published_or_final_version / Transport Policy and Planning / Master / Master of Arts in Transport Policy and Planning

The Greater Toronto and Hamilton Area faces numerous transportation challenges now and in the future: congestion, population growth, and an inadequate public transit network. The metropolitan region has also changed in form in past decades, shifting from a monocentric to a polycentric region, further compounding the challenges. Currently, the public transit service is delivered by 9 different agencies comprised of 6 municipal providers, 2 sub-regional providers, and 1 regional provider. A region possessing a multiplicity of agencies suggests an overabundance of jurisdictional borders - borders that can potentially restrict travel across them. This thesis seeks to determine the impact of the presence or the omission of jurisdictional borders on transit patrons. A comparative approach is employed to investigate the benefits and costs to patrons and agencies through greater integration of specific origin-destination (OD) pairs. The chosen methods selects OD pairs that are known to be transit competitive, possess a high travel demand, and cross a transit-jurisdictional border. The relationship between transportation and land-use is relied upon to select clusters of dense employment or population, called activity centres, where public transit is known to compete well with the private auto. The travel demand between these centres is obtained using the 2006 Transportation Tomorrow Survey and the current optimal transit routing is determined using Google Trip Planner. Three OD pairs are selected that possess the most onerous transfers, a proxy for poor integration. Another three OD pairs are selected that possess seamless or no transfers using a variety of modes. In both cases, the existing transit routing is compared to an alternate routing to understand the benefits achieved through inter-jurisdictional integration; the first compares existing trips to improved inter-jurisdictional routes while the second compares existing trips to exclusively intra-jurisdictional routes. Through identification of 40 employment and 29 population activity centres in the region, and the acquiring of travel demand between them, the six case study OD pairs are selected. The three OD pairs investigated, with onerous transfers, are comprised of trips between Brampton-Mississauga, Hamilton-Burlington, and Brampton-Toronto. The remaining inter-jurisdictional case study OD pairs are made up of three different modes: conventional bus, express bus, and regional rail. They comprise trips between Toronto-York Region, Brampton-Mississauga, and Mississauga-Toronto respectively. This study finds that in all cases, the routes with greater integration reduce total travel time and the generalized cost to patrons. Additionally, the penalty due to transferring is reduced through integration implying a current barrier existing at some jurisdictional borders. For the agencies, the cost of delivering the suggested inter-jurisdictional service varies dramatically. The costs are translated into a quantity of additional patrons necessary to justify the operation investment while maintaining the current revenue/cost ratio. These findings provide insight into the current transit network. Promoting integration throughout the network will help attract new riders as the generalized cost of travel is reduced. Also, when inter-jurisdictional connections are made, such as in the case of the Brampton-Mississauga Zum service, the beneficiaries of that service are widespread and not limited to the corridor in which the service operates.

This study evaluated the influence of changes in public transit routes and hubs on the use and behaviour of people in and adjacent to a public square in a medium-sized southern Ontario city. Pre and post transit change data were obtained using behaviour observation and user interviews over a two-week period. Qualitative data was collected through investigator observations, intercept interviews, and field notes. The removal of the transit hub from the square resulted in 50 percent fewer people post change, although the duration of time spent by users in the square increased. There was also a shift in the type of user. The results indicate that transit hub changes can influence the number and types of users in a public square as well as the duration of use. The implications are that public transit design is an important factor in the planning and design of downtown civic space.

No description available.

This thesis proposes a framework that combines a mode choice model with a large scaled agent-based multimodal traffic microsimulation. Both components are discussed with respect to their development as separate entities. The mode choice model uses a formulation that explicitly considers latent modal captivity despite using conventional travel survey data. An existing multimodal microsimulation traffic assignment model used in the study area is enhanced and partially calibrated for use with the MATSIM traffic assignment tool. Both of the components are then tested independently in terms of statistical and behavioral validity and a conceptual procedure to test the implications of the mode choice model on mode switching behaviour within the traffic assignment model is presented. Other applications of both the travel assignment model and mode choice model are discussed. In order for the framework to become operational, further development with respect to the traffic assignment model is required.

This thesis proposes a framework that combines a mode choice model with a large scaled agent-based multimodal traffic microsimulation. Both components are discussed with respect to their development as separate entities. The mode choice model uses a formulation that explicitly considers latent modal captivity despite using conventional travel survey data. An existing multimodal microsimulation traffic assignment model used in the study area is enhanced and partially calibrated for use with the MATSIM traffic assignment tool. Both of the components are then tested independently in terms of statistical and behavioral validity and a conceptual procedure to test the implications of the mode choice model on mode switching behaviour within the traffic assignment model is presented. Other applications of both the travel assignment model and mode choice model are discussed. In order for the framework to become operational, further development with respect to the traffic assignment model is required.

No description available.

Until recently, transit data lacked a common data format that could be used to share and integrate information among multiple agencies. In 2005, however, Google worked with Tri-Met in Oregon to create the General Transit Feed Specification (GTFS), an open data format now used by all transit agencies that participate in Google Maps. GTFS feeds contain data for scheduled transit service including stop and route locations, schedules and fare information. The broad adoption of GTFS by transit agencies has made it a de facto standard. Those agencies using it are able to participate in a host of traveler services designed for GTFS, most notably transit trip planners. Still, analysts have not widely used GTFS as a data source for transit planning because of the newness of the technology. The objectives of this project are to demonstrate that GTFS feeds are an efficient data source for calculating key transit service metrics and to evaluate the validity of GTFS feeds as a data source. To demonstrate GTFS feeds’ analytic potential, the author created a tool called GTFS Reader, which imports GTFS feeds into a database using open-source products. GTFS Reader also includes a series of queries that calculate metrics like headways, route lengths and stop-spacing. To evaluate the validity of GTFS feeds, annual vehicle revenue miles and hours from the National Transit Database (NTD) are compared to the calculated values from agencies whose GTFS feeds are available. The key finding of this work is that well-formed GTFS feeds are an accurate representation of transit networks and that the method of aggregation presented in this research can be used to effectively and efficiently calculate metrics for transit agencies. The daily aggregation method is more accurate than the weekly aggregation method, both introduced in this thesis, but practical limitations on processing time favor the weekly method. The reliability of GTFS feed data for smaller agencies is less conclusive than that of larger agencies because of discrepancies found in smaller agencies when their GTFS-generated metrics were compared to those in the NTD. This research will be of particular interest to transit and policy analysts, researchers and transit planners.

The relationship between monthly transit pass ownership and home-work trip mode choice in the Greater Toronto Area was explored using a logit formulation. A Nested Logit model was specified with the primary mode (non-motorised, auto, transit, or auto access to transit) in the upper level and the pass ownership in the lower transit nests. Inclusive value parameters confirm the chosen structure, and a range of policy sensitive, statistically significant parameters having expected signs and magnitudes reveal differences between socio-economic characteristics of pass holders and non-pass holders. In particular, pass holders are less likely to possess a driver’s license or automobile than non-pass holders, implying that passes can be thought of as complementary mobility tools. Cost parameters, which include differentiated pricing for pass and non-pass holders, result in plausible in-vehicle values of time of approximately $31/hour, $33/hour and $8/hour for auto, auto access to transit, and transit all way, respectively.

The relationship between monthly transit pass ownership and home-work trip mode choice in the Greater Toronto Area was explored using a logit formulation. A Nested Logit model was specified with the primary mode (non-motorised, auto, transit, or auto access to transit) in the upper level and the pass ownership in the lower transit nests. Inclusive value parameters confirm the chosen structure, and a range of policy sensitive, statistically significant parameters having expected signs and magnitudes reveal differences between socio-economic characteristics of pass holders and non-pass holders. In particular, pass holders are less likely to possess a driver’s license or automobile than non-pass holders, implying that passes can be thought of as complementary mobility tools. Cost parameters, which include differentiated pricing for pass and non-pass holders, result in plausible in-vehicle values of time of approximately $31/hour, $33/hour and $8/hour for auto, auto access to transit, and transit all way, respectively.