Utilizing Big Data for Performance Evaluation of Traffic Signals in Tennessee

No Thumbnail Available
Meleby, Piro A.J.D
Journal Title
Journal ISSN
Volume Title
Middle Tennessee State University
Congestion on roadways has led to increased travel times, fuel costs and emissions. Presented as techniques to mitigate congestion, construction of more lanes and/or roads is expensive while optimization efforts including performance evaluations of traffic signals have been documented to reduce congestion through shorter travel times. This thesis details a data-driven methodology for performance evaluations using third-party segmented probe vehicle data for Tennessee. Intersections were formed by matching the names and geographical information of inbound segments. The speed and travel time data for each segment at an intersection were used to develop metrics for a ranking formula that output a numerical value from 0-10. The results were appended to an online database visible via a webpage with search and sort capability for local traffic agencies to prioritize retiming. The ranking formula results were harsher in comparison to expert rankings from Murfreesboro and Franklin cities. Additionally, testing was conducted on an industry-standard traffic signal controller for data logging and networking purposes. An ATSPM server and website was set up to receive and display performance data from the controller in a lab to emulate a field intersection.