Commuting bicylists’ route choice

Car route choice have been extensively studied (see e.g. Frejinger et al 2009 ), but it remains a question whether the same route choice behavior applies to bicyclists. Being much slower than cars they may be seen as an intermediary private transport form– between the faster cars with long transport ranges, and the slower pedestrians, having a more limited range of activity. Furthermore, biking can also have an element of recreational purpose. As opposed to optimizing wayfinding, based on ‘perfect’ knowledge about the infrastructure, it can by hypothesized that slower types of movement – such as cycling – can better be described by inclusion of ‘locomotion’ where choices are determined by cyclists’ immediately perception, as they move forward (see Montello, 2005). Locomotion may seem more reasonable if we are looking at recreational movements, where immediate pleasure can be observed and the knowledge about the infrastructure and surrounding environment is limited. During commuting – which takes place in more well-known environments and often requires more efficient movement - it is more likely that navigation will be dominated by optimizing wayfinding. Yet we often find that people do not choose the fastest, nor the shortest, route. Modeling route choices may, therefore, take both elements of locomotion and wayfinding choice strategies into account. Consequently, both may be important for analyzing route choice behavior – which is an important component for the construction of bicycle infrastructure. The question we raise in this paper is whether bicyclists’ route choice can be described by a choice between routes connecting starting point A with end point B (as in wayfinding), or whether it can be described by a choice between a series of decisions along the route (as in locomotion). Second, we ask which factors determine the route choice. The factors we look at include the type of road, the presence, and quality of bicycle lanes/tracks, the environmental surroundings and factors influencing pace – e.g. of traffic lights and turns. Data are based on bicyclists in Copenhagen, Denmark. From a global perspective, Copenhagen has a relatively high mode share of biking of 35% (Pucher and Buehler, 2012), and an extensive network of bicyclist facilities, causing the route choice to be very flexible regardless of where you want to go. In this study, we use data from 3 one-week GPS-tracking observations, where the participants’ bicycle movements were recorded. In total 179 bicyclists were observed, taking 1292 trips with an average distance of 5.4 km and an average speed of 14.4 km/h. Apart from the importance of the above-mentioned attributes, we look at the importance of individual route-specific characteristics – the length of the route, whether choices differ between the beginning or the end of the route, whether it is in known or unknown terrain, and whether it differs with respect to socio-demographic characteristics. We find that both locomotion and wayfinding are able to explain cyclists’ navigation and that the attributes that determine the choice are to a large extend identical between the two approaches. The possibility of high speed seems quite important both by looking at the number of stops and turns, and the quality of roads. The presence of green environment plays a smaller role, which seems surprising given the stated focus participants have on the health and recreational effect of biking. This may be different if noncommuting routes were analysed. References: · Frejinger, E., Bierlaire, M. and Ben-Akiva, M. (2009). Sampling of Alternatives for Route Choice Modeling, Transportation Research Part B: Methodological · Montello, D. R. (2005). Navigation. In P. Shah & A. Miyake (Eds.), The Cambridge handbook of visuospatial thinking (pp. 257-294). Cambridge: Cambridge University Press. · Pucher, J. and Buehler, R. 2012. City Cycling. MIT Press