CE 350 Traffic Accident Analysis Lecture

1
CE 350Traffic Accident Analysis Lecture

• Alabama and
• Highway Safety

2
1. Safety Culture

• Research shows it is necessary
• Other nations have it, we need it
• Why is killing 1200 people tolerated?

MADD changed the US alcohol culture Who is going
to change Alabama traffic crash culture?
3
Agenda
Is Enforcement the Key to Safety?

4
Global Safety Comparisons
Australia 9.0 Denmark 9.2 Canada 9.3 USA 9.4 Germ
any 11.1
Deaths/1B kmVT U K 7.6 Sweden 8.3 Norway 8.3 Switz
erland 8.4 Finland 8.5
5
Global Safety Comparisons (continued)

• US - worlds best mileage rate for many years
• Competitive now because we drive miles2

• US one of highest fatality rates/1000 population
  
• Not dropping significantly

• 40,000 per year per year per year

6
Importance of Safety Culture UK - public
supports speeding, safety belt, and cell phone
fines of 1,000 Most EC nations use automated
speed enforcement
2,000
7
Australia - Most like the USGovernment, Road
Systems and Travel Patterns

• Nearly identical rates in 1980
• Now about half U.S. rate

8
Australias Safety Culture Every Drivers
Responsibility

• Impairment
• 0.05 BAL for all drivers
• Zero BAL first 3 years, and taxi drivers
• Immediate license loss for 2nd offenses

• Enforcement
• 500 more random breath testing
• 1 in 3 drivers tested annually
• Camera enforced speed limit

9
(No Transcript)
10
2. Good News

• Safety research is producing better tools, better
  guidance

Highway Safety Manual, FHWA documents, IHSDM,
Safety Analyst, short courses
11
(No Transcript)
12
(No Transcript)
13
Design Webinar - Reference Materials for
Overview of Safety Effects of Geometric Design
Features
14
Soft Side Research Human Factors
Good News A Definitive Countermeasure Listing
15
Greatest potential for safetyDifficult to
predict accident reduction or cost effectiveness

• Alcohol
• Seatbelt use
• Aggressive driving/speeding
• Distracted/fatigued driver
• Motorcycles
• Older driver

16
The Accident Picture
Driver involved in 85 of crash causation
17
Example Motorcycle Countermeasures
Helmet use is only proven countermeasure
18
Example younger driver countermeasures
19
Magnitude of R.O.R. ProblemAlabama Last 4 Years

• 85,249 Crashes
• 45,529 Injuries
• 1,685 Fatalities
• 16 of All Crashes
• 25 of all Injury Crashes
• 42 of All Fatal Crashes

20
ROR Crashes by Highway Class
21
Driver Age
22
Typical ROR ScenarioRamifications

• Delayed Discovery and EMS
• Higher speeds
• Low Restraint Usage
• Widely Distributed Low ADT Roads
• Curve-Superelevation Combinations
• Significant Increase over Four Years

23
Example used by FHWA-NHTSA-AASHTO National
seminar December 2003
24
Example ROR by Age
25
CARE Example cross-tabulation (Age Time)
26
(No Transcript)
27
www.highwaysafetymanual.org
28
 
 
IHSDM Overview The Interactive Highway Safety
Design Model (IHSDM) is a road safety evaluation
software that marshals available knowledge about
safety into a more useful form for highway
planners and designers. The Federal Highway
Administration (FHWA) has been developing IHSDM
with initial focus on two-lane rural highways.
The 2003 release of IHSDM for two-lane rural
highways is now available for testing and
evaluation purposes.
 
The 2003 Release of IHSDM is now available.To
download the software for testing and evaluation
purposes, click here.
29
http//www.safetyanalyst.org/
30
(No Transcript)
31
Overview - Analysis of aDesignated Crash Site(A
short primer)

• Daniel S. Turner
• CE 350

32
Typical Work Steps

• Choose THE Site (Mgmt. by Exception)
• Find Crash Pattern
• Pattern ? Cause

• Countermeasures
• Cost Effectiveness

• Implementation
• Evaluation

33
How to Choose THE Site

• Number of Accidents
• Accident Rate
• Number-Rate
• Severity
• Rate Quality Control
• CARE

34
1) Number Method

• Simple - (pin map)
• Pro - easy, good for small towns
• Con - reflects traffic volume

35
2) Rate Method

• Rate per MV or MVM
• Pro - over-represented locations
• Con - one accident on a low-volume road yields
  a huge rate

36
3) Number-Rate Method

• Combine methods to overcome weaknesses. Set a
  minimum threshold
• Example - For sites with 5 or more crashes, what
  are the 4 with the highest rates?

37
4) Severity Method

• Convert all crashes to equivalent PDOs
• Select a scale 1 fatal crash 12 PDOs, and 1
  injury crash 3 PDOs
• Example Site with 2 fatal, 7 injury and 13
  PDOs
• 2(12) 3(7) 17 (1) 62 EPDOs

38
5) Rate Quality Control Method
Crashes
39
Results of Comparison
40
Results of Comparison
41
Finding Crash PatternsPlot Collision Diagrams
42
Finding Causes of Patterns

• Condition diagram (scaled map)
• Site visit
• Cause-and-effect charts/tables

43
Selection of Countermeasures

• Experience helps
• Use tables of countermeasures and crash
  reduction factors

44
Evaluation - Selection

• Cost Effectiveness
• Cost of continued crashes
• Cost of treatment reduced crashes

• Select
• Best treatment at each site
• Best sites to treat

• Or use more complex optimization

45
Highway Safety Manual (HSM)

• TRB Task Force
• In cooperation with
• AASHTO Joint Task Committee

46
HSM Chapters

• Part III Predictive Methods
• Rural 2 lane
• Rural multilane
• Urban/suburban arterials
• Part IV Safety Mgmt of Systems
• Sites with promise
• Diagnosis
• Countermeasures
• Economic appraisal
• Prioritized rankings
• Part V Safety Evaluation
• Evaluate completed projects

• Part I Intro and fundamentals
• Introduction
• Fundamentals
• Part II Knowledge
• Road segments
• Intersections
• Interchanges
• Special situations
• Road networks

47
TRB Highway Safety Manual(HSM)

• Under development since 2000
• 35 member task force, 100 volunteers
• 7 committees, 5 production groups
• NCHRP panels to guide projects
• Two meetings/year
• Monthly conference calls
• I have read 6000 pages so far

48
(No Transcript)
49
(HSM) Example Chapters 1 7NCHRP Project 7-27

• iTRANS Consulting Ltd. plus eight respected
  safety experts
• Scope find best available knowledge
• Reviewed 573 safety studies
• 678 page summary of what we know

50
NCHRP Project 7-27

• Classified safety reliability
• Stable, high certainty, low variability
• Identified 400 desired safety relationships
  (12-ft lanes are safer than 11-ft
  lanes)
• Found less than 30 statistically significant

51
The SCIENCE

• Regression to the Mean (RTM)
• Empirical Bayes method (EB)
• Accident Modification Factors
• Standard Error
• The big question data quality

52
NCHRP Project 7-27
Mean
Small Std Error Desirable
AMF of 0.50 expected 50 crash reduction AMF of
1.07 expected 7 crash increase
Typical AMF of 0.85 Std Error 0.27 Expected
range of AMF to 0.58 to 1.12
Not Acceptable
53
NCHRP Project 7-27
Large Std Error Not Desirable
Neither is Acceptable
AMF of 0.85 Std Error 0.27 Std error too
large low confidence in result
AMF of 0.94 Std Error 0.13 Expected range to
0.81 to 1.07
54
Much of HSM is traditional, updated with better
science as needed
55
(No Transcript)
56
(No Transcript)
57
(No Transcript)
58
(No Transcript)
59
Adjust for RTM
x
Expected Crashes
x
60
AMF TablesHeart of the HSM
61
(No Transcript)
62
(No Transcript)
63
(No Transcript)
64
(No Transcript)
65
Concern and Commitment

• What have you done in the last month to save a
  life on Alabama highways?

66
There is no end to this presentation

• Put it on and wear it home!