Advancing knowledge of the safety implications of transportation engineering decisions with surrogate safety data and innovative statistical techniques

近年来，交通工程师研究和评价道路和交通安全的方式有了显著的进步。工程师现在可以采用系统和严格的流程来分析道路现场的安全，以确定减少撞车事故和实施潜在安全对策的机会。这一过程主要依赖于几年来研究开发的一套统计技术和模型。反过来，这些模型依赖于碰撞观察作为量化安全水平的主要数据来源。然而，在安全模型中使用碰撞数据有几个限制，主要与质量和可用性问题有关。此外，碰撞观察的使用是被动的：在采取任何行动之前，必须发生大量的碰撞。这些限制也阻碍了该领域方法学进步所提供的新机会的应用。 替代安全数据代表了道路安全分析中完全使用碰撞的一种有前途的替代方案。到目前为止，在估计和验证将替代指标转换为崩溃频率的系数方面开展的研究有限。然而，统计方法的不断进步使研究人员能够从现有数据来源中提取越来越多的信息。为此，拟议的研究计划旨在利用非基于碰撞的安全措施在道路安全方面实施先进的方法，并在碰撞和碰撞代理之间建立功能关系。 拟议研究计划的短期目标是：(1)收集和测量新出现的替代安全数据和指标，以克服与碰撞数据相关的限制；(2)通过与现场测量的替代数据建立关系，验证从交通微观模拟模型获得的替代安全数据的使用；以及(3)利用道路安全统计的新方法前沿，在碰撞和碰撞替代之间建立函数关系。这项研究的结果有望通过提高与交通运营和道路设计选择的安全影响相关的知识水平来提高高速公路的安全性能。这将对我国社会的可持续性产生直接影响，因为道路伤亡造成了巨大的经济和人员损失。

The way transportation engineers study and evaluate road and traffic safety has remarkably advanced in recent years. Engineers can now employ a systematic and rigorous process to analyze safety at road sites that identifies opportunities to reduce crashes and implement potential safety countermeasures. This process mainly relies on a set of statistical techniques and models developed over several years of research. In turn, these models rely on crash observations as the main data source to quantify safety levels. However, the use of crash data in safety models has several limitations, mainly related to quality and availability issues. Moreover, the use of crash observations is reactive: a substantial number of crashes must occur before any action is taken. These limitations also inhibit application of new opportunities offered by methodological advancements in the field. Surrogate safety data represents a promising alternative to the exclusive use of crashes in road safety analysis. So far, limited research has been carried out to estimate and validate coefficients converting surrogate measures to crash frequency. Nevertheless, the continual advance in statistical methodologies has enabled researchers to extract more and more information from existing data sources. For this reason, the proposed research program is aimed at implementing advanced methodologies in road safety using non-crash-based safety measures and establishing a functional relationship between crashes and crash surrogates. The short-term objectives of the proposed research program are to: (1) collect and measure emerging surrogate safety data and indicators to overcome the limitations related to crash data; (2) validate the use of surrogate safety data obtained from traffic microsimulation models by establishing a relationship with field-measured surrogates; and (3) establish a functional relationship between crashes and crash surrogates by taking advantage of the new methodological frontier in road safety statistics. The results of this research are expected to enhance safety performance of highways by increasing the level of knowledge associated with the safety implications of traffic operations and road design choices. This will have direct impact on the sustainability of our society due to the enormous economic and human costs related to road casualties.