. This study aims to address both issues by simultaneously calibrating demand and supply models using real-world multi-source data. To illustrate the performance of the method, it is applied to Tuen Mun network, which is a corridor network of Hong Kong transportation system. It is shown that the proposed approach can fit the observed data well (flow error 1.00% and time error 5.95%). Potentially, the procedure described in this study may facilitate policy makers with a structured and robust way of assessing traffic demand and improving supply services.