Research Product Transfer for Local Calibration Factors of the Highway Safety Manual (HSM) and Integrated Surrogate Safety Assessment Framework

Improving the safety of road users is one of the most important goals of transportation agencies in the United States. Various safety plans and policies have partly contributed to a reduction in fatalities and injuries from vehicle crashes. A recent statistics of the National Highway Traffic Safety Administration (NHTSA) reported that vehicle crash fatalities decreased by 25 percent between 2002 and 2011. More strategies, planning, and policy efforts should be implemented to keep this downward trend. The implementation of latest study findings and procedures in the real world planning would be of help for government agencies to systematically assess roadway safety and prescribe appropriate countermeasures. In this vein, the proposed workshop is to present state transportation officials two successful and closely related projects that were conducted, respectively by Morgan State University and the University of Virginia. A team of researchers at Morgan State University has recently completed a study of developing local calibration factors to adjust the Highway Safety Manual (HSM) to the state of Maryland. The study was sponsored by the Maryland State Highway Administration and the National Transportation Center of Morgan State University. The second study to be provided in the proposed workshop is the surrogate safety assessment model (SSAM) in assessing safety conducted by researchers at the University of Virginia. The HSM is a culmination of decades-long efforts to provide a technical approach which is based on a system analysis frame. The HSM provides tools to facilitate roadway safety planning, design, operations, and maintenance decisions based on explicit consideration of their safety consequences. Once a data set is prepared for the HSM, it is expected that, ultimately, the HSM approach will help government agencies utilize limited resources more efficiently by quantifying and prioritizing the potential safety effects of government actions. To apply the HSM predictive methodology to the study area, one more step should be taken: the calibration of local calibration factors (LCFs). The crash prediction models of the HSM were developed using the data from a number of similar facilities of states of Washington and California. Due to multiple factors that may vary across the country, such as climate, population, traffic, crash reporting system, and others, the estimated crashes from HSM models cannot be directly applied to local agencies. To be effective, LCFs for roadway segments and intersections with various roadway geometry configurations should be developed. This process involve laborious tasks of data collection, generation and compilation which include historical data on crashes, traffic volume, roadway characteristics data, and land use data, as well as necessary procedures such as site selection, model estimation, and calibration. The HSM and any statistical crash models have been widely used in assessing safety for existing transportation networks. This is because one can develop the relationship between actual crash data and covariates including vehicular volume, speed, speed variance, etc. For example, statistical approach develops a regression type model estimating crash frequency based on the Average Annual Daily Traffic (AADT) and speed variables. However, these tools are not applicable for the untried and/or new strategies (i.e., no historical crashes are available). Surrogate safety measures were proposed to assess safety based on "conflicts" even from microscopic traffic simulation tools. Many studies have used the surrogate safety assessment model (SSAM) in assessing safety as it complements the HSM and statistical models for untried conditions. It should be noted that this 3 approach could be extremely useful if there is few number of crashes where the statistical approach cannot provide significant difference in crash frequencies/rates. However, one of limitations in the SSAM approach is the microscopic traffic simulator model that does not explicitly consider later movements within the lane during the lane change and the aggressiveness in the lane change durations. A recent Federal Highway Administration (FHWA) funded Exploratory Advanced Research project developed an enhanced integrated safety assessment framework that overcame such limitations. A validation study with actual crashes showed statistically better performance than that of traditional approach. This workshop will highlight the general overview of the traditional SSAM model, the enhanced approach and its validation, and a tool implementing the proposed approach. While the HSM provides a very detailed guidance and example, a real world application is more complicated than what is described in the manual. In addition, surrogate safety assessment model (SSAM) was developed to assess safety in untried conditions and a recent research enhanced the SSAM by integrating vehicle dynamics model and lane change aggressiveness. The goal of this workshop is to provide the participants with a workshop of Local Calibration Factor development process as well as SSAM application. The objectives of this workshop are to: (1) review the HSM predictive method analysis process; (2) discuss data collection and related issues; (3) provide examples of complementary data collection methods; (4) explain local calibration factor development process and implications; (5) provide the significance of surrogate safety measures; (6) teach how to use SSAM tool and explain its limitations; (7) discuss the need for the proposed approach and explain how it works; (8) conduct a case study comparing the proposed and traditional approaches; and (9) develop a tool implementing the proposed approach for practical use.

    Language

    • English

    Project

    • Status: Active
    • Funding: $86266.00
    • Contract Numbers:

      DTRT012-G-UTC03

      MAUTC-2013-06

    • Sponsor Organizations:

      Mid-Atlantic Universities Transportation Center

      Pennsylvania State University
      201 Transportation Research Building
      University Park, PA  United States  16802-4710
    • Performing Organizations:

      University of Virginia, Charlottesville

      Center for Transportation Studies
      P.O. Box 400742, Thornton Hall, D228
      Charlottesville, VA  United States  22903

      National Center for Transportation Management, Research and Development

      Morgan State University
      1700 East Cold Spring Lane, Montebello D-206
      Baltimore, MD  United States  21251
    • Principal Investigators:

      Shin, Hyeon-Shic

      Lee, Young-Jae

    • Start Date: 20130901
    • Expected Completion Date: 0
    • Actual Completion Date: 20141231
    • Source Data: RiP Project 35295

    Subject/Index Terms

    Filing Info

    • Accession Number: 01514348
    • Record Type: Research project
    • Source Agency: Mid-Atlantic Universities Transportation Center
    • Contract Numbers: DTRT012-G-UTC03, MAUTC-2013-06
    • Files: UTC, RiP
    • Created Date: Feb 14 2014 1:00AM