详情

Understanding and Communicating Reliability of Crash Prediction Models

项目名称：	Understanding and Communicating Reliability of Crash Prediction Models
摘要：	The current Highway Safety Manual (HSM), published in 2010, provides models for predicting crashes for several common roadway facilities and methods for calibrating the models and some qualitative guidance on model reliability. However, the manual does not include methods to consistently convey model reliability. Transparency and credibility are essential to any form of analysis being performed for the public welfare. During initial implementation of the HSM, model results have been generated and often utilized to make major capital improvement decisions without fully understanding and communicating the accuracy of the model results, which can erode the credibility of this new and rapidly growing field. The state of the art of safety analysis has progressed and more has been learned about the impact on accuracy of assumptions made during the development of crash prediction models using HSM procedures. Reliability, accuracy, and appropriateness of safety models may not always conform to necessary requirements relative to facility type. Practitioners also appear to be struggling to fully understand and appropriately communicate the benefits of the HSM and results derived from the included methods. There are many factors that affect reliability, for example model accuracy. Case studies presented at various conferences, including the Transportation Research Board (TRB) Annual Meeting, and through other initiatives demonstrate that some practitioners are utilizing the models incorrectly or in ways not recommended, and are also displaying crash prediction results without properly understanding the model reliability. Some crash prediction models have been reported with measures of model reliability while others have not. Understanding and communicating consistently reliable crash prediction results is critical to credible analysis and a barrier for some transportation agencies or professionals utilizing these models. The objectives of this research are to develop guidance for: (1) the quantification of the reliability of crash prediction models including crash modification factors and/or functions (CMFs) and safety performance functions (SPFs) for practitioner use; (2) user interpretation of model reliability; and (3) the application of crash prediction models accounting for, but not limited to assumptions, data ranges, and intended and unintended uses.
状态：	Active
资金：	300000
资助组织：	National Cooperative Highway Research Program
项目负责人：	Bush, Mark S
执行机构：	University of North Carolina, Chapel Hill
主要研究人员：	Srinivasan, Raghavan
开始时间：	20160920
预计完成日期：	20190319
实际结束时间：	0