As part of the Sustainable Safety programme in the Netherlands the Dutch road network has been reclassified comprising essentially only two urban and three rural road categories or classes. Integral to this is an ongoing research project aimed at defining road and other design elements that visually distinguish the different road classes, making them recognisable to the road user and thereby provoking the correct road user behaviour. The present study is concerned with how the road lay-out can be best designed to ensure that the intended classification corresponds to the subjective categorisation of motorists. The study will also determine the effect that the environment, in which a road find itself, has on the perceived classification. In the experiment three road layout alternatives were combined with three different environments. All combinations were tested in three different tasks in which subjects were presented with photographs of road scenes. One of the road layout alternatives (indicated with WOV1) was a baseline design layout based on (preliminary) CROW specifications for the sustainable safe road categories (CROW, 1997). Another road layout alternative was the best alternative from research of Janssen, Claessens and Muermans (1999). In this alternative (as in Janssen et al. called WOV3), edge markings and 'counter stream markings' were added to increase the differences between the different road categories. A third road layout alternative is tested to investigate whether an alternative version of WOV3 would lead to better categorisation. This alternative, in which reflector posts were added, is called WOV3 +. To investigate the effect of the environment on road classification the different road layout alternatives were embedded in three different environments, called 6-4, 9-1, and neutral. In the 6-4 and 9-1 conditions the environment was altered such that the compatibility between the environment and the road category varied. The compatibility between environment and road category was higher in the 9-1 condition than in the 6-4 condition. In the condition with the neutral environment the complete environment of the road was eliminated with the exception of the relevant road characteristics. Each combination of road layout alternative and environment was investigated with three different tasks, a sorting task, a learning task, and an expectations task. In the sorting task the subjects had to sort photographs of road scenes on basis of expected road user behaviour (i.e., roads where similar behaviour was expected were to be sorted into one category). There were no restrictions. So, subjects could sort the material in any number of road classes they deemed fit and the number of photographs in a class could differ. The results showed that categories made in WOV3 and WOV3+ were more in agreement with the official classification than WOV1. The categories made in the 9-1 condition were more in agreement than in the 6-4 condition. The difference between the neutral condition and the 9-1 condition was small. The learning task was used to investigate whether road layout could be used to learn the different road categories. The results showed that the WOV3 layout was the best learned alternative. The best performance in the environment conditions was obtained in the 9-1 condition. The expectations task was used to investigate the correctness of the subject's expectations regarding the presence of other road users and maximum speed on roads. Furthermore, subjects had no indicate, irrespective of the speed limit. at what speed would drive on the roads. No effect of road layout alternative was obtained. In the neutral condition less questions were correctly answered than in the 6-4 and 9-1 condition. With respect to the subjects' own driving speed no effect was obtained from road layout alternatives. An effect of environment was obtained. Especially on roads within the urban area the choice of driving speed depended upon the absence or presence of `buildings'. On a number of road categories the own driving speed was (far) above the desired speed limit. This result can be ascribed to the expectations and knowledge of subjects regarding the present Dutch road system. On basis of the combined results of the three tasks the conclusion has to be that both WOV3 and WOV3+ are more in agreement with the official classification than WOV1. The difference between WOV3 and WOV3 + is small and in favour of WOV3. Moreover, the road environment influences the perceived classification of roads. This latter result is of importance for the recategorization of the Dutch road system. (A)
Samenvatting