Zhong-Ren Peng, Ph.D., Department of Urban & Regional Planning
Title: Understanding Transit Service Reliability and Service Restoration
Abstract: This study investigates important issues in transit service reliability, namely large service gaps and bus bunching. We try to understand the spatial and temporal patterns of large service gaps along the route, whether and to what degree large gaps propagate down to the road, and how to identify and establish conditions that indicate a large service gap is imminent. We use the automatic vehicle location (AVL) data from Route 20 at the Chicago Transit Authority (CTA) as a case study. We found that bus bunching from one time point did not propagate to the next time point, while the large gap at one time point led to a large gap in the next time point but did not propagate to the following time points. This indicates service restoration strategies conducted by field supervisors could and did help to restore services. We further found that the most common condition that forms a large gap is when one bus runs slower, while the previous bus runs quicker than the schedule running time between two time points. We further demonstrated that conditions or threshold could be identified that could lead to large gaps by analyzing the AVL data. We recommend the implementation of a warning system to notify control center personnel and the operators that a large service gap is likely to occur so that preventative actions can be taken by service centers, field supervisors and operators. The findings in this study can also have important implications for intelligent signal priority design.