CMF / CRF Details

CMF ID: 7043

Install median barrier

Description:

Prior Condition:  No median barrier present

Category: Roadside

Study: NCHRP Report 794: Median Cross-Section Design for Rural Divided Highways, Graham et al., 2014

 
Star Quality Rating:3 Stars  [View score details]
Rating Points Total:80
Crash Modification Factor (CMF)
Value:0.04
Adjusted Standard Error:
Unadjusted Standard Error:0.014
Crash Reduction Factor (CRF)
Value:96  (This value indicates a decrease in crashes)
Adjusted Standard Error:
Unadjusted Standard Error:1.44
Applicability
Crash Type:Cross median
Crash Severity:All
Roadway Types:Principal Arterial Other Freeways and Expressways
Street Type:
Minimum Number of Lanes:4
Maximum Number of Lanes:4
Number of Lanes Direction:
Number of Lanes Comment:
Crash Weather:Not specified
Road Division Type:Divided by Median
Minimum Speed Limit:
Maximum Speed Limit:
Speed Unit:
Speed Limit Comment:
Area Type:Rural
Traffic Volume:
Average Traffic Volume:
Time of Day:All
If countermeasure is intersection-based
Intersection Type:
Intersection Geometry:
Traffic Control:
Major Road Traffic Volume:
Minor Road Traffic Volume:
Average Major Road Volume :
Average Minor Road Volume :
Development Details
Date Range of Data Used:2000 to 2007
Municipality:
State:MO,NC,OH
Country:USA
Type of Methodology Used:Before/after using empirical Bayes or full Bayes
Sample Size (sites):238 sites before, 238 sites after
Other Details
Included in Highway Safety Manual?No
Date Added to Clearinghouse:Nov 01, 2015
Comments:This CMF applies to all cross-median crashes for flexible barriers installed on four-lane freeways. The number of crashes in the after period were not reported in this study, however, they have been recorded as 300 to give 10 points as a beneift of doubt for one or more of the following: (1) number of miles/sites in the reference/treatment group, (2) number of crashes in the references/treatment group, (3) reporting AADTs for the aggregate dataset but not for the disaggragate dataset used for CMF development.