EVALUATION OF A PREDICTION ALGORITHM FOR A REAL-TIME RAMP CONTROL SYSTEM. VOLUME 3. (FREEWAY AND RAMP REAL TIME FORECASTING)
Volume III evaluates a statistical pattern recognition-based predictive algorithm that was tested on-line with the ramp metering computer system of the Washington State Department of Transportation (WSDOT) in Spring 1990. The purpose of this algorithm was to forecast bottleneck formation 1 or 2 minutes in advance of its occurrence, and to adjust ramp metering rates in order to avoid or decrease bottleneck formation. The evaluation of this algorithm was conducted by applying multiple linear regression techniques to traffic volumes and occupancy time series data collected by the WSDOT's Transportation System Management Center (TSMC). In addition to the statistical analysis, the accuracy of the algorithm's predicting ability was evaluated using the computer generated prediction messages printed during the study period.
- Record URL:
- See also Volume 2, PB94-106549 and Volume 4, PB94-106564. Sponsored by Washington State Department of Transportation, Olympia, TransNow, Seattle, and Federal Highway Administration, Olympia, WA. Also available in set of 5 reports, PB94-106523.
Washington State Transportation CenterUniversity of Washington, 1107 NE 45th Street, Suite 535
Seattle, WA United States 98105
Washington, DC United States 20590
- Nihan, N L
- Cabrera-Gonzalez, I
- Publication Date: 1993-6
- Pagination: 99 p.
- TRT Terms: Algorithms; Bottlenecks; Mathematical prediction; Multiple regression analysis; Pattern recognition systems; Ramp metering; Real time control; Regression analysis
- Old TRIS Terms: Multiple regression; Real time operations
- Subject Areas: Data and Information Technology; Highways; Operations and Traffic Management; I73: Traffic Control;
- Accession Number: 00666390
- Record Type: Publication
- Report/Paper Numbers: WA-RD-288.3, TNW-93-05.3
- Files: TRIS, USDOT
- Created Date: Sep 19 1994 12:00AM