| dc.description.abstract | The assumption in traditional route choice models focused on motorized vehicles is that people take the shortest available route to their destination. This assumption however, is not applicable to cyclists. For them, other factors might be more important than the shortest distance or shortest travel time. One of these factors is infrastructure. For some cyclists, the infrastructure may be the reason why a certain route is chosen over the shortest route. Understanding cyclist route choice and generating fitting cyclist route choice models is key to introduce new policies that improve the experience for current cyclists and incentivize more people to cycle. By getting more people to use their bicycle as their main mode of transport instead of their car, large reductions to the emission of greenhouses gasses can be achieved while simultaneously improving public health by a large margin.
In this research, GPS data collected by the B-riders program in the Dutch province of Noord-Brabant is used to find out if the route choice behaviour can be explained by infrastructural factors. This route choice behaviour was studied by comparing the difference between the observed routes and the shortest route in distance with the difference in infrastructural factors along these routes. Studies have shown that there are several infrastructural factors that influence cyclist route choice. These are the road type, type of road surface, quality of the road and intersections and traffic signal control. The data on these infrastructural factors was sourced from the Fietsersbond road network and was added to the routes by using several GIS methods.
When comparing the shortest routes to the observed routes, the observed routes were indeed significantly longer. All but one of the infrastructural factors were also significantly different between the two types of routes. However, when adding the amount of deviation and the difference in infrastructural factors to a multiple linear regression model, only a small amount of the difference between the two types of routes could be explained by the infrastructural factors. It turns out that the one infrastructural factor that significantly influences a cyclist route is traffic signal control at an intersection. This is in contrast with previous research on cyclist route choice, where cyclists preferred to cycle on separated bicycle paths and roads with a certain surface and road quality.
The fact that the multiple linear regression models and the infrastructural factors analysed in this research explained only a small amount of cyclist route choice behaviour can perhaps be used an indication that cyclist route behaviour should be researched using different (non) linear models. To better predict cyclist route choice behaviour in the future, other factors such as the type of landscapes and personal characteristics should also be included. | |
