Go to  Advanced Search

Sustainable road safety : developing community-based macro-level collision prediction models of increased bicycle use in the Regional District of Central Okanagan

Show full item record

Files in this item

Files Size Format Description   View
ubc_2013_spring_feng_wei.pdf 4.980Mb Adobe Portable Document Format   View/Open
 
Title: Sustainable road safety : developing community-based macro-level collision prediction models of increased bicycle use in the Regional District of Central Okanagan
Author: Wei, Feng
Degree Master of Applied Science - MASc
Program Civil Engineering
Copyright Date: 2012
Publicly Available in cIRcle 2012-11-01
Abstract: Since the energy crisis, global warming, transportation congestion, and road safety issues, governments and individuals are more willing to use sustainable transportation modes to support the development of compact neighbourhoods. Bicycling, as one of the most effective modes for short-distance trips, is widely encouraged. However, as vulnerable road users (VRUs), cyclists are more likely to be injured when they are involved in collisions. In order to prevent bicycle collisions from occurring, it is necessary to conduct proactive evaluation of cyclists’ road safety. Community-based, macro-level collision prediction models, as prospective empirical tools to evaluate and predict road safety, have been suggested in reactive road safety applications and road safety planning. This research reviews previous studies on road safety and bicycle use and summarizes different regression methods for collision prediction model (CPM) development. On the basis of insights gained in the literature review, community-based, macro-level CPMs related to bicycle use were developed using the generalized linear regression (GLM), zero-inflated count regression (ZIC), geographically weighted regression (GWR), and full Bayesian (FB) methods, based on data from the Regional District of Central Okanagan (RDCO) of British Columbia, Canada. The statistical associations of total/severe/bike-vehicle collisions and their neighbourhood traits, which were derived from these model results, are reasonable. In the reactive road safety application of macro-level black spot study, bike-vehicle collision prone zones (CPZs) were identified and ranked in the RDCO with the developed GLM and FB collision prediction models and preliminary diagnoses and remedies for these CPZs are suggested. Finally, data gaps for macro-level CPM development were identified. In order to improve data quality and linkage, several suggestions about how to build an integrated data warehouse for vulnerable road user collisions are proposed.
URI: http://hdl.handle.net/2429/43565
Scholarly Level: Graduate

This item appears in the following Collection(s)

Show full item record

All items in cIRcle are protected by copyright, with all rights reserved.

UBC Library
1961 East Mall
Vancouver, B.C.
Canada V6T 1Z1
Tel: 604-822-6375
Fax: 604-822-3893