Return to the VTRC Home Page
Click here to print the printer friendly version of this page.
Page Title: VTRC Report Detail

The contents of this report reflect the views of the author(s), who is responsible for the facts and the accuracy of the data presented herein. The contents do not necessarily reflect the official views or policies of the Virginia Department of Transportation, the Commonwealth Transportation Board, or the Federal Highway Administration. This report does not constitute a standard, specification, or regulation. Any inclusion of manufacturer names, trade names, or trademarks is for identification purposes only and is not to be considered an endorsement.


Calibration of Highway Safety Manual Safety Performance Functions for Freeway Ramp Terminals in Virginia
Justice Appiah, Ph.D., P.E.
Justice Appiah
Justice Appiah
Year: 2019
VTRC No.: 23-R17

Chapter 19 of the Highway Safety Manual (HSM) provides safety performance functions (SPFs) for freeway ramps and crossroad terminals.  The chapter includes 56 predictive models for ramp terminals characterized by terminal type, intersection control, crash severity, area type, and number of crossroad lanes.  These SPFs were developed with data from other states and need to be calibrated to Virginia conditions to ensure that they accurately reflect the driver population and environment.  The application of uncalibrated SPFs may produce misleading results, compromise safety outcomes, and lead to inappropriate design decisions.

This study conducted systematic calibration of the HSM ramp terminal SPFs to account for conditions in Virginia.  This involved determining appropriate multipliers or functions that aligned the expected average crash frequencies estimated using HSM methodologies with field-observed crash frequencies from selected sites.  A review of cumulative residual plots for fitted values suggested that using a single calibration factor as a multiplier to adjust the HSM ramp terminal SPF predictions did not provide a good fit to Virginia data.  Consequently, calibration functions were developed that provided a better fit of Virginia data to the HSM ramp terminal SPF predictions.  Limiting the number of crash modification factors to 3 instead of using all 11 applicable crash modification factors resulted in a marginally better fit of the data.