Pedestrian and Pedal Cyclist Route Choice Criteria

A Multi Attribute Utility Model was used to reveal more information on pedestrian and pedal cyclist route choice behaviour. The main objective for this experiment was to ascertain the relative importance of and the interrelation between several attributes that were mentioned as important route choice criteria. Another objective was to reveal the difference between three countries, Great Britain, Sweden and the Netherlands. Other objectives were to reveal the differences between men and women and between younger and older vulnerable road users.

Subjects were asked to select four routes from a map, for a frequently made trip. Subjects were familiar with the routes and were able to assign preferences to the routes. Preferences were assigned to the routes (global preference) and to each route for just a single attribute. The weight elicitation technique used was a modified indifference method.

The differences in importance of the attributes between the countries were small. Attributes that get high weights are thought to be important attributes in the route choice behaviour of a subject. Important attributes were rated high in all the countries and unimportant attributes were always rated low. Differences for gender and age were found to be marginal.

Distance and pleasantness were found to be important attributes for pedestrians. These attributes are also the minimal set that produce high correlation coefficients between the aggregated values for the routes and the global preferences. For pedal cyclists the attributes that formed the minimal set and the attributes that are rated important are distance, pleasantness and traffic safety. This minimal set correlated with the global preference at .70 for pedestrians and .71 for pedal cyclists. Distance as a single attribute could not produce these correlation coefficients.

The attribute weights are related to ranges in the objective world. This means that other attributes can be rescaled to distance. This revealed that for pedestrians on average a distance of more than 160 meters can be played off against one point on a pleasantness scale ( a 7-point scale ). For pedal cyclists on average a distance of 200 meters can be played off against one point on the pleasantness scale or a distance of 250 meters can be played off against one point on a scale for traffic safety.

An important conclusion from the experiment is the fact that pleasantness can be played off against distance. Subjects are prepared to walk or cycle further to have a more pleasant route. This means that the use of safer routes can be encouraged, even if they are longer, by making the safer routes more pleasant.