EXECUTIVE
SUMMARY
This report
analyzes the problem of pedestrian crashes in the United States
(U.S.) to support the development and assessment of effective
pedestrian crash avoidance systems as part of the Intelligent
Vehicle Initiative. Pedestrian crashes are defined as those
involving one moving vehicle striking a pedestrian. In 1998,
about 70,000 such crashes or 1.1% of all police-reported crashes
occurred in the U.S. These crashes resulted in 5,294 fatal crashes
or 14.3% of all fatal motor vehicle crashes during that year.
This analysis identifies and counts these crashes by their pre-crash
scenarios that represent vehicle maneuvers and pedestrian actions
immediately prior to impact. Moreover, these pre-crash scenarios
are individually described in terms of their physical setting,
crash contributing factors, and crash characteristics such as
the age of people involved and maximum injury severity. The
analysis was conducted using a four-year data set from the 1995-1998
National Automotive Sampling System/General Estimates System
(NASS/GES) and Fatality Analysis Reporting System (FARS) crash
databases of the National Highway Traffic Safety Administration.
Eight “basic”
pre-crash scenarios were found to be the most common in pedestrian
crashes, accounting for nearly 90% of all police-reported pedestrian
crashes over the 1995-1998 time period. These scenarios combine
vehicle maneuvers such as going straight, turning right or left,
and backing up, with pedestrian actions such as crossing, darting
onto, walking along, playing, and working in the roadway. The
majority or 55% of all pedestrian crashes happened away from
junctions, a junction being the area formed by the connection
of two roadways. About 40% of pedestrian crashes were associated
with intersections. The following ten “specific”
pedestrian pre-crash scenarios were obtained by correlating
the eight basic pre-crash scenarios with information about the
crash relation to junction (percentages shown below refer to
the frequency of each scenario relative to the size of all pedestrian
crashes):
- Vehicle
is going straight and pedestrian is crossing the roadway at
nonjunction (25.9%).
- Vehicle is going straight and pedestrian is crossing the
roadway at intersection (18.5%).
- Vehicle is going straight and pedestrian is darting onto
the roadway at nonjunction (16.0%).
- Vehicle is turning left and pedestrian is crossing the roadway
at intersection (8.6%).
- Vehicle is turning right and pedestrian is crossing the roadway
at intersection (6.2%).
- Vehicle is going straight and pedestrian is walking along
the roadway at nonjunction (3.7%).
- Vehicle is going straight and pedestrian is darting onto
the roadway at intersection (2.5%).
- Vehicle is backing up (2.5%).
- Vehicle is going straight and pedestrian is not in the roadway
at non-junction (1.2%).
- Vehicle is going straight and pedestrian is playing or working
in the roadway at non-junction (1.2%).
The crash statistical description provided
in this report focuses on the above ten specific pre-crash scenarios
that account for 86.4% of all police-reported pedestrian crashes.
The majority of these pre-crash scenarios at non-junctions occurred
on straight, non-hillcrest roadways with posted speed limits
between 25 mph and 35 mph. The 3-color signal was reported as
the traffic control device present in 45% of these scenarios
at intersections, while “no controls” were coded
in 36% of these crashes. It should be noted that “no controls”
coding in the GES refers only to the direction of the road the
vehicle is traveling on. About 17% of all pedestrians involved
in the 10 pre-crash scenarios were in the crosswalk at the time
of impact.
The analysis of crash contributing factors
in the ten specific scenarios revealed that a very high percentage
of drivers reported vision obscurity in pre-crash scenarios
where the pedestrian darted onto the roadway (scenarios 3 and
7). Alcohol involvement was particularly high for drivers in
scenarios where the pedestrian was either walking along the
roadway at non-junctions or simply not in the roadway (scenarios
6 and 9). On the other hand, a high percentage of drunken pedestrians
was observed in scenarios 1, 2, and 6 where the pedestrian was
either crossing the roadway or walking along the roadway. Almost
60% of pedestrian crashes in which the pedestrian was walking
along the roadway at a non-junction occurred at nighttime (scenario
6). Hit and run cases were prevalent in 15% to 19% of pedestrian
crashes across the ten scenarios.
Younger pedestrians, especially those aged
from 5 to 9 years old, were the most susceptible to vehicle-pedestrian
crashes, accounting for nearly 14% of all pedestrians involved.
This age group had about the same relative frequency in the
two pre-crash scenarios where pedestrians darted onto the roadway,
35% in scenario 3 and 37% in scenario 7. The pedestrian age
group of 5-24 years old composed about 46% of the pedestrian
crashes and was the only age group over-represented in terms
of the U.S. population. It should be noted that the focus of
this analysis is on crashes, not injuries and fatalities, and
that different age groups may emerge as over-represented if
injuries or fatalities were the focus of the analysis. Roughly
43% of pedestrians in the pre-crash scenario where the pedestrian
was walking along the roadway at a non-junction were aged from
10 to 24 years old. About 22% were in the 15 to 19 years old
age group. The age group of 30-34 years old had the greatest
frequency of drivers involved in pedestrian crashes, accounting
for about 14% of all drivers involved in pedestrian crashes.
Relative to the licensed driver population, drivers under the
age of 20 years old were most likely to be involved in pedestrian
crashes. Such drivers comprised about 5.3% of the total licensed
driver population and yet were involved in more than 11% of
pedestrian-related crashes.
Pedestrian injuries tended to be more severe
away from junctions. At intersections, injuries were much more
severe in the scenario where the vehicle was turning left versus
the scenario in which the vehicle was turning right. The analysis
of FARS data indicated that 25% of fatalities occurred in pre-crash
scenarios where the vehicle is going straight and the pedestrian
is walking along, playing, or working in the roadway.
The
Full Report is available for download in PDF. The size is 442K
and is 96 pages.
Download
the Full Report
