Advancing the science of road safety management through innovation in crash and non-crash based analysis

根据加拿大汽车运输管理委员会(CCMTA)的数据，在加拿大，每年有近2000人在使用加拿大道路时死亡，16.5万人受伤(重伤1万人)，社会损失370亿美元。CCMTA网站维护一份可供用户搜索的道路安全倡议清单，道路安全利益攸关方可以采纳或调整这些倡议，以应对其具体的道路安全挑战。虽然其中一些措施已被“证明”有效，但值得注意的是，对另一些措施来说，尚不具备衡量的效力，因此这些措施仅被视为“有希望的”。需要了解衡量的安全效力，以便将这些有希望的措施，特别是与基础设施有关的措施重新归类为“已证实的”，并在制定碰撞修正系数的过程中，这些系数对于在规划基础设施安全措施的成本效益分析和项目优先次序中估计安全效益至关重要。在满足这一需求方面，传统的道路安全分析方法所能达到的效果是有限的，尽管知识匮乏，但这一领域的研究成果正在迅速减少。为了推动科学解决这一空白，我提出了一项为期五年的研究计划，其长期目标是填补关于道路基础设施决策安全影响的难以捉摸的知识空白。其中一些知识差距可能会在未来出现，因为道路上的交通流量越来越多地受到不同水平和混合的自动化的影响，这些自动化现在看来在不远的将来是不可避免的。这个五年计划将包括三个相互关联的组成部分，它们构成了短期目标。第一个组件将继续我现有的研究，开发碰撞修改功能，使用碰撞和道路数据将工程决策的安全影响与应用环境联系起来。这一知识不仅可以在规划干预措施中对道路基础设施决策的安全影响进行更准确的估计，而且还将促进所得出的知识的可转移性。第二部分将利用最近在美国和加拿大获得的大量自然主义驾驶数据来调查使用碰撞替代物来获得关于道路设计特征的安全影响的知识的可行性。第三个组成部分旨在量化人工和自动车辆混合使用的道路的安全性能，认识到完全自动化和完全普及可能在遥远的未来。拟议的研究不仅将通过促进安全分析科学而使道路安全研究界受益，而且还将有助于实现CCMTA最近制定的加拿大道路安全战略(RSS)2025的目标，以帮助实现该战略使加拿大道路成为世界上最安全的目标。

According to the Canadian Council of Motor Transportation Administrators (CCMTA), each year in Canada, almost 2,000 people are killed and 165,000 are injured (10,000 seriously) while using Canada's roads, costing society $37 billion. The CCMTA website maintains a user searchable inventory of road safety initiatives which road safety stakeholders can adopt or adapt to address their specific road safety challenges. Although some of these have been “proven” effective, it is noted that for others, measured effectiveness is not yet available, and these are thus considered only as “promising”. There is a need for knowledge on measured safety effectiveness in order to reclassify those promising measures, particularly infrastructure related ones, to “proven”, and in the process to develop crash modification factors that are vital for estimating safety benefits in cost-benefit analysis and project prioritization in planning infrastructure safety measures. In fulfilling this need, there are limits to what can be achieved with traditional road safety analytical methods, and research returns in this area are rapidly diminishing despite the dearth of knowledge. To advance the science in addressing this void, I propose a five-year research plan with a long-term goal of filling elusive knowledge gaps on the safety implications of road infrastructure decisions. Some of these knowledge gaps are likely to occur in the future as traffic flow on roads is increasingly impacted by the various levels and mixtures of automation that now seem inevitable in the not so distant future. The five-year program will have three inter-related components that form the short-term objectives. The first component will continue my existing research on developing crash modification functions that relate the safety impacts of an engineering decision to application circumstances using crash and roadway data. This knowledge will not only allow more precise estimation of safety effects in making road infrastructure decisions in planning interventions, but will also facilitate the transferability of the derived knowledge. The second component will utilize the extensive naturalistic driving data recently acquired in the United States and Canada to investigate the viability of using crash surrogates to derive knowledge on safety effects of road design characteristics. The third component aims to quantify the safety performance of roads with mixes of manual and automated vehicles, recognizing that full automation and full penetration may be in the distant future. The proposed research will not only benefit the road safety research community through advancing the science in safety analysis, but it will also contribute towards fulfilling the objectives of the Canada's Road Safety Strategy (RSS) 2025 recently developed by the CCMTA in helping to achieve the Strategy's goal of making Canada's roads the safest in the world.