This study focused on selected Provincial Roads in the South Holland Procince and sought to establish if the practice of designing horizontal curves in the Netherlands has led to increased accident rates. Two main objectives were drawn with respect to this task. Firstly to evaluate the quality of design of the horizontal curves in accordance with the Lamm Method, Relation Design Method and on the basis of Accident Rates. The other was to ascertain the influence of geometric design consistency indicators on the accident rates. The geometric design consistency indicators considered were alignment indices and the reduction in predicted operating speeds. The study found the quality of design of the horizontal curves to be in the good to fair range and none as poorly designed. This finding was generally consistent for the three evaluation techniques. The alignment indices had practically no correlation with the accident rates. Speed reduction had a relatively high correlation accounting for 34% of the accidents but this was still weak. As such speed reductions too did not influence the accident rates. This lack of correlation further reinforced the evaluation that the horizontal curves were in the good to fair range with respect to their geometric design consistency quality. Therefore the practice of designing horizontal curves in the Netherlands has not led to increased accident rates. On the contrary it has also guaranteed consistency in the geometric alignment of horizontal curves. However, based on the scope of this study and the methods of predicting vehicle operating speeds it is recommended that speed prediction equations need to be calibrated to suit the Dutch situation. Further study into other aspects of design consistency that relate to intersections, interchanges and general highway aesthetics is necessary. It should also be established if the Provincial Roads in the South Holland Province are, in this context, representative for the whole Netherlands. (Author/publisher)
Abstract