Prediction of Safety Opportunities Using Automated Traffic Signal Performance Measures
-
2025-06-01
-
Details
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Report
-
Corporate Publisher:
-
Abstract:This report presents the Smart Signal Performance Monitor (SSPM) system, developed to enhance intersection safety in Florida. The SSPM utilizes Automated Traffic Signal Performance Measures (ATSPM) to provide data-driven signalization recommendations. Raw ATSPM data is transformed into performance measures like Signal Phasing and Timing (SPaT) metrics, vehicle volume, conflicts, red-light running (RLR), and pedestrian activity. Algorithms were developed to recommend six key strategies: Yellow/Red Clearance Time Adjustment using a causal forest model; Protected vs. Permitted Left-Turn Phasing via gap analysis and volume estimation; Pedestrian Recall using a Beta-Binomial model and k-means clustering; and Leading Pedestrian Interval (LPI) with No Right Turn on Red (NRTOR) based on pedestrian-vehicle conflict propensity. The SSPM architecture features a Transform-Recommend-Rank (TRR) Server (for processing, recommendations, and ranking intersections by a composite safety score), a Database Server, and a Frontend Server with interactive dashboards. Tested with June 2024 data from 19 Seminole County intersections, the SSPM demonstrates an end-to-end pipeline for monitoring intersection safety and recommendations for signal strategies. Future work includes integrating live ATSPM feeds and expanding system coverage, which will provide FDOT with a network-wide proactive management tool for intersection safety.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:urn:sha-512:51b35d8c6d17f180c1f751f8380a36a251c9886458b79d2b8640c06df449d0d6bd1c34c8a5f36094921fb191a01a8f75029f293a5e334e3594fa43d5562da3b3
-
Download URL:
-
File Type: