Single-level approach to estimate origin-destination matrix: exploiting turning proportions and partial OD flows

ABSTRACT

This research proposes a novel single-level approach to traffic counts-based OD estimation using observed turning proportions and partial OD flows. The key idea is to relax the need for explicit assignment-based bi-level models and exploit the observations from big traffic data. The observed turning proportions are used to estimate traffic counts; and the structural knowledge of partial OD flows is used to improve the quality of OD estimates. Numerical experiments considering different spatial coverage of partial OD from Bluetooth are conducted on two networks. The study findings indicated that the quality of estimates improve with higher percentage of partial OD information. The proposed approach is computationally faster and yielded slightly better results than the traditional bi-level. The methodology is not limited to data from Bluetooth but adaptive with other sources such as GPS, mobile phone, and travel surveys that can provide turning proportions and/or partial OD flows.

KEYWORDS:

• OD matrix estimation
• single-level
• turning proportions
• partial od flows
• od structure
• bluetooth
• brisbane

Acknowledgments

The authors are thankful to the Queensland Department of Transport and Main Roads (TMR), the Brisbane City Council (BCC), and the Queensland University of Technology for supporting this research. The conclusions of this paper reflect understandings of the authors, who are responsible for the accuracy of the research findings.

Disclosure of potential conflicts of interest

No potential conflict of interest was reported by the author(s).

Notes

1. An abridged version of this research was first presented for a toy network in Australasian Transport Research Forum (ATRF) but was not published (Behara, Bhaskar, and Chung 2018b).

2. Australian Bureau of Statistics (ASGS 2017) defines hierarchy of geographical areas into four statistical areas (SA): SA1 (200 to 800 persons), SA2 (3000 to 25,000 persons), SA3 (30,000 to 130,000 persons), and SA4 (a minimum of 100,000 individuals).

3. Studies suggest that the termination criterion can be either based on maximum relative change in the elements of estimated OD flows at successive iterations (Maher, Zhang, and Van Vliet 2001) or observed convergence for a fixed number of iterations (Bullejos, Barceló Bugeda, and Montero Mercadé 2014).

4. GSSI computes statistics on group of OD pairs belonging to same geographical windows. These windows are defined based on higher zonal level boundaries. Refer (Behara, Bhaskar, and Chung 2020b) for more details.