Risk of Fatal RearEnd Collisions Is there more to it than attention - PowerPoint PPT Presentation

Loading...

PPT – Risk of Fatal RearEnd Collisions Is there more to it than attention PowerPoint presentation | free to view - id: 3f313-NzI1O



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Risk of Fatal RearEnd Collisions Is there more to it than attention

Description:

BMW. DaimlerChrysler. DBM Reflex. Denso. Exatec. Federal-Mogul. Fiat. Ford. GE. Gentex. General Motors ... Motorcycles and buses [After Wiacek & Najm, 1999] ... – PowerPoint PPT presentation

Number of Views:105
Avg rating:3.0/5.0
Slides: 28
Provided by: john1088
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Risk of Fatal RearEnd Collisions Is there more to it than attention


1
Risk of Fatal Rear-End CollisionsIs there more
to it than attention?
  • John M Sullivan
  • Michael J Flannagan
  • University of Michigan Transportation Research
    Institute

2
Industry Affiliation Program for Human Factors in
Transportation Safety
  • AGC America
  • Autoliv
  • Automotive Lighting
  • Avery Dennison
  • BMW
  • DaimlerChrysler
  • DBM Reflex
  • Denso
  • Exatec
  • Federal-Mogul
  • Fiat
  • Ford
  • GE
  • Gentex
  • General Motors
  • Guardian Industries
  • Guide Corporation
  • Hella
  • Honda
  • Magna International
  • Mitsubishi Motors
  • Nichia America
  • North American Lighting
  • OSRAM Sylvania
  • Philips Lighting
  • PPG Industries
  • Reflec USA
  • Reflexite
  • Renault
  • Samlip
  • Schefenacker International
  • Sisecam
  • Solutia Performance Films
  • Stanley Electric
  • TG North America
  • Toyota Technical Center, USA
  • Valeo
  • Vidrio Plano

3
Overview
  • What is known about causes of rear-end
    collisions?
  • Stimulus factors
  • Driver inattention
  • Method to assess crash risk as a consequence of
    darkness
  • Light level and fatal rear-end collisions
  • Conclusions

4
Rear-end collision facts
  • NHTSA GES for 2001
  • 1.9 million rear-end crashes
  • 30 of all crashes
  • FARS 2001
  • Rear-end crashes are 5 of all fatal crashes
  • While numerous, rear-end crashes are not
    particularly lethal

5
Causal analysis
  • Indiana Tri-Level Study (Treat, 1979 Knipling,
    Wang, Yin, 1993)
  • 68 of rear-end crashes caused by driver
    inattention and other forms of delayed
    recognition
  • NASS analysis (Knipling et al., 1993)
  • 66 of crashes caused by driver inattention
  • 75 occurred in daylight
  • 90 on straight roadways
  • Less than 10 of rear-end collisions were
    attributed to environmental factors
  • Icy roads, fog, rain, obscured vision

6
Can we conclude anything about driver attention?
  • Does attention stray more in the daylight?
  • Do straight roads invite rear-end collisions?
  • Does this suggest driver complacency?

7
No
  • A rear-end crash is the product of both
  • Driver behavior and
  • Crash opportunity (i.e., exposure)
  • Rear-end collisions occur between 2 (or more)
    vehicles with one vehicle behind the other
  • Traffic is most dense during daylight
  • There are more straight sections of roadway than
    curved
  • A crash is more likely on a straight roadway

8
Does poor visibility affect rear-end crash
likelihood?
  • Our ongoing work on driver vision at night
  • Previous work showed major effect of light on
    fatal pedestrian crashes
  • BUT
  • There was a slight effect on multiple-vehicle
    crashes
  • Closer look indicates that the effect mostly
    involves rear-end collisions

9
Isolating effects of light
  • Simple day/night comparisons fail to account for
    exposure differences and other risk factors
  • Fatigue, alcohol use, demographics
  • Compare two driving conditions which only differ
    with respect to light level
  • Daylight Saving Time Analysis (Ferguson et al.,
    1995 Sullivan Flannagan, 2002)

10
Daylight Saving Time Analysis
  • Uses abrupt seasonal change in clock time in
    spring and fall
  • Compare one-hour periods that transition from
    darkness to daylight and vice-versa

11
For example
  • April in Detroit
  • 730-830pm before DST is dark
  • 730-830pm after DST is light
  • Compile crash data in the weeks before (dark) and
    in the weeks after (light)
  • Compare the distribution of different crash types
  • Null hypothesis would suggest that they are
    equally distributed across dark and light periods

12
Assumptions
  • Road use is tied to clock time, not the position
    of the sun
  • Maybe not for children, retirees, animals
  • Observed differences between these two periods
    are related to differences in ambient light level

13
Annual Solar Cycle(End of Civil Twilight)
14
Civil Twilight (Fall PM)
15
United States Counties
  • Clock time of civil twilight end is tied to
    geographical location
  • Tagged all crashes that fell into time window
    based on county lat and lon
  • Determined weeks relative to DST changeover from
    crash date

For further detals refer to Sullivan and
Flannagan (2002). Accident Analysis and
Prevention, 34(4), 487-498.
16
Effect of ambient light
  • If light level has no influence, fatalities
    should be equally distributed in dark and light
    intervals
  • (Fatalsdark/Fatalslight) 1
  • Fatalsdark 50
  • Reliable departures indicate an effect of light
    level

17
Fatal Crashes During Fall PM
Light
Dark
18
Ambient light effect on crashes between vehicles
  • FARS Manner of Collision
  • Rear-end collisions
  • Front-to-Side collisions (Side)
  • Front-end collisions
  • Accumulated fatal crashes
  • FARS 15 years (1987-2001)
  • 5-week periods before and after changeovers in
    spring and fall
  • PM only

19
Fatal Crashes by Manner of Collision
20
Result
  • Fatal rear-end collisions are less frequent than
    front-end and side-impact
  • Rear-end collisions are more than twice as
    frequent in darkness as in daylight

21
Is there a pattern?
  • Is risk in darkness related to vehicle type and
    role?
  • Trucks are overrepresented as struck vehicles in
    nighttime crashes (Green, Kubaki, Olson, Sivak,
    1979).

22
Vehicle Type
  • Light vehicles
  • Automobiles, light trucks, SUVs, Vans
  • Trucks
  • Medium and heavy
  • Excluded vehicles
  • Motorcycles and buses

After Wiacek Najm, 1999
23
Role
  • Only fatal rear-end crashes in which an
    unambiguous role could be determined
  • Striking
  • Struck
  • Excluded
  • Crashes involving more than 2 vehicles

24
DST rear-end crash breakdown
25
Results
  • Fatal rear-end collisions involving striking
    light vehicles are 3 times as likely in darkness
  • Medium and heavy trucks appear to be 8 times as
    likely to be rear-ended in darkness
  • Questions
  • Why are trucks difficult to see at night?
  • Do drivers of light vehicles have faulty night
    driving strategies?

26
Conclusions
  • Perhaps both attention and stimulus factors
    contribute to the effect of light level
  • Approach detection in the visual periphery may be
    poorer in the darkness
  • Attention may be re-directed in daylight, but not
    as well in darkness
  • Drivers may reallocate attention away from the
    roadway if a threat is not accurately perceived

27
Thank You
About PowerShow.com