Application of Crowdsourced Travel Data in Identifying Potential Opportunities for Transportation Demand Management

Abstract

Traffic congestion is a major concern that many urban regions contend with. Since the 1970s, transport practitioners have sought out different ways to alleviate the negative impacts of congestion. The concept of Transportation Demand Management (TDM) was born out of this era as a means to address congestion through demand-side approaches that encourage adjustments in travel behaviour. Given its broad mandate, a wide variety of policies and programs fall within the realm of TDM. To assist in the selection of appropriate TDM strategies prior to implementation, several TDM evaluation models have been developed. However, many analysis techniques require data from past examples of TDM strategy implementation, which are at times difficult to obtain, and would not be available for newly developed and untested policy and program approaches. This thesis builds on the techniques used in past models to present a new model framework that can be used in the identification and evaluation of opportunities to apply TDM strategies. Employing a pivot-point logit model, the model relies only on quantifications of traveller utility through generalized cost calculations. Crowdsourced travel data collected from the Google Maps web-mapping application has been used as the primary source of data behind the generalized costs. The model was applied to a set of test corridors in the Greater Toronto and Hamilton Area (GTHA) which represent major cross regional travel flows. Results of the model were found to be directionally consistent with known conditions of the transportation system in the study area, though adjustments are needed to the model coefficients in order to more accurately reflect the magnitude of behavioural change to be expected from travellers.