Novel approach for OD estimation based on observed turning proportions and Bluetooth structural information: Proof of the concept

This research proposes a novel approach for estimating OD matrices based on observed turning proportions and Bluetooth structural knowledge. The key idea is to relax the need for explicit assignment-based models and exploit the observations from big traffic data. In the proposed methodology, observed turning proportions implicitly consider path assignment. The methodology is tested on synthetic data from a small network with sufficient route choice options. Numerical experiments are conducted considering different Bluetooth connectivity to OD pairs. The findings from the experimental study indicate that the quality of OD estimates improves as the knowledge of Bluetooth OD increases. The proposed research does not depend on traditional bi-level formulation and is computationally faster. The methodology is adaptive for any data source that can provide OD structural information. The testing of the proposed methodology on a large-scale network is part of the ongoing study