eprintid: 4715
rev_number: 5
eprint_status: archive
userid: 1
dir: disk0/00/00/47/15
datestamp: 2024-08-21 13:47:10
lastmod: 2024-08-21 13:47:10
status_changed: 2024-08-21 13:47:10
type: thesis
metadata_visibility: show
sword_depositor: 1
creators_name: Berge, Luuk ten
creators_id: S4924606
creators_email: luuk.t.berge@gmail.com
title: Evaluating the Predictive Ability of Least-Cost Analysis for Bicycle Route Choices
ispublished: unpub
full_text_status: public
abstract: The Netherlands is facing increasing congestion and traffic accidents on cycle paths, necessitating a better understanding of cyclists' behaviour for effective policy and urban design interventions. This study aims to provide insights into cyclists' behaviour through a literature review that identifies factors such as infrastructure and land use that influence route choice. These factors are incorporated into a Weighted Raster Network (WRM) for Groningen, using regression coefficients from a previous study in Enschede. Data from TalkingBikes was used, resulting in 50 GPS trips after a careful selection. Three routes were compared: Observed Route (OBR) based on GPS data, Shortest Path (SHP) representing the minimum possible distance and Least Cost Path (LCP) representing the minimum cost according to the WRM. This study uniquely examines cycling routes using an LCP, unlike previous studies that often compare SHP routes, which do not take into account the factors considered by an LCP. Visual and cost comparisons show that cyclists do not strictly follow LCP or SHP routes, with the OBR showing significant variation. Differences in total costs between routes suggest that context-dependent weights, such as the over-emphasis on segregated cycle path (72.9% of LCP length overlaps with segregated cycle lanes), do not fit well with Groningen context. This study highlights the complexity of accurately predicting cycling behaviour. In order to determine whether an LCP can effectively predict route choice, specific weight coefficients need to be developed for the city of Groningen, and a mixed methods approach should be used to gain deeper insights into cyclists' route choices through interviews.
date: 2024
pages: 49
thesis_type: bachelor
degree_programme: TP
tutors_name: Vos, D.
tutors_organization: Fac. Ruimtelijke wetenschappen, Basiseenheid Culturele Geografie
tutors_email: D.Vos@rug.nl
security: public
keywords_local: Cycling behaviour
keywords_local: Route choice modelling
keywords_local: Least-cost path analysis
keywords_local: Infrastructure factors
keywords_local: Land-use factors
language_iso: en
date_issued: 2024-08-21
comment: Niet vertrouwelijk. (wanneer ik 'nee' selecteer wordt dit veld verplicht in te vullen, selecteer ik 'ja' dan is het niet meer verplicht in te vullen)
citation:   Berge, Luuk ten  (2024) Evaluating the Predictive Ability of Least-Cost Analysis for Bicycle Route Choices.  Bachelor thesis.   
document_url: https://frw.studenttheses.ub.rug.nl/4715/1/BachelorProjectLMtenBergeS4924606.pdf