RISK-BASED MANAGEMENT OF GUARDRAILS: SITE SELECTION AND UPGRADING

1
RISK-BASED MANAGEMENT OF GUARDRAILS SITE
SELECTION AND UPGRADING

• Presented to
• Project Steering Committee
• Virginia Department of Transportation
• by the
• Center for Risk Management of Engineering Systems
• April 17, 2000

2
Agenda

• Introduction
• Risk-based screening of corridors
• Data representation for site screening
• Multiple objectives in the selection among
  candidate sites
• Software demonstrations
• Discussion

3
Project Team
Center for Risk Management of Engineering
Systems James H. Lambert, Research Assistant
Professor of Systems Engineering, Center
Associate Director Yacov Y. Haimes, Quarles
Professor of Systems Engineering and Civil
Engineering and Center Director Jeffrey A. Baker,
BS/MS Student Christian R. Baldwin Irene A.
Jacoub Mike R. Raker Virginia Transportation
Research Council Wayne S. Ferguson, Research
Manager VDOT Richmond District Travis Bridewell,
District Traffic Engineer, Richmond District Jeff
Wilkinson, Transportation Engineer, Traffic
Engineering Section, Richmond District Baron
Gissendaner
4
Project Team (cont.)
Additional Steering Committee Steve Edwards,
Transportation Engineer Senior, Traffic
Engineering Division, Central Office Paul Kelley,
Transportation Engineer, Location and Design
Division, Central Office Charlie Kilpatrick,
Fredericksburg Resident Engineer, Fredericksburg
District Bob McCarty, Senior Field Operations
Engineer, Federal Highway Administration -
Richmond Ginger Quinn, District Safety Officer,
Traffic Engineering Section, Salem District Nancy
Berry, Transportation Engineering Program
Supervisor, Location and Design Division, Central
Office Bill Bushman, Virginia Transportation
Research Council Angela Tucker, Resident
Engineer Willie Gentry, Resident Engineer Alan
Leatherwood, Resident Engineer
5
Problem Statement

• Public and transportation-agency values
  concerning the location of roadway guardrails in
  need of clarification
• Concerns of Virginians for adequate guardrails
  high relative to the national norms
• VDOT Districts select locations for new
  guardrails based on citizen complaints, a general
  knowledge of roadway needs from local engineers,
  and accident history

6
Problem Statement (cont.)

• Kentucky hazard-index point system (Kentucky
  Transportation Center Report KTC-89-39 "Warrants
  and Guidelines for Installation of Guardrail")
• Hundreds of candidate locations on the
  thirteen-county secondary system of Richmond
  District
• New Kent and Charles City County the focus of a
  related preliminary study in Richmond District

7
Purpose and Scope

• Identify attributes and develop associated
  cost-benefit-risk tradeoff methodology to support
  screening and evaluation for guardrail site
  selection and upgrading with limited available
  funding

8
Purpose and Scope (cont.)

• Objectives
• Review and evaluate what others have done
• Adopt assessment methods
• Develop tradeoff methodology
• Specify and develop prototype databases

9
Screening of Corridors
10
Evaluation of Sites
11
Risk Based Screening of Corridors
12
Motivation

• A data-driven approach to assess accident risk
  and associated guardrail needs across Districts
  and Residencies

13
Data Needs

• Screening
• Guardrail inventories
• Percent unprotected hazards
• Percent guardrail coverage
• Percent substandard guardrail
• Accident histories
• FO accidents per DVMT
• Fatalities caused by FO accidents
• Average daily traffic
• Complaint records

14
Data Needs (cont.)

• Evaluation
• Cost (installation, upgrade)
• Length of hazard
• Severity of hazard
• Shoulder width
• Slope
• Curvature

15
Accident Statistics

• Disadvantages
• Unreported accidents
• Severity iceberg
• First and most harmful event
• Fatalities do not occur frequently enough to be
  statistically predictive
• Random nature of road accidents
• (Adams, 1996), (Michie and Bronsted, 1994)

16
Accident Statistics (cont.)

• Advantages
• Available
• Factual
• Public interest

(Adams, 1996), (Michie and Bronsted, 1994)
17
New Kent Case Study

• Initial data collection
• Routes 600-608
• Corridor analysis
• Collect data
• Perform calculations (accidents per mile,
  accidents per DVMT)
• Generate tables and graphs

18
Corridor Analysis

• Compare routes for frequency and severity of
  accidents
• Compare accident statistics with guardrail
  coverage
• Advantages
• Reduce randomness of individual accidents
• Use summary statistics available in HTRIS

19
Corridor Analysis (cont.)

• Disadvantages
• Does not focus on individual locations
• E.g., many locations of mediocre severity vs. one
  location of very high severity

20
Data Acquisition
21
Corridor Screening
Similar treatment of accident counts and
accidents per DVMT
22
Corridor Screening (cont.)

• Example graph
• Run off the road accidents per DVMT

23
Data Representation for Site Screening
24
Motivation

• Need to organize data on many hazards protected
  and unprotected by guardrail on 40,000 miles of
  roadway across Virginia

25
Sample of Guardrail Inventory