Document Type
Technical Report
Publication Date
11-23-2021
Keywords
Proactive management, recurring congestion, non-recurring congestion
Digital Object Identifier (DOI)
https://doi.org/10.5038/CUTR-NICR-RR-2022-4-2
Abstract
This report details the development and implementation of a method to proactively detect and mitigate congestion on freeways and arterials. To accomplish this goal, the research utilizes data fusion between conventional sources, such as radar detectors and traditional probe-based data, with newer sources, such as Bluetooth and connected vehicle (CV) data, to identify conditions that signal impending congestion. Data-driven and signal processing techniques are explored and developed to produce an algorithm that relies on near-real- or real-time traffic measurements capable of generating predictions proactively, using complex and often subtle factors that trigger congestion. The algorithm is validated and calibrated using traffic and other relevant data from comparable roadway facilities located in Florida and Texas. The algorithm is robust enough to function on both traditional and CV-based datasets and provides distinction between four intensity levels of congestion. The algorithm is applied within a microsimulation model to test the effectiveness of congestion mitigation strategies ranging from speed harmonization to dynamic rerouting, implemented individually and simultaneously. Performance measurement benchmarks show how these strategies prove to be effective in proactively reducing recurring and non-recurring congestion while providing additional safety benefits. Finally, this project demonstrates the clear advantage of using CV-based travel time estimates to calibrate microsimulation models over fixed point-based derivations of travel time from spot speeds.
Scholar Commons Citation
Concas, Sisinnio; Kamrani, Mohsen; and Kummetha, Vishal C., "Proactive Congestion Management" (2021). Research Reports. 5.
https://digitalcommons.usf.edu/cutr_nicr/5
Policy Brief
