Deriving Operational Traffic Signal Performance Measures from Vehicle Trajectory Data

Operations-oriented traffic signal performance measures are important for identifying the need retiming to improve operations. Currently, most obtained from high-resolution controller event data, which provides information on an intersection-by-intersection basis and requires significant initial capital investment. Over 400 billion vehicle trajectory points generated each month in United States. This paper proposes using high-fidelity data produce such as: split failure, downstream blockage, quality of progression, as well traditional Highway Capacity Manual level service. Geo-fences created at specific signalized intersections filter waypoints that lie within boundaries. These then converted into trajectories relative intersection. A case study is presented summarizes eight-intersection corridor with four different timing plans over 160,000 1.4 million GPS samples collected during weekdays July 2019 between 5:00 a.m. 10:00 p.m. The concludes by commenting current probe penetration rates, indicating these techniques can be applied corridors annual average daily ~15,000 vehicles per day mainline approaches, discussing cloud-based implementation opportunities.