Reliability-Based Highway Geometric Design

公路碰撞事故通常归因于公路系统三个要素中的一个或多个发生故障：道路本身、车辆和道路使用者。然而，由于90%-95%的撞车事故是由司机失误造成的，人们强烈强调教育和执法举措，但收效甚微。另一方面，在通过改进公路几何设计来减少驾驶员错误和管理驾驶员行为方面的努力相对较少。公路几何设计涉及到不同设计要素的合理尺寸确定和协调。尽管自20世纪初引入机动车辆和铺设道路以来，设计准则不断演变，但基本原则并没有太大变化。 尽管公路系统的三个要素(道路、车辆和驾驶员)的特性差异很大，但公路设计遵循了一种确定性的方法，即每个参数都减少到一个接近最坏的值。这种保守值的累积效应可能会大大高估设计。此外，这些参数是根据主要收集于20世纪30年代和40年代的数据估计的，这些数据显然已经过时。 在本方案中，申请人采用了基于可靠性分析的概率公路几何设计方法，其中通过每个设计参数的分布来明确考虑驾驶员和车辆能力的可变性。研究了水平曲线和高速公路变速车道的设计模型。还将收集数据，以更好地反映当前司机、车辆和道路的行为、能力和限制。然后，研究将调查设计不充分的可能性与道路安全性能之间的关系。

Highway collisions are generally attributed to the failure of one or more of the three elements of the highway system: the road itself, the vehicle, and the road user. However, with driver error being blamed for 90-95% of collisions, a strong emphasis has been placed on education and enforcement initiatives with relatively little success. On the other hand, comparatively very little effort has been directed towards reducing driver error and managing driver behaviour through improving the highway geometric design. Highway geometric design involves the proper dimensioning and coordination of the different design elements. Despite the continuous evolvement of the design guidelines since the introduction of motorized vehicles and paved roads in the early 1900s, the fundamentals have not considerably changed. Despite the wide variability in the characteristics of the three elements of the highway system (road, vehicle, and driver), highway design has followed a deterministic approach where each parameter is reduced to one near-worst-case value. The cumulative effect of such conservative values can significantly overestimate the design. Furthermore, these parameters are estimated based on data mostly collected in the 1930s and 1940s, which are clearly outdated. In this proposal, the applicant adopts a probabilistic highway geometric design approach based on reliability analysis, where the variability in driver and vehicle capabilities are explicitly considered through the distribution of each design parameter. Models are investigated for the design of horizontal curves and freeway speed-change lanes. Data will also be collected to better reflect the behaviour, capabilities, and limitations of current drivers, vehicles, and roads. The research will then investigate the relationship between the probability of an inadequate design and the road safety performance.