Accounting for composite travel time distributions within a traffic stream in determining Level-of-Service

Abstract

This paper presents a procedure for assigning multiple Level-of-Service (LOS) ratings to travel time data sets containing multiple (composite) distributions. Different LOS modes within a mixed data set, the travel time data were fitted to multiple gamma-type distributions using an Expectation Maximization (EM) iterative process. The EM process was enhanced with a Monte Carlo style method, wherein the EM process was run 100 times with different random starting values and the best fit according to an R-squared value was taken. The number of underlying distributions was determined by fitting 1 to 5 distributions and using the Akaike Information Criterion to determine which number of fits maximized the information content of the fitted function. The resulting final posterior probabilities were then used to separate the data into their respective distributions. Reported here are the results of applying this procedure to travel time data collected in the Metro Atlanta area. It is believed that this method can provide enhanced LOS information, especially when the data contain multiple overlapping travel time distributions. This multiple LOS rating method is not intended to replace the current method, instead it is being developed as a supplementary tool to provide more detailed LOS information, for example providing a more detailed view that captures the underlying performance of each subgroup in addition to a single aggregate LOS measure.

Keywords

Travel time distribution expectation maximization iterative process Monte Carlo simulation

Get full access to this article

View all access options for this article.

References

Ji and

H.M.

Zhang , Travel Time Distributions on Urban Streets: Their Estimation with a Hierarchical Bayesian Mixture Model and Application to Traffic Analysis Using High-Resolution Bus Probe Data, In Transportation Research Board 92nd Annual Meeting, Transportation Research Board of the National Academies, Washington, D.C. 2013.

Yang ,

Yun and

Yang , Travel Time Distributions under Interrupted Flow and Its Application to Travel Time Reliability, In Transportation Research Board 93rd Annual Meeting, Transportation Research Board of the National Academies, Washington, D.C. 2014.

Park ,

Rakha and

Guo , Calibration Issues for Multistate Model of Travel Time Reliability. In Transportation Research Record: Journal of the Transportation Research Board, No. 2188, Transportation Research Board of the National Academies, Washington, D.C. 2010, pp. 74–84.

R.S.

Chalumuri and*

Asakura , Modelling travel time distribution under various uncertainties on Hanshin expressway of Japan, European Transport Research Review 6 (2012), 85–92.

Colberg

, Suh

, Anderson

, Zinner

, Guin

, Hunter

and Guensler

, Lane Bias Issues in Work Zone Travel Time Measurement and Reporting, Transportation Research Record: Journal of the Transportation Research Board, No. 2458, pp. 78–87, 2014.

Benaglia ,

Chauveau ,

Hunter and

Young , Mixtools: An R Package for analyzing finite mixture models, In Journal of Statistical Software 32(6) (2009).

Urban Street Facilities. Highway Capacity Manual 2010. Transportation Research Board of the National Academies. Washington, D.C. 2010, pp. 16-1-16-18.

Henclewood

, Suh

, Rodgers

, Fujimoto

and Hunter

, A calibration procedure for increasing the accuracy of Microscopic Traffic Simulation Models, Simulation: Transactions of the Society for Modeling and Simulation International 93(1) (2017), 35–47.

Zinner , A Methodology for Using Bluetooth to Measure Real-time Work Zone Travel Time. Master's Thesis. Georgia Institute of Technology, 2012.

10.

Box , Arterial Travel Time Data Collection Using Bluetooth Technology. Master's Thesis. Georgia Institute of Technology, 2011.

11.

Wasson ,

Haseman and

Bullock , Real-Time Measurement of Work Zone Travel Time Delay and Evaluation Metrics Using Bluetooth Probe Tracking, in Transportation Research Board 89th Annual Meeting, 2010.

12.

Colberg , Investigating the Ability of Automated License Plate Recognition Camera Systems to Measure Travel Times in Work Zones. Master's Thesis. Georgia Institute of Technology, 2013.

13.

Urban Street Segments. Highway Capacity Manual 2010. Transportation Research Board of the National Academies. Washington, D.C., 2010, pp. 17-1-17-44.

14.

Urban Street Segments: Supplemental. Highway Capacity Manual 2010. Transportation Research Board of the National Academies. Washington, D.C. 2010, pp. 3035-30-38.