As congestion worsens, the importance of rigorous methodologies to estimate travel time reliability increases. Exploiting fine-granularity transit GPS data, this research proposes a novel method to estimate travel time percentiles and confidence intervals. Novel transit reliability measures based on travel time percentiles are proposed to identify and rank low-performance hot spots; the proposed reliability measures can be utilized to distinguish peak-hour low performance from whole-day low performance. As a case study, the methodology is applied to a bus transit corridor in Portland, Oregon. Time–space speed profiles, heat maps, and visualizations are employed to highlight sections and intersections with high travel time variability and low transit performance. Segment and intersection travel time reliability are contrasted against analytical delay formulas at intersections—with positive results. If bus stop delays are removed, this methodology can also be applied to estimate regular traffic travel time variability.
Traffic and Transit Travel Time Reliability Indexes and Confidence Intervals
Novel Methodologies for the Corridor and Segment Levels
Transportation Research Record
Transportation Research Record: Journal of the Transportation Research Board ; 2649 , 1 ; 28-41
2017-01-01
Article (Journal)
Electronic Resource
English
Confidence intervals for real-time freeway travel
IEEE | 2003
|Transit Travel Time Reliability
Transportation Research Record | 2019
|Confidence Intervals for Real Time Freeway Travel Time Prediction
British Library Conference Proceedings | 2003
|Confidence intervals and the social value of travel time savings
British Library Conference Proceedings | 1995
|Transit network design based on travel time reliability
Online Contents | 2014
|