Public transit is becoming an increasing important field of study to combat global issues such as traffic congestion and climate change. Accurate simulation of public transit is therefore likewise vital, as it is an important tool for understanding potential impacts of public transit policies. The research presented in this thesis describes the implementation of a multimodal, dynamic, agent-based supply-side simulation model of public transit implemented in the open-source platform MATSim for the city of Toronto. Transit schedule data was converted from Google Transit Feed Specification (GTFS) and map-matched to a region-wide road network to obtain a congestion-based multimodal assignment for transit. Volume-based results from the assignment showed under-prediction of subway volumes and slight over-prediction of bus volumes, but were generally comparable with static EMME/3 assignment for the same data. Travel time analysis indicated that further calibration of network specification is needed.
| Identifer | oai:union.ndltd.org:TORONTO/oai:tspace.library.utoronto.ca:1807/33277 |
| Date | 20 November 2012 |
| Creators | Kucirek, Peter |
| Contributors | Miller, Eric |
| Source Sets | University of Toronto |
| Language | en_ca |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0019 seconds