Abstract:
Road crashes are mainly caused by three concurrent factors: infrastructure, vehicle, and human factors. The interaction between the infrastructure and human factors leads to the concept of geometric design consistency. Recently, a global consistency model was developed based on the difference between the inertial operating speed profile and the operating speed profile. The first was defined as the weighted average operating speed of the previous road section based on distance, and represents drivers’ expectancies, whereas the second represents road behavior. However, drivers’ expectancies are related to short-term memory which declines gradually and depends on time. Thus, a time-based inertial operating speed would allow a more accurate estimation of the phenomenon. This research analyzes different periods of time and weighting distributions to identify how drivers’ expectancies should be estimated. A set of 71 homogeneous road segments located in Italy were considered in the study. As a result, 25 seconds and a convex parabolic distribution should be used to calculate the inertial operating speed profile. This new way of estimating drivers’ expectancies showed better results than those obtained based on distance. Finally, the proposed consistency model was compared with the previous models and was found to be able to assess more accurately the geometric design consistency. Therefore, the proposed consistency model is a useful tool for engineers to estimate the number of crashes so that they can incorporate road safety considerations into the geometric design of either new two-lane rural roads, or improving the existing ones.
