Applied geospatial modelling of the relationship between mode share, rail transit infrastructure and urban structure

摘要

This thesis presents the development and application of a suite of models to examine transit-orientation – measured primarily as mode share – of metropolitan areas. Transit-orientation in urban planning relates to empirical or qualitative relationships between urban structure and use of public transport (‘transit’) and to development that encourages transit use. Studies are usually at a local level, or have utilised overly simple parameters such as population density. Mode share modelling in transport planning is commonly based on behaviourally-based logistic models. The research described here has used geospatial and regression techniques to estimate transit-orientation based on demographics, urban structure, transit infrastructure and transit service at a detailed (travel zone) metropolitan-wide level. The core of this comprises a two stage logistic model suite to predict car ownership and rail, transit and active mode share and travel. The research has also included improvements to an associated road traffic assignment model used to measure the car travel impacts. The models were developed using data from Sydney, Australia, but the approach has international application. The thesis includes a Sydney case study, applied in the context of a paradigm of deferred transit investment common in cities with post World War automobile dominance. The models are applied to backcast and forecast scenarios to measure demand and operational effects, and to calculate economic sustainability impacts. The case study recognises the path dependence and resultant long term lock-in of the Sydney case, and offers comment on the current Sydney transport plan.