Identifying the Underlying Causes of Transit Ridership
In 2017, bus ridership in the United States was at its lowest point since the American Public Transportation Association started keeping track in 1965. To understand what causes this unprecedented decline, I have analyzed ridership at different levels of aggregation. For the TCRP study J-11, I modeled transit ridership based on the amount of service and population at the regional level with colleagues from Georgia Tech. In a separate research project, I am analyzing data from Automated Passenger Counts at 50,000 bus stops in five cities. To explain changes in ridership, I use population and demographic data from the census, amount of service data from GTFS, and ride-hailing data. This research identifies ridership trends on a highly specific spatial and temporal scale and allows to quantify the impact of gentrification, service allocation policies, transit mode cannibalism, and ride-hailing competition. For these projects, I am supervising three graduate students who specialized in GIS, statistics, and data science.
• Ederer, D., Berrebi, S., Diffee, C., Gibbs, T., Watkins, K., (2019) Comparing Transit Agency Peer Groups Using Cluster Analysis. Transportation Research Record. (Forthcoming)
• Berrebi, S., Diffee, C., Watkins, K., (2019) TCRP report J-11 Task 28: Analysis of Recent
Transit Ridership Trends. Transportation Research Board, Washington, DC. (Forthcoming)
A Dynamic Holding Method to Avoid Bus Bunching on High-Frequency Transit Routes: From Theory to Practice
I have developed a new dispatching method to stabilize headways and stop bus bunching. The method uses real-time vehicle information to replace schedules on high-frequency transit routes. In a series of simulation experiments, I found that the proposed holding method dispatched transit vehicles with more stable headways and used less holding time than other methods used in practice and recommended in the literature. I established research partnerships with transit agencies and successfully implemented the holding method on the Atlanta Streetcar, the VIA Bus Rapid Transit Primo Route in San Antonio, TX, and the Georgia Tech Red Stinger Route. I developed an open-source platform to compute holding times and trained operators to receive holding instructions from the software instead of their regular schedule. The method helped reduce bus bunching and passenger waiting time; it is currently being considered for permanent implementation.
• Berrebi, S. J., Crudden, O., S., & Watkins, K. E. (2018). Translating research to practice: Implementing real-time control on high-frequency transit routes. Transportation Research Part A: Policy and Practice.
• Berrebi, S. J., Hans, E., Chiabaut, N., Laval, J. A., Leclercq, L., & Watkins, K. E. (2017). Comparing bus holding methods with and without real-time predictions. Transportation Research Part C: Emerging Technologies.
• Berrebi, S. J., Watkins, K. E., & Laval, J. A. (2015). A real-time bus dispatching policy to minimize passenger wait on a high frequency route. Transportation Research Part B: Methodological. Special Issue: Optimization of Urban Transportation Service Networks
Minutes Matter: A Guide to Bus Transit Service Reliability, Transit Cooperative Research Program Report A-42
The TCRP Report A-42 is a guide for transit agencies to diagnose and manage bus transit service reliability. Based on the literature and on a survey of 150 transit agencies, I have drafted a report on factors of unreliability and possible treatments. I categorized factors of reliability by type and discussed their effect on customer experience and operations. I reported on cost and effectiveness of available treatments to address causes of reliability. The report will be published in early 2019.
• Danaher, A., Wensley, J., Watkins, K., Dunham, A., Berrebi, S., Connor, M., Queen, C., Orosz, T. (2019) TCRP report A-42: Minutes Matter: A Guide To Bus Transit Service Reliability Transportation Research Board, Washington, DC, (Forthcoming)