An Empirical Study of Bicyclists' Turning Behaviour at Signalized Intersections

dc.contributor.authorFraser, Adam
dc.date.accessioned2018-08-23T15:36:46Z
dc.date.available2018-08-23T15:36:46Z
dc.date.issued2018-08-23
dc.date.submitted2018
dc.description.abstractEfforts to increase cycling mode share have seen some success in North America, though challenges persist due to real and perceived safety issues. Of particular concern are left turns at signalized intersections. Left turns can be particularly challenging to traverse and often leave cyclists feeling unsafe, especially those who are less experienced. To reduce conflict and enhance safe left-turn maneuvering, the City of Philadelphia, Pennsylvania has initiated a pilot study for the installation of two-phase left turn bike boxes. This thesis investigates how the installation of two-phase left turn bike boxes influence left-turning behaviour at signalized intersections. A literature review found few studies that demonstrate the benefits of two-phase left turn bike boxes, and generally few studies that document left turn behaviour in a North American context. Similarly, few studies differentiate between signal control infractions and road space infractions. The approach used a before and after video analysis of five unique left-turning scenarios of installation of two-phase left turn bike boxes. A novel method of defining a series of left turn maneuvers was applied to analyze how these turns are conducted in the before and after stages. The method considers road positioning in the approach and departure, as well as the manner in which the bicyclists maneuvered through the intersection. The video footage also produced sufficient data to investigate general cyclist behaviour regarding road space positioning (using proper lanes, sidewalk riding, switching in between) and red light running behaviour for all travel directions of the 6,786 observed cyclists at the study signalized intersections. Through adapting classifications from literature, road space positioning was grouped into three categories: vehicular behaviour, opportunistic behaviour, and pedestrian conflict behaviour. For the red light running behaviour, mean gap times were captured for select travel directions. The research found that red light running rates were highly correlated with mean gap time in cross traffic (R2 = 0.95) and that left turns at signalized intersections produce the most unpredictable behaviour relative to through or right turn movements. The study also found that improved predictability in behaviour of left-turning cyclists is possible with two-phase left turn bike boxes, though understanding treatment context is necessary to see behavioural changes. Cyclists tend to favour directness and reduced delay over predictability and law compliance. Use of two-phase turn boxes occur when cyclists desire to follow road rules and prioritize reducing conflict with other road users over directness and minimizing delay.en
dc.identifier.urihttp://hdl.handle.net/10012/13650
dc.language.isoenen
dc.pendingfalse
dc.publisherUniversity of Waterlooen
dc.subjectcyclingen
dc.subjectbehaviouren
dc.subjectsignalized intersectionsen
dc.subjecttwo-phase left turn bike boxen
dc.subjectred light runningen
dc.titleAn Empirical Study of Bicyclists' Turning Behaviour at Signalized Intersectionsen
dc.typeMaster Thesisen
uws-etd.degreeMaster of Applied Scienceen
uws-etd.degree.departmentCivil and Environmental Engineeringen
uws-etd.degree.disciplineCivil Engineeringen
uws-etd.degree.grantorUniversity of Waterlooen
uws.contributor.advisorCasello, Jeffrey
uws.contributor.affiliation1Faculty of Engineeringen
uws.peerReviewStatusUnrevieweden
uws.published.cityWaterlooen
uws.published.countryCanadaen
uws.published.provinceOntarioen
uws.scholarLevelGraduateen
uws.typeOfResourceTexten

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Fraser_Adam.pdf
Size:
13.89 MB
Format:
Adobe Portable Document Format
Description:

License bundle

Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
6.08 KB
Format:
Item-specific license agreed upon to submission
Description: