ArticlePDF Available

Overview of Wrong-Way Driving Fatal Crashes in the United States

Authors:

Abstract and Figures

In this study, 8 years (2004–2011) of wrong-way driving (WWD) fatal crash data were extracted from the National Highway Traffic Safety Administration Fatality Analysis Reporting System database. The objectives of this study are to (1) provide an overview of the general trend of WWD fatal crashes in the United States; (2) discuss general characteristics of WWD fatal crashes; and (3) delineate significant contributing factors (e.g., crash location, driver gender, age, and impairment). The results will serve to inform national and state efforts to reduce WWD fatal crashes.
Content may be subject to copyright.
Overview of Wrong-Way Driving
Fatal Crashes in the United States
In this study, 8 years (2004–2011) of wrong-way driving (WWD) fatal crash data were
extracted from the National Highway Trac Safety Administration Fatality Analysis
Reporting System database. e objectives of this study are to (1) provide an overview of
the general trend of WWD fatal crashes in the United States; (2) discuss general charac-
teristics of WWD fatal crashes; and (3) delineate signicant contributing factors (e.g., crash
location, driver gender, age, and impairment). e results will serve to inform national and state
eorts to reduce WWD fatal crashes.
B F B-G, P.D.,
H Z, P.D., P.E.,  J S, P.E., PTP, PTOE
Wrong-way driving (WWD), by denition, happens when a driver,
inadvertently or deliberately, drives in the opposite direction of
trac ow along a physically divided highway (freeway, expressway,
or interstate highway) or its access ramps. Although WWD crashes
are relatively infrequent compared to other crash types, the nature
of the crashes—multiple vehicles traveling at very high speeds and
usually head-on impact—creates a very high likelihood of fatal
injury. As a matter of road safety, concerns about WWD have
existed since the advent of access-controlled, divided highways.
Precursors to the American Association of State Highway and
Transportation Ocials Green Book going back to the 1960s
included language aimed at designers about how to mitigate for
potential wrong-way maneuvers, and the current edition of the
Green Book continues to touch on these issues.
1
Additionally,
various states have sporadically examined WWD crashes and
related countermeasures within their respective jurisdictions
since the 1970s. However, aside from a report on this topic by the
National Transportation Safety Board (NTSB), most recently in
2012, there is scant literature that addresses WWD on the national
level in the United States. One previous analysis of WWD fatal
crashes on U.S. highways based on the Fatality Analysis Reporting
System (FARS) database in 2008 revealed that about 350 fatalities
and thousands of injuries were reported in the crashes caused by
WWD.
2
e objectives of this study are to give an overview of
WWD fatal crashes in the United States., identify general trends,
and determine key contributing factors.
Literature Review
In the United States, few past studies focused on WWD crashes at the
national level. Most of the existing research has been limited to experience
within a particular state or jurisdiction. Taken together, the following
studies help to identify likely WWD trends in the United States.
A 2012 Special Investigation Report by NTSB summarized past
investigations of nine fatal WWD crashes from dierent states
going back to 1968.
3
It concluded that impairment by alcohol is
a major cause of WWD crashes. NTSB also issued recommenda-
tions to dierent transportation agencies and law enforcement
on various countermeasures for reducing WWD crashes.
Many states and local agencies have conducted studies to examine
factors that contribute to WWD crashes, including the role of alcohol
and/or drug involvement by drivers; the proportion of elderly drivers;
the role of the driver’s sex; and the crash time, in order to determine
appropriate engineering and enforcement countermeasures. Some of
their ndings are summarized as follows:
Copelan reported that impaired drivers on California freeways
accounted for almost 60 percent of all wrong-way crashes and
almost 77 percent of fatal WWD crashes from 1983 to 1987.
4
A Washington State Department of Transportation (DOT) study
found that 50 percent of the 30 WWD crashes in the I-82 Yakima-
to-Tri-Cities corridor study were alcohol- or drug-related.
5
Researchers in Indiana determined that out of 77 WWD crashes
in the 1970 to 1972 time period, 42 involved driving under the
inuence (DUI).
6
www.ite.org August 2014 41
feature |
Texas DOT used data from previous studies and conducted a
detailed study of 4 years of WWD crashes on Texas freeways, and
found that almost 61 percent of WWD crashes occurred as a result
of using alcohol and/or drugs and 80 percent occurred at night.
7
Zhou et al. examined WWD crashes on Illinois freeways over
a 6-year period from 2004 to 2009 and concluded that a large
portion of the 217 actual WWD crashes occurred on weekends;
approximately 60 percent of the crashes were the fault of impaired
drivers, and older drivers and male drivers were overrepresented
in the WWD crashes compared to all other fatal crashes.
8
At a substate level, an analysis of WWD crashes by the North
Texas Tollway Authority revealed that from 2007 to 2009,
almost 94 percent involved some degree of alcohol or controlled
substance impairment and occurred between 11 p.m. and 4 a.m.
9
e literature review results imply that an overview of WWD fatal
crashes at the national level will be valuable to better understand
and dene the challenges associated with WWD, as well as to
compare and contrast WWD among the states. is overview
will also serve as a catalyst for sustained eorts to reduce WWD
fatalities in the United States.
Data Collection
Crash Database
FARS is a nationwide census of fatal trac crashes maintained by
the National Highway Trac Safety Administration (NHTSA).
Crashes that involve motor vehicles traveling on a public trac way
and cause the death of the driver, an occupant, or a non-occupant
(such as pedestrians) within 30 days of the crash are included in the
FARS database.
10
Each state gathers data from various sources such
as police crash reports, driver licensing les, vehicle records, death
certicates, and emergency medical service reports, converts them
into a common format, and transmits the data to NHTSA to expand
the FARS database.
Data Extraction Process
To extract WWD-related data from the FARS database les, the
following elds and corresponding attribute codes were used:
1. Roadway Function Class eld, inclusive of
attribute codes for
Rural Principal Arterial Interstate, Rural Principal Arterial
Other, Urban Principal Arterial Interstate, Urban Principal
Arterial Other Freeways or Expressways, and Urban Other
Principal Arterial
;
2. Tracway Flow eld, inclusive of attribute codes for Divided
Highway—Median Strip (Without Trac Barrier), Divided
Highway—Median Strip (With Trac Barrier),
One-way Tracway, and Entrance/Exit Ramp;
3. Sequence of Events (1 through 6) elds, exclusive of attribute
code for Cross Median/Centerline crashes;
4. Violations Charged (1 through 3) elds, inclusive of attribute
codes for Driving Wrong Way on One-way Road and Driving on
Le—Wrong Side of Road; and
5. Driver-Related Factors (1 through 4) elds, inclusive of attribute
codes for Driving Wrong Way on One-way Trac and Driving
on Wrong Side of Road (Intentional or Unintentional).
Cases that satisfy the criteria specied above are considered as
WWD fatal crashes. Each case report consists of various character-
istics of crashes, drivers, vehicles, and persons involved in the fatal
crash, such as number of fatalities, crash location, driver condition,
and driver age and gender. Data were extracted for an 8-year period
covering the years 2004 through 2011.
WWD Data Analysis
National Fatal Crashes and Fatalities
According to the WWD data, an average of 269 fatal crashes resulting
in 359 deaths occurred annually in the United States during the study
period. In terms of magnitude, WWD fatal crashes and fatalities
are comparable to those occurring at highway-rail grade crossings
Total Number o
f
Fatalities
Number of WWD Fatalities
2004 2005 2006 2007 2008 2009 2010 2011
42836 43510 42708 41259 37423 33883 32999 32367
397 336 372 370 370 334 336 357
Total number of fatal crashes
WWD fatal crashes
0
10000
20000
30000
40000
50000
200
400
600
800
1000
0
5000
10000
15000
20000
25000
30000
35000
40000
200
400
600
800
1000
Tota
l
Num
b
er o
f
Fata
l
Cras
h
es
Number of WWD Fatal Crashes
2004 2005 2006 2007 2008 2009 2010 2011
38444 39252 38648 37435 34172 30862 30196 29757
238 253 270 266 271 249 276 281
Total number of fatal crashes
WWD fatal crashes
Figure 1. U.S. Overall Fatal Crashes vs. WWD Fatal Crashes Figure 2. U.S. Overall Fatalities vs. WWD Fatalities
42 August 2014 ite journal
Table 1. U.S. WWD Fatal Crashes by State (2004–2011)
State
Average Frequency
Percent of U.S. Total
Texas 38 14%
California 26 10%
Florida 21 8%
Pennsylvania 11 4%
Missouri 10 4%
Mississippi 9 3%
Tenne ssee 8 3%
Illinois 8 3%
Georgia 8 3%
Arizona 8 3%
Alabama 7 2%
Michigan 6 2%
New Jersey 6 2%
Oklahoma 6 2%
Washington 6 2%
Virginia 5 2%
Louisiana 5 2%
North Carolina 5 2%
Kansas 5 2%
Colorado 5 2%
New York 5 2%
Ohio 4 2%
South Carolina 4 2%
Nevada 4 1%
Maryland 4 1%
Arkansas 4 1%
State
Average Frequency
Percent of U.S. Total
Minnesota 4 1%
West Virginia 4 1%
Utah 3 1%
Connecticut 3 1%
Iowa 3 1%
Kentucky 3 1%
Massachusetts 3 1%
Oregon 3 1%
Indiana 2 1%
Wisconsin 2 1%
Idaho 2 1%
Montana 1 1%
Rhode Island 1 0%
Delaware 1 0%
New Mexico 1 0%
Hawaii 1 0%
Wyoming 1 0%
New Hampshire 1 0%
South Dakota 1 0%
Maine 1 0%
Vermont 0 0%
North Dakota 0 0%
Alaska 0 0%
Nebraska 0 0%
District of Columbia 0 0%
(HRGX). However, unlike the HRGX experience, WWD has not been
declining over the years. Figures 1 and 2 compare the overall trends
of trac fatal crashes and fatalities to those that are specically
WWD-related in the United States over the 8-year study period.
Figures 1 and 2 both illustrate that overall trac fatal crashes
and fatalities declined substantially in the study period, decreasing
about 4 percent per year on average, while WWD fatal crashes and
fatalities remained fairly constant. at there has been no sustained,
coordinated national campaign to address WWD may somewhat
explain this dierence. It also suggests that such an eort could
make a positive dierence by helping to align WWD with overall
national trac safety trends.
State-Level Experience
Sorting the data by state reveals that WWD fatal crashes and
fatalities are spread very unevenly across the United States, with
some states experiencing none at all. However, it is notable that the
top 10 states account for more than 50 percent of the national totals.
Table 1 lists the fatal crashes by state, and Table 2 lists the fatalities
by state in the United States.
Table 2. U.S. WWD Fatalities by State (2004–2011)
State
Average Frequency
Percent of U.S. Total
Texas 51 14%
California 35 10%
Florida 28 8%
Pennsylvania 14 4%
Missouri 13 4%
Illinois 12 3%
Georgia 11 3%
Mississippi 11 3%
Tenne ssee 11 3%
Arizona 10 3%
Alabama 9 3%
Michigan 8 2%
Oklahoma 8 2%
Louisiana 7 2%
New Jersey 7 2%
New York 7 2%
North Carolina 7 2%
Virginia 7 2%
Washington 7 2%
Colorado 6 2%
Kansas 6 2%
Ohio 6 2%
Arkansas 5 1%
Maryland 5 1%
Minnesota 5 1%
Nevada 5 1%
State
Average Frequency
Percent of U.S. Total
South Carolina 5 1%
Utah 5 1%
West Virginia 5 1%
Connecticut 4 1%
Iowa 4 1%
Kentucky 4 1%
Massachusetts 4 1%
Oregon 4 1%
Idaho 3 1%
Indiana 3 1%
New Mexico 3 1%
Wisconsin 3 1%
Delaware 2 1%
Montana 2 1%
Hawaii 1 0.3%
Maine 1 0.3%
New Hampshire 1 0.3%
Rhode Island 1 0.3%
South Dakota 1 0.3%
Vermont 1 0.3%
Wyoming 1 0.3%
North Dakota 0 0.0%
Alaska 0 0.0%
Nebraska 0 0.0%
District of Columbia 0 0.0%
www.ite.org August 2014 43
To measure the severity of this type of crash, the number of
lost lives, the most serious consequence of the trac crash, per
WWD fatal crash was calculated. Analyses showed that on average
four persons die in every three WWD fatal crashes in the United
States (359/269). Another measure used to assess the signicance
of the WWD issue is the frequency of its fatalities out of the total
number of crash fatalities in each state. For the study period
(2004–2011), the average number of total freeway fatalities was
15,738 in the United States. Accordingly, WWD fatalities account
for about 2.3 percent of total fatalities occurring on freeways as
a national average. Fatality rates vary widely between states; for
example, there were 48 WWD fatalities per 1,000 people killed
in fatal crashes in Mississippi as compared to no WWD fatality
in the District of Columbia (see Table 3). (Using Mississippi
as the example, a total of 1,810 trac fatalities occurred in the
state during the 8-year study period, of which 87 fatalities were
WWD-related, which is 4.8 percent of the total and the highest
among all states.)
Crash Locations
e location of WWD fatal crashes can rst be characterized as
rural or urban. e data analyzed for this study, summarized in
Table 4, show that urban roads represent a higher proportion of
WWD fatal crashes, with about 57 percent occurring on urban
roads and the remaining 43 percent occurring on rural roads. In
contrast, according to the Federal Highway Administration, only
about 24 percent of highway miles are characterized as urban, as
shown in Table 5.
11
It is important to note that for classifying roads,
the designation of “urban” includes urbanized and small urban
areas. ese relationships are illustrated graphically in Figure 3.
Driver Characteristics
Impaired Driving
According to Trac Safety Facts (TSF), published annually by
NHTSA, a driver (the operator of any motor vehicle, including
a motorcycle) is considered to be alcohol-impaired if their blood
alcohol concentration (BAC) equals or exceeds 0.08 percent.
12
Furthermore, TSF shows that the percentage of drivers in
fatal crashes with a BAC≥0.08 percent was reported to be 21
percent and 22 percent for 2001 and 2010, respectively, while
the percentage of fatalities where driver BAC was ≥0.08 percent
was 31 percent in both 2001 and 2010. Each year there is a large
State
Percentage
Mississippi 4.8%
Delaware 3.8%
Utah 3.7%
West Virginia 3.7%
Texas 3.6%
Nevada 3.5%
Indiana 3.4%
Washington 3.3%
Kansas 3.2%
Rhode Island 3.2%
New Hampshire 3.1%
Missouri 3.1%
Minnesota 3.0%
Connecticut 3.0%
Oklahoma 2.8%
Pennsylvania 2.7%
Tenne ssee 2.7%
State
Percentage
Illinois 2.6%
Massachusetts 2.6%
Idaho 2.4%
Vermont 2.4%
Louisiana 2.3%
Iowa 2.3%
Alabama 2.3%
Hawaii 2.3%
Colorado 2.2%
Arizona 2.2%
New Jersey 2.1%
Georgia 2.1%
Michigan 2.1%
Virginia 2.0%
Arkansas 2.0%
Maine 2.0%
Oregon 2.0%
State
Percentage
Florida 1.9%
California 1.8%
North Carolina 1.7%
Maryland 1.7%
Ohio 1.6%
Montana 1.6%
South Carolina 1.4%
South Dakota 1.4%
Kentucky 1.3%
Wisconsin 1.3%
New York 1.2%
Wyoming 1.1%
New Mexico 1.1%
North Dakota 0.5%
Alaska 0.3%
Nebraska 0.1%
District of
Columbia 0.0%
Table 3. WWD Fatalities as a Percentage of Total Fatalities for States
(2004–2011)
Table 4. WWD Fatal Crashes by Rural and Urban Areas (U.S.)
2004 2005 2006 2007 2008 2009 2010 2011 Average Percentage
Rural Areas 12 7 97 115 110 117 116 12 6 115 117 43.5%
Urban Areas 156 156 155 15 6 154 133 15 0 16 6 152 56. 5%
Tot al 283 253 270 266 271 249 276 281 269 100%
Table 5. Highway Miles by Rural and Urban Areas (U.S.)
2000 2002 2004 2006 2008 Average Percentage
Rural Areas 3,091,733 3,079,757 3,003,441 2,990,482 2,980,333 3,029,149 75.7%
Urban Areas 859,368 901,913 994,021 1,042,526 1,079,007 975,367 24.3%
Tot al 3,951,101 3,981,670 3,997,462 4,033,008 4,059,340 4,004,516 100%
44 August 2014 ite journal
amount of missing data (about 60 percent) for the Alcohol Test
Result variable in FARS due to drivers’ refusal to submit to an
alcohol test.
12,13
As part of this WWD study, data from the FARS Driver and
Person reports were analyzed for information related to driver
intoxication by or consumption of alcohol or drugs. Specically,
the following FARS elds and attributes were examined:
1. From the Person report, inclusive of the Police-Reported
Alcohol Involvement and Drug Involvement, which is based on
the investigating ocer’s judgment;
2. From the Driver report, inclusive of the Investigating Alcohol
Test Result, which is restricted to drivers with objective BAC
measurement cited by the investigating ocer;
3. From the Driver report, inclusive of the Driver-Related Factor
5, which is used if the driver was reported to be Under the
Inuence of Alcohol, Drugs, or Medication; and
4. From the Driver report, inclusive of Violations Charged
Codes for Driving While Intoxicated (Alcohol or Drugs), BAC
Above Limit (Any Detectable BAC for CDLs), Driving While
Impaired, Driving Under Inuence of Substance Not Intended
to Intoxicate, Drinking While Operating, Illegal Possession
of Alcohol or Drugs, Driving With Detectable Alcohol, and
Alcohol, Drug, or Impairment Violations Generally.
When one or more of the above conditions are satised for a driver
involved in a WWD fatal crash, it is counted as an impaired driving
or DUI event.
e frequency and percentage of DUI-related WWD fatal
crashes from the study period are provided in Table 6. e
long-term averages suggest that about 58 percent of WWD crashes
are DUI-related, which is nearly twice the rate of 31 percent alcohol
or drug involvement for all fatal crashes as reported by NHTSA.
Past studies for various states and various years reported similar
results.
3,7,9
is arms that impaired driving is a major factor
in WWD fatal crashes and suggests that eorts to curb WWD
should include education and enforcement activities targeting DUI
behaviors, in addition to deploying engineering countermeasures.
Gender and Age
Strategies to improve trac safety are oen informed by human
factors such as gender and age. is information also tends to be
the most commonly available demographic data in crash reports.
14
erefore, a focus on these variables is to be expected.
Sorting the WWD fatal crash data by gender reveals that
during the study period, male drivers outnumbered female drivers
by more than 2 to 1, as shown in Table 7. Further analysis that
considered gender and DUI status found that impaired male
drivers outnumbered impaired female drivers by nearly 3 to 1.
While these ndings are consistent with other past state-level
research indicating that males are generally more likely to
be involved in fatal crashes, the dierences are even more
pronounced for W WD.
Table 6. U.S. WWD Fatal Crashes and Drivers Under the Inuence
2004 2005 2006 2007 2008 2009 2010 2011 Average
DUI 131 147 156 170 162 151 150 168 154
Tot al 283 253 270 266 271 249 276 281 269
Percentage 46% 58% 58% 64% 60% 61% 54% 60% 58%
Table 7. U.S. WWD Fatal Crashes by Gender
2004 2005 2006 2007 2008 2009 2010 2011 Average Percentage
Male 194 183 204 185 204 175 175 205 191 71%
Female 89 70 66 81 67 74 101 76 78 29%
Tot al 283 253 270 266 271 249 276 281 269 100%
Rural Areas Urban Areas
Percentage of highway miles
Percentage of WWD crash
0
20
40
60
80
100
Figure 3. Percentages of Highway Miles and WWD Crashes by Rural and
Urban Areas
www.ite.org August 2014 45
When sorting the WWD fatal crash data by age, dierentiating
between “younger” and “older” drivers was important, as some
previous WWD studies have found age-related overrepresentation.
erefore, the following three groupings were used: Younger than
Age 24, Ages 24 to 64, and Older than Age 64.
is analysis shows that about 15 percent of wrong-way drivers
are age 65 or older, as indicated in Table 8. In their report, NTSB
reported the same percentage of wrong-way drivers over the age
of 70 between 2004 and 2009.
3
A broader review of U.S. freeway
fatal crashes for the same 8-year period shows that an average of
just over 10 percent involved an older driver. Comparing these
results suggests that older drivers are overrepresented in WWD
fatal crashes. Cooner et al. and Zhou et al. found similar results for
dierent data years in Illinois and Texas.
7,8
For the other age groups, there does not appear to be the same
extent of overrepresentation.e “younger” driver age group
(younger than age 24) represents 18 percent of WWD fatal crashes,
as shown in Table 8, and represents 19 percent of all freeway fatal
crashes for the same time period.
e further driver characteristic-based analysis accounted
for all three of the above variables—gender, age, and DUI, and
the results are provided in Table 9. According to these results,
there were signicantly fewer “older” drivers that were involved
in WWD fatal crashes and also impaired—only 50 out of a total
of 329 older driver WWD fatal crashes, or 16 percent. However,
nearly two-thirds of WWD fatal crash drivers under the age of
65 were also reported as impaired by alcohol or drugs. An earlier
study that examined data for the years 1982–2000 also reported
that drivers age 65 and above had the lowest proportion of
alcohol-involved fatal crashes.
13
ese ndings can serve to inform
the DUI education and enforcement strategies for reducing WWD
fatal crashes, suggesting that a focus on age groups younger than
65 would be most benecial. Conversely, since WWD fatal crashes
with older drivers tend to be less likely to also be DUI, considering
engineering countermeasures that are known to benet older
drivers may be quite helpful in mitigating WWD risk.
Conclusions
is article summarizes the results of a study on WWD fatal
crashes and fatalities in the United States for an 8-year period
(2004–2011) to give an overview of general WWD-related fatal
crash trends and characteristics. e analyses revealed that
Table 9. U.S. WWD Fatal Crashes by Age, Gender, and DUI
DUI Non-DUI
Male Female Male Female
Year Age <24 Age
24–65 Age >65 Age <24 Age
24–65 Age >65 Age <24 Age
24–65 Age >65 Age <24 Age
24–65 Age >65
2004 177066320146819122811
2005 327751119312401781910
2006 2290692901352211918
2007 19 92 5 14 39 1 10 4 0 19 6 14 7
2008 20 102 7 10 22 1 6 41 28 4 13 17
2009 18 84 3 12 34 0 7 45 18 4 16 8
20101883393619323072919
2011 25 9 2 6 13 29 3 11 4 6 2 5 7 12 12
Total
171 690 41 84 240 9 8 2 364 177 49 14 0 102
902 333 623 291
1,235 914
Table 8. U.S. WWD Fatal Crashes by Age
Age of Drivers 2004 2005 2006 2007 2008 2009 2010 2011 Average Percentage
Age<24 49 63 45 49 40 41 43 56 48 18%
Age 24–65 198 155 180 185 178 179 180 179 17 9 67 %
Age>65 36 35 45 32 53 29 53 46 41 15%
Tot al 283 253 270 266 271 249 276 281 269 100%
46 August 2014 ite journal
although the total numbers of trac fatal crashes and fatalities
has fallen by roughly 4 percent per year, the percentages of WWD
fatal crashes and fatalities have remained nearly unchanged over
the same period. It was found that Texas, California, and Florida
account for the highest number of WWD fatal crashes and fatalities
and represent almost one-third of the national totals.
Several additional valuable ndings:
e proportions of WWD fatal crashes that occurred on urban
and rural roads were 57 percent and 43 percent, respectively;
e rate of impaired driving for WWD fatal crashes is about 58
percent, which is nearly twice the rate of 31 percent for all types
of fatal crashes;
Across all age groups, and whether impaired or not, involvement
rates for WWD fatal crashes are higher for male drivers versus
female drivers at a ratio of more than 2 to 1; and
Older drivers appear to be overrepresented, accounting for 15
percent of WWD fatal crashes and just over 10 percent of all
freeway-related fatal crashes.
e results of this study are intended to serve as an important
resource for national and state eorts to reduce WWD incidents
and deaths. e authors hope that the understanding of WWD will
continue to evolve through further research of trac safety data, as
well as robust evaluations of safety strategies designed to speci-
cally target WWD. itej
References
1. American Association of State Highway and Transportation Officials. A
Policy on the Geometric Design of Highways and Streets. Washington, DC:
AASHTO, 2011.
2. North Texas Tollway Authority. “Keeping NTTA Roadways Safe: Wrong-Way
Driver Task Force Staff Analysis. Presented to the NTTA Board of Directors,
September 23, 2009.
3. National Transportation Safety Board. Wrong-Way Driving. Highway
Special Investigation Report. NTSB/SIR-12/01, PB2012-917003, Notation
8453. December 11, 2012.
4. Copelan, J. Prevention of Wrong-Way Accidents on Freeways. Report No.
FHWA/CA-TE-89-2. California Department of Transportation, Traffic
Operations Division, 1989, p. 95.
5. Moler, S. “Stop. You’re Going the Wrong Way!” Public Roads, September/
October 2002, pp. 24–29 [Online]. Available: www.tfhrc.gov/
pubrds/02sep/06.htm. [Accessed September 19, 2013.]
6. Scifres, P. and R. Loutzenheiser. “Wrong-Way Movements on Divided
Highways.” Purdue University Joint Highway Research Project No. JHRP-
13–75, 1975, p. 46.
7. Cooner, S.A., A.S. Cothron, and S.E. Ranft. Countermeasures for Wrong-way
Movement on Freeways:Guidelines and Recommended Practices. FHWA/TX-
04/4128-2. Texas A&M Transportation Institute, January 2004.
8. Zhou H., J. Zhao, R. Fries, L. Wang, and M.R. Gahrooei. “Investigation of
Contributing Factors Regarding Wrong-way Driving on Freeways Final
Report.” Submitted to ICT, June 2012.
9. Kaminski Leduc, J.L. Wrong Way Driving Countermeasures. Old Research
Report 2008-R-0491. September 22, 2008.
10. U.S. Department of Transportation, National Highway Traffic Safety
Administration, and National Center for Statistics and Analysis. Agency
Procedures for Release and Security of Research Data Collected Under the
Fatality Analysis Reporting System (FARS). Washington, DC: March 2004.
11. Federal Highway Administration. 2010 Status of the Nation’s Highways,
Bridges, and Transit: Conditions & Performance, “Chapter 2: System
Characteristics.” [Online]. Available: www.fhwa.dot.gov/policy/2010cpr/
chap2.htm#body. [Accessed September 19, 2013.]
12. National Highway Traffic Safety Administration. Traffic Safety Facts, 2010
Data: State Alcohol-Impaired Driving Estimates. DOT HS 811 612. April
2012.
13. National Highway Traffic Safety Administration. Transitioning to Multiple
Imputation—A New Method to Impute Missing Blood Alcohol Concentration
(BAC) values in FARS. Technical Report, DOT HS 809 403. National Center
for Statistics and Analysis Research and Development, January 2002.
14. Haight, F.A. “The Place of Safety Research in Transportation Research.
Transportation Research, Vol. 19 (1985): 373–376.
Huaguo Zhou, Ph.D., P.E. is an associate professor
in the Department of Civil Engineering at Auburn
University. He earned a doctorate of philosophy
degree in civil engineering from the University of
South Florida. He is a fellow of ITE and a licensed
professional engineer in Florida.
Jerey Shaw, P.E., PTP, PTOE manages the
Intersection Safety Program for the Federal Highway
Administration, Oce of Safety. He is a registered
professional engineer in Illinois and has been board-cer-
tied as a Professional Trac Operations Engineer and
Professional Transportation Planner. He is a member of
ITE and past chair of the Transportation Safety Council.
Fatemeh Baratian-Ghorghi, Ph.D., S.M.ASCE is
currently a doctoral student at Auburn University. She
works under the supervision of Dr. Huaguo Zhou. Her
main research interests include trac operations and
safety, travel behavior modeling, statistical modeling
of crash data, and intelligent transportation systems.
She is a student member of ITE.
www.ite.org August 2014 47
... According to the Fatality Analysis Reporting System crash database, from 2004 to 2011, an average of 269 WWD fatal crashes resulted in 359 fatalities annually in the United States (7). Even though total fatalities declined, WWD fatalities remained constant over the past years, as shown in Figure 1. ...
... In fact, 60% or above WWD crash occurs due to driving under the influence of alcohol or drugs (7). WWD crash in urban area is higher than in rural areas. ...
... Time of the day is also an essential contributing factor in WWD crash analysis. WWD crash more frequently occurs between 12 AM to 6 AM (7,8). Male drivers are overrepresented in WWD crashes compared with their female counterparts. ...
Thesis
Full-text available
The wrong way driving (WWD) crash differs significantly from all other crashes in terms of severity. Despite much valuable research on WWD crash severity, there is no robust study on developing a WWD crash severity distribution table that could be used for crash predictive methods by the Highway Safety Manual (HSM). HSM has severity distribution tables for all crashes that are used in quantitative safety analyses. Using the same tables for quantitative WWD crash analysis may produce inaccurate results since WWD crash severity is significantly different. Current severity distribution tables in the HSM have been prepared using Highway Safety Information System (HSIS) crash database. Unfortunately, a structured guideline that could extract true WWD crashes from the HSIS database is still absent. Therefore, the current study attempts to extract true WWD crashes from HSIS crash database. And using the extracted WWD crash database, the study proposes different WWD crash severity distribution tables for freeway facilities. Finally, the study shows practical applications of extracted WWD crash database and proposed WWD crash severity distribution tables. These applications are expected to help future researchers of WWD crashes, safety analysts, and transportation policymakers.
... Wrong-way driving (WWD) crashes happen when a driver inadvertently or deliberately 2 drives against the main direction of traffic flow (1). Researchers have been working relentlessly to 3 combat the WWD phenomenon since the 1950s, when the first interstate highway system was built 4 (2). Still, FARS (Fatality Analysis and Reporting System) data from 2004 to 2011 indicates nearly 5 a constant trend of WWD crashes (3). ...
... Researchers have been working relentlessly to 3 combat the WWD phenomenon since the 1950s, when the first interstate highway system was built 4 (2). Still, FARS (Fatality Analysis and Reporting System) data from 2004 to 2011 indicates nearly 5 a constant trend of WWD crashes (3). The most recent FARS data used in this study also supports 6 this trend. ...
... Studies on non-access control highways are also available since a 1 certain percentage of WWD movements occur when a driver takes a U-turn on the mainline. (9). 2 Additionally, previous research has shown that the percentage of WWD crashes in an urban area 3 is higher than in rural areas (3,8,10,11), while 76 % of highway miles are rural, according to 4 FHWA (3). 5 Intensive research disclosed that the main reasons that contribute to WWD crashes are 6 impaired drivers, nighttime, urban areas, early morning hours, male drivers, older drivers, 7 weekend, dark-but-not-lighted, geometric design of exit/entrance road and crossroads, location of 8 W.W. sign, and pavement arrow, high annual average daily traffic (AADT) on entrance ramp 9 (1,4,9,10,(12)(13)(14)(15)(16). Numerous methods are applied to find crash contributing factors: descriptive 10 statistics, logistic regression, multiple correspondence analysis (MCA), machine learning, and data 11 mining approaches. ...
Conference Paper
Full-text available
Extensive research has been conducted on Wrong-Way Driving (WWD) prone crash locations, crash contributing factors and safety countermeasures. Still, the number of WWD crashes remains nearly constant over the past years, necessitating further investigation from different perspectives. Past studies identified various crash contributing factors that directly contributed to WWD crash frequency and severity. Prime factors include driver characteristics, environmental and temporal characteristics, and interchange layouts. However, the impact of locality and nonlocality on WWD crash propensity is seldom investigated. Therefore, this study explores the major WWD crash contributing factors related to local and non-local drivers. A total of 1,048 WWD fatal crashes from 2015 to 2017 were collected from the Fatality Analysis Report System (FARS) database. Drivers’ locality and non-locality were defined using the distance traveled from home to crash location using the Geographic Information System. Descriptive statistics and a multinomial logit model were developed to analyze the significant contributing factors specific to driver groups. The results demonstrated that factors such as rural settings, unprotected median types, and dark but not lighted conditions significantly contributed to WWD fatal crashes when the driver was non-local. In contrast, local drivers are more prone to be involved in WWD crashes in urban areas and while driving under the influence of alcohol or drugs. Based on the results, different safety countermeasures related to WWD crashes were recommended targeting local and non-local drivers.
... On average, WWD crashes account for 357 fatalities /year in the US [2]. National and state level data associate WWD crashes to a variety of factors including; age impaired driving, roadway geometry and orientation, lighting, weather conditions, day of the week, and time of day [3][4][5][6][7][8][9][10][11]. High-speeds and impaired driving; with the high propensity for head-on collisions is cause for multiple fatalities (to persons in either the "wrong-way" or "right-way" vehicles) and significant vehicle damage. ...
Chapter
The fatality rate in Wrong-Way Driving (WWD) crashes is 12 times greater than all other roadway crash types. Recently, WWD countermeasures that provide alerts of oncoming wrong-way vehicles to “right-way” drivers are gradually being implemented across the US. Using a driving simulator, this study examined the behavior/performance of “right-way” drivers during a WWD event; and subsequently seeks to evaluate the potential effectiveness of an ITS countermeasure that targets the driver in the legal direction of travel. Findings showed that, overall; the use of an overhead dynamic message sign (DMS) “wrong-way warning” system can prevent fatal WWD crashes and/or reduce their severity. The likelihood for a “right-way” driver to avoid an oncoming WWD is 19.4 times greater when they are provided a DMS alert. Participants demonstrated use of lower brake force; indicating they were cautious as they approached the WWD. Even among participants that did collide with the WWD, their collision velocities were lower.
Article
Roadway improvements to reduce the frequency of crashes are of the utmost priority to transportation agencies. To a great extent, the implementation of improvement programs depends on the reliable identification of roadway segments with high crash risk. Among all crash types, wrong-way driving (WWD) crashes are considered random in nature and are a major safety concern. The Federal Highway Administration defines WWD specifically for high-speed divided highways and access ramps. This definition excludes all other roadway classifications when a crash occurs in the opposing direction to the legal flow of traffic. Screening 5 years of crash data in Minnesota revealed that WWD resulted in crashes on other types of roadway functional classes. This work aimed to (1) introduce a new term/acronym to the literature for driving in the wrong direction (DWD) on all roadway functional classes, (2) apply a set of count data models to estimate the occurrence of DWD crashes, (3) identify roadway geometric features of high-risk segments for DWD crashes, (4) investigate random effects of covariates due to unobserved factors, and (5) calculate elasticity effects of variables. The final models’ specifications indicate that the negative binomial (NB) mixed effect model was found to be the best-fit model. Focusing on DWD crashes, we uncovered the factors contributing to higher DWD crash-risk segments: log of average annual daily traffic (AADT), number of lanes, sidewalk, and shoulder type. The change in frequency of crashes is also investigated using marginal effects, and safety interventions for preventing DWD crashes are also discussed. Transportation agencies can use the findings of this research, in terms of contributing factors and their relative effects on DWD crashes, to deploy appropriate countermeasures at high-risk locations.
Article
Introduction: Partial cloverleaf (parclo) interchanges with closely spaced parallel entrance and exit ramps are more prone to wrong-way driving (WWD) compared to other interchange types. In this study, a logistic regression model was developed to predict the risk of WWD at the exit ramp terminals of parclo interchanges. Method: The logistic regression model was developed using Firth's penalized likelihood techniques based on the predictor variables such as exit ramp geometric design features, wrong-way related traffic control devices, area type, and traffic volume. Results: According to the model, the significant predictors of WWD at parclo exit ramp terminals include corner radius from crossroad to entrance ramp, type of median on crossroad, width of median on two-way ramp, channelizing island, distance to the nearest access point, "Keep Right" sign, wrong-way arrow, intersection signalization, and traffic volume at the exit and entrance ramps. This model was used to conduct network screening for all the exit ramp terminals of parclo interchanges in Alabama and Georgia to identify high-risk locations in these two states. Seven high-risk locations were monitored by video cameras for 48-hours to observe the occurrences of WWD incidents. Results suggest that two locations in Alabama and two locations in Georgia experienced multiple WWD incidents within 48-hours of a typical weekend. Conclusion: The observation of WWD incidents at high-risk locations demonstrates strong evidence that the model could identify the exit ramp terminals with high risk of WWD. Practical applications: Transportation agencies can use this model to assess the risk of WWD at the exit ramp terminals within their jurisdictions and identify the high-risk locations for countermeasures implementation.
Article
Hundreds of fatal accidents occur each year due to wrong‐way driving (WWD). Although several methods have been developed to detect WWD using existing closed‐circuit television (CCTV) data, they all require manual recalibration whenever a camera rotates, and are thus not scalable across statewide CCTV networks. This paper, therefore, proposes an end‐to‐end deep‐learning‐based model that considers camera orientation as a variable, detecting camera rotation automatically and learning new decision criteria accordingly using a neural network model. We show that our proposed solution can detect WWD with a precision of 0.99 and a recall of 0.97. Due to its cheap computational cost and high error tolerance, our solution is easily scalable for statewide surveillance on a real‐time basis to help decision‐makers reduce fatalities due to WWD.
Article
The objective of this study is to identify clusters of contributing factors associated with the occurrence of wrong-way driving (WWD) fatal crashes on freeways using the multiple correspondence analysis (MCA) method based on the Burt matrix with an adjustment of inertias. A total of 14 years (2004–2017) of WWD fatal crash data were extracted from the National Highway Traffic Safety Administration (NHTSA) Fatality Analysis Reporting System (FARS) database. A standard procedure was developed to extract the WWD crash information (including a total of 3,817 crashes) on freeways from the FARS. Each crash contains various characteristics of crashes, vehicles, and drivers, for example, crash time, crash location, vehicle type, driver age, and so forth. The MCA analysis used a total of 19 key variables with 67 defined categories. The results of this study indicate that four clusters of factors which, when combined, might contribute to the occurrence of some WWD fatal crashes. These four clusters were: (1) younger drivers, driving under the influence (DUI), midnight/early morning, lower speed limit (45–50 mph), urban areas, and street lighting; (2) older drivers, non-DUI drivers, and daylight; (3) dark/no light, 18:00 to 23:59 p.m., higher speed limits (65 mph or more), and rural areas; and (4) rain/snow/sleet/hail/fog, and wet road surface.
Article
Wrong-way Driving (WWD) is the movement of a vehicle in a direction opposite to the one designated for travel. WWD studies and mitigation strategies have exclusively been focused on limited-access facilities. However, it has been established that WWD crashes on arterial corridors are also severe and relatively more common. As such, this study focused on determining factors influencing the severity of WWD crashes on arterials. The analysis was based on five years of WWD crashes (2012–2016) that occurred on state-maintained arterial corridors in Florida. Police reports of 2,879 crashes flagged as “wrong-way” were downloaded and individually reviewed. The manual review of the police reports revealed that of the 2,879 flagged WWD crashes, only 1,890 (i.e., 65.6 %) occurred as a result of a vehicle traveling the wrong way. The Bayesian partial proportional odds (PPO) model was used to establish the relationship between the severity of these WWD crashes and different driver attributes, temporal factors, and roadway characteristics. The following variables were significant at the 90 % Bayesian Credible Interval (BCI): day of the week, lighting condition, presence of work zone, crash location, age and gender of the wrong-way driver, airbag deployment, alcohol use, posted speed limit, speed ratio (i.e., driver’s speed over the posted speed limit), and the manner of collision. Based on the model results, specific countermeasures on Education, Engineering, Enforcement, and Emergency response are discussed. Potential Transportation Systems Management and Operations (TSM&O) strategies for WWD detection systems on arterials to minimize WWD frequency and severity are also proposed.
Article
Although wrong-way driving (WWD) crashes are rare compared with other crashes, the severe outcomes resulting from these crashes make them an important traffic safety issue, especially on freeways. The initial point of origin for the majority of WWD crashes is the exit ramp terminal. During 2009–2013, the exit ramp terminals of full diamond and partial cloverleaf interchanges in Alabama experienced considerably higher frequency of wrong-way entries compared with those in Illinois. In this study, a comparative analysis is conducted between these two states to compare the differences in exit ramp terminal design practices and determine the differences that might have caused the relatively higher frequency of wrong-way entries at the exit ramp terminals in Alabama. The comparative analysis revealed that there is relatively higher usage of certain geometric design features and traffic control devices in Illinois, which contributed to deterring WWD. The lessons learned from this study can potentially benefit the 4E community (i.e., enforcement, education, emergency response, and engineering) in further understanding the best practices on off-ramp terminal design and traffic control devices features that have the potential to reduce WWD crashes on freeways.
Investigation of Contributing Factors Regarding Wrong-way Driving on Freeways Final Report
  • H Zhou
  • J Zhao
  • R Fries
  • L Wang
  • M R Gahrooei
Zhou H., J. Zhao, R. Fries, L. Wang, and M.R. Gahrooei. "Investigation of Contributing Factors Regarding Wrong-way Driving on Freeways Final Report. " Submitted to ICT, June 2012.
Keeping NTTA Roadways Safe: Wrong-Way Driver Task Force Staff Analysis
  • North Texas
  • Tollway Authority
North Texas Tollway Authority. " Keeping NTTA Roadways Safe: Wrong-Way Driver Task Force Staff Analysis. " Presented to the NTTA Board of Directors, September 23, 2009.
Status of the Nation' s Highways, Bridges, and Transit: Conditions & Performance System Characteristics
  • Federal Highway Administration
Federal Highway Administration. 2010 Status of the Nation' s Highways, Bridges, and Transit: Conditions & Performance, " Chapter 2: System Characteristics. " [Online]. Available: www.fhwa.dot.gov/policy/2010cpr/ chap2.htm#body. [Accessed September 19, 2013.]
Report No. FHWA/CA-TE-89-2. California Department of Transportation
  • J Copelan
Copelan, J. Prevention of Wrong-Way Accidents on Freeways. Report No. FHWA/CA-TE-89-2. California Department of Transportation, Traffic Operations Division, 1989, p. 95.
National Highway Traffic Safety Administration Traffic Safety Facts, 2010 Data: State Alcohol-Impaired Driving Estimates
National Highway Traffic Safety Administration. Traffic Safety Facts, 2010 Data: State Alcohol-Impaired Driving Estimates. DOT HS 811 612. April 2012.
Wrong Way Driving Countermeasures
  • J L Kaminski Leduc
Kaminski Leduc, J.L. Wrong Way Driving Countermeasures. Old Research Report 2008-R-0491. September 22, 2008.