Transit Analytics Lab (TAL) 2023 Research Day

The Transit Analytics Lab (TAL) of the University of Toronto brings together transportation and technology researchers from across the University of Toronto, transit systems in the Golden Horseshoe area, and private sector technology providers and consultants.

Among its objectives are to: foster innovation in transit data-driven tools (analytics) using advanced methods of data science, machine learning, artificial intelligence, simulation and statistics; expose the professional community through knowledge transfer activities to advanced analytics; build collaborations with public and private organizations; and establish U of T as a national and international leader in transit data analytics.

TAL was launched in 2020 with the International Symposium on Automated Transit Data. Since then, TAL has organized workshops and has been involved in a number of research pursuits, many of which have practical applications.

The time is appropriate to host our annual TAL Research Day workshop that will provide a high-level overview of the many research activities being pursued by the TAL team; please see the Draft Program below.

The TAL Research Day will be held virtually and is free, though registration is required.

Agenda

9:00 a.m.         Introduction to the Transit Analytics Lab (TAL)

• • • Words of Welcome and Update on TAL Activities (Amer Shalaby)

9:15                  Transit Analytics to Support Planning (Moderator: Brendon Hemily)

• • • Constructing Origin-Destination Demand Matrix using Wi-Fi and AFC Gate Count Data: A Case Study of Toronto’s Subway Network (Diego Da Silva)
    • Trends in Toronto’s Transit Ridership Recovery: Insights from Subway Wi-Fi Records (Roger Chen)
    • Modelling On-Demand Transit Ridership (Alaa Itani)

10:15                  Break

10:30                 Operations Analytics to Improve Rail Performance (Moderator: Amer Shalaby)

• • • Impact of Subway Service Disruption on User Mobility: Analysis and Visualization Using Customer Facing Wi-Fi Data in Toronto (Aidan Grenville)
    • Short-term Prediction of Crowd Flows at Subway Stations using Wi-Fi Data and Graph Neural Networks Modelling (Diego Da Silva)
    • Mitigating Subway Station Overcrowding via Passenger Inflow Control (Chandler White)
    • SPUR: Modular, Data-Driven Mesoscopic Simulation Platform to Analyze Stochastic Railway Networks (Peter Lai)

11:30              Lunch Break

12:30 p.m.     Keynote: A Conversation with John Levin on Transit Data and Analytics

• John Levin, Director-Strategic Initiatives, Metro Transit (Minneapolis)

1:30                 Analytics to Support Bus Operations (Moderator: Brendon Hemily)

• • • Leveraging Large Language Models (LLMs) for Improving Public Transit Systems: An Exploration of GPT Models and State-of-the-Art Applications (Jiahao Wang)
    • Extraboard Operator Planning and Scheduling Under Uncertainty (Jilin Song)
    • Impacts of Transit Driver Advisory System with Space and Time Priorities on Route Performance (Kareem Othman)

2:30                Break

2:45                 Analytics to Support Planning and Deployment of Zero Emission Buses (ZEBs (Moderator: Amer Shalaby)

• • • Insights from the Research on ZEB Deployment (Diego Da Silva)
    • Optimization Model for Planning On-Route Charging Infrastructure and Schedules of ZEB Fleets (Lorna Licollari)
    • Data-Driven Prediction of e-Bus Battery Consumption Rates using Machine Learning (Kareem Othman)

3:45                 Wrap-Up

4:00                 End of Research Day

Download the TAL 2023 Research Day agenda.

Registration

Register on Eventbrite for this virtual event.

Free. All are welcome.

If any specific accommodations are needed, please contact mobilitynetwork@utoronto.ca. Requests should be made as early as possible.