ACTIVE TRAVEL CHOICES IN CONSTRAINED URBAN ENVIRONMENTS - A NEW ANALYTICAL FRAMEWORK FOR INTRODUCING RADICAL INTERVENTIONS IN URBAN AND SUBURBAN CYCLY

The main research areas are: infrastructure and urban systems and statistics and applied probability. The related research areas are: built environment, energy efficiency and end use energy demand, engineering design.
Across the UK there is a growing need to re-think about how urban spaces best serve the needs to improve the quality of life for people and enhance the eco-system services. This often leads to the need to re-design urban infrastructure and spaces in both high growth areas (to respond to pressures of expansion) and declining areas (to shrink the urban footprint). In some cities, there have already been attempts to initiate interventions, e.g. in areas of high traffic congestion, there have been proposals to close main urban roads during peak times, to remove in-town car parking etc. However, most of such initiatives have not fared well in practice. The lack of success in practical implementation reflects the fact that the effects of the radical interventions are not well understood in fundamental theories of the choices made by citizens and businesses. For instance, existing transport and land use models, which have been developed in past EPSRC sustainable built environment programmes, can no longer accurately represent the factors that influence current travel behaviour. We hypothesise that by utilising new data sources that have emerged in the last decade, we are able to build a new theory and model that captures these new influences. Objectives of the Research is to develop a new modelling framework that account for the new factors of influence on travel behaviour. This research project is focused on cycling which is the least well understood among all means of transport for journeys to work. The specific objectives is to 1 assemble all new data sources that support the modelling of cycle travel to work; 2 develop appropriate theoretical models; 3 empirically estimate and test the models; 4 demonstrate the validity and use of the models in real world case studies. The approach for data gathering is data fusion through combining traditional transport modelling data and new data sources including crowd-sourced, more granular cycling routes and social media data on the environmental characteristics of the routes. In particular, as new framework of mapping the routes is developed using the Open Street Map as the base, along with a wide range of data on land use characteristics, the ease and suitability of cycling for different age groups, and the socio-demographic profile of the neighbourhoods and social media crowd presence along each route.
The approach for modelling addresses the new issues of the numerical and categorical information that emerge having incorporated the wider data sources. This has led to the development of new statistical and machine-learning models which provide appropriate empirical robustness and prediction capabilities. Two ongoing case studies are carried out. 1 - investigates highly variable rates of cycling across the UK in order to understand the factors that influence these variations, among 36 UK cities - the first output of this research has been published in the top journal of the field, Transport Policy. 2- changes to travel behaviour observed between 2001 and 2011 in the Cambridge City region. During this timeframe, major transport measures were implemented in the city, such as the widespread closure of central streets and a significant programme of investment in cycling infrastructure. Novel engineering and/or physical sciences content - the novel contents of the research are 1 a brand new database framework 2 new methods to estimate statistical models that represent observed choices and provide robust predictions for future interventions in planning and design; 3 empirical estimates across 36 UK cities and the temporal dynamics 2001-2011 for Greater Cambridge; 4 new ways to assess and validate the models that set a new benchmark for future studies in this field.