Calibration of the Two-Fluid Model with Third-Party Crowdsourced Data - A Procedure and Evaluation
The two-fluid model is a well-established relation of network and corridor level trip time to the proportion of stopped and running vehicles. Historically, data required by the two-fluid model is obtained through chase-car experiments where vehicles in a network or corridor are followed. Recently, this technique is being supplemented or replaced by GPS probe data. Using a third-party GPS vendor’s (TomTom) crowdsourced, aggregated and processed data, a simple methodology is adapted from previous literature in order to obtain the two-fluid model parameters more efficiently. The advantage of this method over more conventional methods is that it does not need labor or time intensive procedures, or complex data cleaning methodologies. Statistical analysis reveals that the distribution of the stopped fraction of vehicles obtained from the method follow similar travel time behavior from other related studies such as right-skew and heavy tailing. Heteroscedasticity was observed, and controlled for by the use of weighted least-squares with differing weights. The resulting two-fluid model parameters are comparable to literature, and provide results relatively close to observed values of the running and average speed. This suggests that the methodology proposed can be a quick, and efficient tool for calibrating the two-fluid model for both research and practice.
-
Supplemental Notes:
- This paper was sponsored by TRB committee AHB45 Standing Committee on Traffic Flow Theory and Characteristics. Alternate title: Calibration of the Two-Fluid Model with Third-Party Crowdsourced Data: A Procedure and Evaluation
-
Authors:
- Manuel, Aaron
- Kattan, Lina
- Tahmasseby, Shahram
- de Barros, Alexandre
-
Conference:
- Transportation Research Board 97th Annual Meeting
- Location: Washington DC, United States
- Date: 2018-1-7 to 2018-1-11
- Date: 2018
Language
- English
Media Info
- Media Type: Digital/other
- Features: Figures; Maps; References; Tables;
- Pagination: 21p
Subject/Index Terms
- TRT Terms: Automatic data collection systems; Calibration; Cluster analysis; Crowdsourcing; Evaluation; Global Positioning System; Highway corridors; Least squares method; Operating speed; Regression analysis; Running speed; Stopping; Traffic models; Travel time
- Identifier Terms: TomTom
- Uncontrolled Terms: Two-fluid model
- Subject Areas: Data and Information Technology; Highways; Operations and Traffic Management; Vehicles and Equipment;
Filing Info
- Accession Number: 01660939
- Record Type: Publication
- Report/Paper Numbers: 18-05683
- Files: TRIS, TRB, ATRI
- Created Date: Feb 22 2018 9:18AM