10 King's College Rd
Toronto, ON M5S 3G8
Pedestrian simulation models are becoming more frequently relied upon by researchers and practitioners when designing or evaluating transit stations, event venues, and other large facilities. However, the computational performance of these models can vary greatly and their relative accuracy is generally unexplored, leaving users with little guidance when selecting a model for their needs.
Here, models based on the Social Forces, Optimal Steps, and static graph concepts are coded into MassMotion, then calibrated against observed pedestrian flow data using Genetic Algorithms. Using two transit station scenarios, each model’s ability to match observed and increased volume pedestrian flows is explored and their relative speeds are compared, generally showing significant speed advantages over the default MassMotion model, but also indicating that the graph and Optimal Steps models are not sufficiently sensitive to congestion.
Additionally, this work explores the behaviour of these models within various facility elements, including corridors, corners, and open spaces.
Greg Hoy is an MASc student in the Department of Civil Engineering at the University of Toronto. Bridging the gap between research and practice, his work focuses on pedestrian microsimulation in the context of public transit and is supervised by Professor Amer Shalaby at University of Toronto and Erin Morrow at Arup. Greg received his BASc in Engineering Science, specializing in Infrastructure Engineering, from the University of Toronto in 2015. When not hard at work, Greg enjoys exploring the city on foot, listening to music, and catching up on the latest Netflix shows.
This seminar is presented by the University of Toronto Institute of Transportation Engineers (ITE) Student Chapter.