Cycling network discontinuities and their effects on cyclist behaviour and safety

Cycling is widely considered to be the riskiest mode of transport since collisions with vehicles are more likely to result in serious injuries or even death than other road users except pedestrians.Given its many environmental and social benefits, cities are encouraging cycling as an affordable mode of transport and are expanding their cycling infrastructure.While cities are aiming to increase cycling mode share, their alarming safety statistics have compelled transportation researchers and planners as well as city officials and decision makers to invest resources in designing, implementing and improving the cycling network to safely accommodate cyclists.Improving the cycling network to increase cycling mode share and safety relies on detailed quantitative information on performance indicators.One of the dimensions of cycling network analysis is its continuity.Network continuity provides a set of possible routes that are connected and accessible to all road users.However, cycling networks are usually implemented on the already existing road network, which results in locations where there are changes in road and cycling network characteristics.These changes are interruptions in the cycling network, also referred to as discontinuities.Despite the many infrastructural, traffic and environmental measures studied in cycling literature, the systematic definition of cycling network discontinuities has been overlooked.In this dissertation, four research gaps have been identified in cyclist behaviour and safety literature as well as cycling network performance studies: the definition and presentation of cycling network discontinuity indicators, the effects of road lighting discontinuities on nighttime cyclist safety, the cyclist behaviour and safety analysis at discontinuity locations in the cycling network.To address the first gap, different categories of discontinuity measures are proposed and defined, where there are 1) intrinsic changes in the cycling network (end of cycling facility, change in cycling facility type, change in cycling facility width, change in cycling facility location on road, change in pavement condition, change in road lighting, change in road grade, closure/rerouting of cycling facility due to construction or maintenance), 2) changes to the road network (change in road class, change in number of road lanes, intersections) and traffic characteristics (change in traffic volume, change in traffic speed), and 3) other changes (driveways, bus stops, parking allowed on road).Moreover, an automated methodology is proposed that can be applied to any area using its georeferenced cycling network data to identify and quantify infrastructural discontinuities along the cycling network.The methodology is applied to a case study of four North American cities, to compare their discontinuity levels in a uniform and systematic way.The areas under study are viii ranked based on two cycling network discontinuity indicators from worst to best as: Portland, Vancouver, Washington D.C., and Montréal.To close the second research gap, a methodology is proposed to perform a nighttime road lighting audit collecting illuminance measurements at an intersection or link level to identify locations with discontinuous lighting.Past studies using illuminance measurements relied on inconsistent and cumbersome methods for collecting data.The proposed methodology in this dissertation provides a uniform methodology that can be applied to any area to collect nighttime illuminance data.The methodology is applied to case study locations in Montréal and a statistical analysis of historical accident data showed that locations with higher illuminance levels are associated with an increase in the chance of a severe cyclist accident at nighttime.The third research gap addressed in the dissertation is the analysis of cyclist behaviour at locations where there is a cycling network discontinuity.Adopting the proposed methodology, two pairs of discontinuity and control sites are selected in Montréal.The in-depth analysis of cyclist behaviour requires large amounts of microscopic data, i.e. road user trajectories at a fine temporal scale.To this end, computer vision techniques and trajectory clustering methods are applied to video data.Hence, an automated video analysis tool is adopted to extract road user trajectories and cluster similar cyclist trajectories to compare the movements of cyclists traveling through the cycling network discontinuity compared to a control site.The methodology identifies valuable microscopic information on cyclist movements that can be applied to any location to evaluate cyclist behaviour.Results from this study indicated a higher variation in number of cyclist maneuver and speeds at locations of cycling network discontinuity compared to their control site.Finally, the safety implications of discontinuity locations on cyclists is studied using surrogate measures of safety (SMoS) adopting a probabilistic method (PSMoS) of predicting future positions of road users.The two pairs of discontinuity and control sites from Montréal are further analysed in a case study.Time-to-collision (TTC) is computed for cyclist-vehicle interactions and summarised per cyclist maneuver to identify the specific risky maneuver cyclists make at discontinuity locations.This novel movement-based PSMoS approach has not been adopted in literature for the safety analysis of all cyclist maneuvers.The approach is a useful tool to identify the exact movements that influence the safety of cyclists.Results show that the discontinuity locations have a higher number of unsafe cyclist motion patterns compared to their control sites.At the discontinuity site where the physically separated cycling facility location changes from one side of the road to another, cyclists ix who originate and end in the cycling facility on the opposite ends of the intersection have the lowest TTC.In Summary, this dissertation closes the gaps in literature by defining and proposing cycling network discontinuity indicators and evaluating their effects on cycling network performance, cyclist behaviour, and safety.Results of all the studies in this dissertation confirm the importance of including discontinuity indicators in the planning and evaluation of cyclist networks.The lack of these indicators in current planning and evaluation stages provides a partial image of the quality of a cycling network and cycling experience and leaves transportation departments unable to fully address the effect of discontinuities on cyclists.The information obtained from the cyclist behavior and safety studies will help planners and city officials make better informed decisions by improving the infrastructural design of the cycling network discontinuity locations to eliminate unsafe movements and safely accommodate all cyclists.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.