Spatial analysis of bicycle use and accident risks for cyclists

Most developed countries nowadays face environmental, health and mobility problems as a consequence of widespread car use. Policies are now being reappraised in favour of more sustainable modes of transport. In particular, bicycle use holds the potential to provide a ‘green’ and healthy alternative to car commuting. There are however still important barriers that discourage people cycling. This thesis aims at identifying some of the main factors that influence cycle commuting and cycling accidents. Identifying such factors would in turn provide greater support to enable policy makers developing supportive environmental conditions for cycling. In the first part of this thesis, we examine which factors influence the spatial variation of bicycle use for commuting to work at the level of the municipalities in Belgium. Special attention is paid to bicycle-specific factors and spatial econometric methods are used to account for the presence of spatial effects in the data. The second part of this thesis examines which factors are associated with cycling accidents in Brussels. Spatial point pattern methods extended to networks are used to compare the ‘locational tendencies’ of cycling accidents officially registered by the police with those that are unregistered. An innovative case-control approach, based on a rigorous sampling design of controls and an exhaustive data collection of spatial factors, is also proposed to allow modelling the risk of cycling accident along the Brussels’ road network. This thesis not only provides sound recommendations helping planners and policy makers to encourage bicycle use, but it also offers new research directions for pinpointing locations where accidents are more likely to occur.