Traffic Performance Measurement Using High-Resolution Data – The SMART-SIGNAL System

1. Traffic Performance Measurement Using High-Resolution Data –The SMART-SIGNAL System Dr. Henry Liu Department of Civil Engineering University of Minnesota — Twin Cities March 1, 2011

2. An automatic and continuous data collection system from existing traffic signals A performance measurement system for intersection queue length and arterial travel time, especially under congested traffic conditions A performance tuning system for optimization of traffic signal parameters SMART-Signal: Systematic Monitoring of Arterial Road Traffic Signals

3. SMART-SIGNAL System Architecture Signal Detectors Signal Detectors Signal Detectors ... ... Local Data Collection Unit Local Data Collection Unit Local Data Collection Unit Data Server at Master Cabinet FIELD DSL Communication Database Preprocessed Data Performance Measures TMC Direct/Internet Access Internet Access Traffic Engineers Road Travelers Monitor Diagnosis Fine-tuning Travel Decision USERS

4. Terminal Box Data Collection DAC

5. Serial Port #1 SDLC Ethernet Serial Port #2 Power Plug-and-Play Device for TS2 Cabinet

6. http://signal.umn.edu

7. 11 intersections on France Ave. in Bloomington (March 07 – June 09) 6 intersections on TH55 in Golden Valley (Feb. 08 – Sept. 09) 3 intersections on PCD in Eden Prairie (Current) 6 intersections in the City of Pasadena, California (Iteris, Spring 2011, On-going) 14 intersections on TH13 (Spring 2011, Expected) 10 intersections on TH55 (Spring 2011, Expected) SMART-Signal Implementation Sites

8. Prairie Center Dr. and Technology Dr. (Eden Prairie) Feb 4, 2010

9. Queue length estimation Delay, Level of Services, number of stops Identification of oversaturated conditions Oversaturation Severity Index (OSI) Travel time estimation Personal trip delay, number of stops, carbon footprint on travel Developed Algorithms

10. Vehicle classification / speed estimation Both arterial and freeway applications Queue length / travel time prediction Modeling of Arterial Traffic Flow Fine-tuning signal timing parameters Offsets, Green Splits, “Break Points” for time of day … Algorithms Under Development

11. Lessons Learned from SMART-SIGNAL • Although traffic is traditionally modeled as “continuous flow”, traffic, after all, is discrete. • Measuring traffic flow parameters using the data collected at the individual vehicle level • Don’t aggregate data before useful information being derived • Technological advances support such data collection at affordable prices

12. Publications • Liu, H. and Ma, W., (2009) A virtual vehicle probe model for time-dependent travel time estimation on signalized arterials, Transp. Res. Part C, 17(1), 11-26. • Liu, H., Wu, X., Ma, W., and Hu, H., (2009) Real-Time queue length estimation for congested signalized intersections, Transp. Res. Part C, 17(4), 412-427. • Wu, X., Liu, H. and Gettman, D. (2010) Identification of Oversaturated Intersections Using High-Resolution Traffic Signal Data, Transp. Res. Part C, 18(4), 626-638. • Wu, X., Liu, H, and Geroliminis, N. (2010) An Empirical Analysis on the Arterial Fundamental Diagram, Transp. Res. Part B, 45, 255-266.

13. Acknowledgements

14. THANK YOU!Dr. Henry Liu612-625-6347henryliu@umn.eduSMART-Signal Web Site: http://signal.umn.edu