Browsing by Subject "cycling"
Now showing items 1-2 of 2
-
(2021)Collisions and near accidents between pedestrians and cyclists can result in serious injuries and death but have received limited academic attention. Using an online survey, this thesis aimed to increase knowledge of such events, assess the sense of safety of pedestrians and cyclists in traffic as well as identify safety-related constraints to the uptake of walking and cycling with practice theory. Practice theory considers human behaviour to be guided via participation in established social practices constituted by interconnected elements of meaning, material and competence. As such, this thesis contributes to debates concerning barriers to walking and cycling from a safety perspective. The survey was directed to Finnish cities with over 100,000 population and asked frequent pedestrians and cyclists to report details of collisions and near accidents between pedestrians and cyclists that they had experienced in the previous three years. Additionally, the survey asked questions concerning respondents’ sense of safety in traffic when walking or cycling. Survey data was analysed with chi-square tests of independence and ordinal logistic regression. Constraints to the uptake of cycling and walking and ways to overcome them were identified with a practice theory analysis. This involved examining the implications of survey results for the elements constituting the practices, their interrelations and how the practices influenced each other. According to the results, near accidents are roughly 50 times more frequent than collisions. Only 16 respondents had experienced a collision, whereas roughly a third had experienced at least one near accident. Additionally, shared paths were associated with more collisions and near accidents compared to separated spaces, and respondents felt less safe and less willing to travel on them compared to separated paths. The most common type of collision and near accident involved both road users travelling in the same direction. Constraints to cycling and walking were found to surface from meanings of danger associated particularly with shared infrastructure, a material element of the practices. These issues are evidenced by a high near accident frequency, low sense of safety and low willingness to travel on shared spaces. In addition, these issues were exacerbated by a lack of competences concerning space sharing, resulting in poor rapport and respect between pedestrians and cyclists. Significant effects regarding sense of safety were detected between pedestrians and cyclists and across age and genders with ordinal logistic regression, suggesting variance in how different groups experience meanings of danger. Intervening in the material element of the practices by preferring the provision of spatially separated infrastructure was considered to have potential to help overcome these constraints due to their associated safety benefits and respondents’ more favourable position toward them. In addition, working to develop a shared code of conduct for travel on shared environments could further mitigate constraints. Overcoming these constraints could assist the promotion of active travel and help improve the sustainability of transport while improving traffic safety and increasing physical activity.
-
(2017)Transportation in cities is facing the challenges of congestion and environmental impact caused by the increase in traffic flows. These issues can be reduced by promoting more sustainable transport modes, such as cycling. To increase its modal share, cycling has to be an attractive and competitive choice compared to other travel modes. Digital Geography Lab in University of Helsinki has developed comparable measures for modelling accessibility with different travel modes in Helsinki region. However, cycling is missing from the data because it has been previously modelled with simplistic assumptions of constant travel speed. Little research has been carried out to assess the applicability of this assumption. The main objective of this thesis is to develop a more realistic GIS model for calculating optimal routes and travel times of cycling in Helsinki region taking into account the feasibility of the model. Other objectives are to find out what factors affect cyclists' travel speed and can the environmental factors be used as impedances in the travel time model, what kind of spatial differences the cycling speeds have, and how realistic it is to model cyclists' travel times with constant speed on a regional scale. According to previous research, among the various things affecting cycling some of the main environmental factors are slope, junctions and traffic lights. The effects of these factors to cycling speeds in Helsinki region were analysed based on individual cycling routes and on a route and segment level from the whole data with linear regression models. GPS data of cycling was collected from volunteers who had been tracking their cycling in Helsinki region with mobile sports applications. Basic background information of the cyclists was also collected to analyse the variations in speed between different background variables. Road network for cycling and walking by Helsinki Region Transport was used as the modelling network. A GIS-based map-matching method for the cycling GPS data was developed by applying a method developed for map-matching GPS data of cars. Slope was calculated for route segments using NLS 2 meter digital elevation model and the traffic light information was derived from Digiroad. Python scripts used in modelling are available on GitHub. The cycling speeds vary by cycling frequency: cyclists who stated to cycle almost every day of the week, 3-5 times a week, or a few times a week have median speeds of 24 km/h, 22 km/h and 18 km/h, respectively. Uphill slope and signalized junctions decelerate and downhill slopes accelerate cycling speeds on individual routes. Looking at the whole data, speed has a weak negative correlation between slope and different junction types. On a regional scale the effect of signalized junctions is the greatest, whereas uphill slope has the greatest effect on route-based mean speeds. The regression models do not explain the variation in cycling speeds very well (R2 ≈ 0.1) so a travel time model based on constant speeds corresponding to the different median speeds of frequent and less frequent cyclists was implemented on the network. Spatial examination shows that mean cycling speeds in parts of central Helsinki are 0.8 times slower than in rest of the area, so the cycling speeds of the model were slowed down on those segments. Slope, traffic lights and other junctions affect cycling speeds on an individual level but not on the regional scale. Based on model validation the travel times of the constant speed model correlate strongly with the real travel times of the GPS data. The model taking into account the slower parts of central Helsinki is marginally better but the difference is only slight and affecting only the routes going via the city centre. The difference in travel times caused by different constant speeds is much greater. Constant speed can hence be seen as an adequate assumption to model cyclists' travel times in Helsinki region but the personal and spatial differences in cycling speeds should be taken into account.
Now showing items 1-2 of 2