Content uploaded by Mahdi Pour-Rouholamin
Author content
All content in this area was uploaded by Mahdi Pour-Rouholamin on Nov 10, 2014
Content may be subject to copyright.
STATISTICAL CHARACTERISTICS OF WRONG-WAY DRIVING CRASHES ON
ILLINOIS FREEWAYS
Huaguo Zhou*
Associate Professor, Dept. of Civil Engineering
Auburn University
Auburn, AL 36849-5337
Phone: 334-844-4320
Email: hhz0001@auburn.edu
Jiguang Zhao
Transportation Engineer, CH2M HILL Inc.
8735 W Higgins Road Suite 400
Chicago, IL 60631
Phone: 773-458-2885
Email: Jiguang.zhao@ch2m.com
Ryan Fries
Assistant Professor, Dept. of Civil & Environmental Engineering
Southern Illinois University Edwardsville
Edwardsville, IL 62026-1800
Phone: 618-650-5026
Email: rfries@siue.edu
Mahdi Pour Rouholamin
Ph.D. Graduate Student, Dept. of Civil Engineering
Auburn University
Auburn, AL 36849
Phone: 618-660-4123
Email: mahdipn@auburn.edu
* Corresponding Author
Word Count = 4,712 + 250×11 Tables and Figures = 7,462
A paper submitted for presentation at the 93
rd
Transportation Research Board Annual Meeting
and publication in the Transportation Research Record: Journal of the Transportation Research
Board
TRB 2014 Annual Meeting Paper revised from original submittal.
Zhou, Zhao, Fries, and Pour Rouholamin
2
ABSTRACT 1
This study collected and analyzed the wrong-way crashes data for a six-year time period, from 2
2004 to 2009, on Illinois freeways. The objective of the study was to characterize the statistical 3
characteristics of wrong-way crashes from these three aspects: crash, vehicle, and person. The 4
temporal distributions, geographical distribution, roadway characteristics, and crash 5
characteristics were analyzed for wrong-way crashes. The wrong-way driver demographic 6
information, driver physical condition, and driver injury severity were analyzed for wrong-way 7
drivers. The vehicle characteristics, vehicle operation, and collision results were analyzed for 8
wrong-way driving vehicles. General statistical characteristics of wrong-way crashes were 9
analyzed, and the findings revealed that a large proportion of wrong-way crashes occurred during 10
the weekend from midnight to 5 a.m. Approximately 80 percent of wrong-way crashes were 11
located in urban areas. Nearly 70 percent of wrong-way vehicles were passenger cars. 12
Approximately 60% of wrong-way drivers were driving under the influence (DUI). Of those, 13
nearly 50% were confirmed to be impaired by alcohol, about 5% were impaired by drugs, and 14
more than 3% had been drinking. Wrong-way entry points were analyzed for different 15
interchange types as well. Compressed diamond interchanges, SPUI, partial cloverleaf 16
interchanges, and freeway feeders had the highest wrong-way crash rates (wrong-way crashes 17
per 100 interchanges per year). The contribution of this study was a new method to predict the 18
possible wrong-way entry points and rank the high-frequency crash locations for field reviews 19
based on the number of recorded or estimated wrong-way freeway entries. 20
21
Keywords: Wrong-Way Driving, Crashes, Statistical Characteristics, Freeways 22
TRB 2014 Annual Meeting Paper revised from original submittal.
Zhou, Zhao, Fries, and Pour Rouholamin
3
INTRODUCTION 1
Driving the wrong-way on freeways has been a consistent traffic safety problem since the 2
interstate system was opened in the 1950s. Herein, a wrong-way driver is considered to be 3
traveling in the wrong direction on a physically separated motorway or a driver traveling in the 4
opposite direction along a one-way street (1). Recent statistics show that about 350 people are 5
killed each year nationwide due to wrong-way driving crashes based on the National Highway 6
Traffic Safety Administration (NHTSA)’s Fatality Analysis Reporting System (FARS) (2). In 7
Illinois, 217 freeway wrong-way crashes occurred from 2004 to 2009 resulting in 44 killed and 8
248 injured (3). The IDOT decided that an in-depth investigation of wrong-way crashes could 9
provide a better understanding of such events. The purpose of this research was to review such 10
severe crashes in depth, to determine what contributing factors are most commonly involved, and 11
to generate ideas to consider in reducing the frequency and severity of these crashes. 12
Wrong-way crashes tend to be more severe and have a greater likelihood to result in 13
death or injury when compared with other types of crashes. Past studies (4, 5, and 6) showed that 14
although a very small percentage of overall traffic crashes were caused by wrong-way driving, a 15
relatively large percentage of fatal crashes were. Drivers and passengers in both wrong-way and 16
right-way vehicles can be killed in wrong-way crashes (7). For example, of the 49 fatal wrong-17
way crashes on the New Mexico interstate highway system between 1990 and 2004, 35 drivers 18
and 11 passengers in the wrong-way vehicles were killed; 18 drivers and 15 passengers in 19
vehicles traveling in the correct direction were killed as well (7). 20
Wrong-way crashes are more prevalent during non-daylight hours, particularly in the 21
early morning. In Texas, the six hours from 12:00 midnight to 6:00 a.m. were when 52 percent of 22
all wrong-way crashes occurred; however, only 10.4 percent of overall freeway crashes occurred 23
during that time period. Past studies (4, 5, 8, and 9) indicated that wrong-way crashes occurred 24
more frequently during the weekends. Monthly distribution of wrong-way crashes varies among 25
different states (8, and 10) and countries (11), showing no consistent trend. 26
Research operations conducted in both California (4) and Texas (5, and 6) have found 27
that urban areas have many more wrong-way crashes than rural areas. Studies in Texas (5, and 6) 28
also found that most of the wrong-way collisions occurred in the inside lane of the correct 29
direction and at locations with left-side exit ramps or one-way streets that transitioned into a 30
freeway section. A study in the Netherlands from 1983 to 1998 found that 79 percent of wrong-31
way crashes took place on the main line of freeway, 5 percent on merge/diverge lanes, and 17 32
percent on ramps (12). 33
The characteristics of wrong-way drivers, such as driver sobriety, age, and gender have 34
been discussed in many past studies. A significant portion of wrong-way crashes on freeways 35
was caused by driving under the influence (DUI) of alcohol or drugs. Most past studies 36
concluded that young drivers and older drivers are overrepresented in the wrong-way crashes. 37
Most of the crashes caused by drivers in the young and middle age range were brought about by 38
inattention, while most crashes caused by drivers in the senior age range occurred because of 39
some physical illnesses such as dementia or confusion (11). The overwhelming majority of 40
wrong-way crashes involved male drivers, and most of the female drivers were in the young age 41
groups (11). 42
Past studies indicated that wrong-way driving crashes were random events, and it was 43
very difficult to identify high crash locations for improvements. In this study, a new method is 44
developed to identify high crash locations for field review and countermeasure developments. 45
TRB 2014 Annual Meeting Paper revised from original submittal.
Zhou, Zhao, Fries, and Pour Rouholamin
4
This method has been implemented by the IDOT districts and proved to be an effective way to 1
develop site specific countermeasures to mitigate wrong-way driving errors and crashes. 2
3
METHODS 4
Data Collection 5
The IDOT original crash data in text format was studied and used in the wrong-way crash data 6
collection. The IDOT original crash data (text file) contains three different sub-databases: crash 7
file, vehicle file and person file. Each file contains various information relevant to this study. For 8
instance, the crash file includes all the information on the crash characteristics such as crash 9
location, time of day, severity, etc. 10
Altogether, 632 possible wrong-way crashes on freeways were identified from the total 11
2,387,877 crashes from 2004 to 2009 in Illinois. After reviewing the hardcopy crash reports, 217 12
wrong-way crashes were confirmed. Table 1 lists the number of crashes per year. It should be 13
noted that the total crashes significantly decreased in 2009 as a result of the Illinois crash 14
reporting threshold changing from $500 to $1,500 in that year. 15
16
TABLE 1 Number of Crashes under Different Categories 17
Category Year
2004 2005 2006 2007 2008 2009 Total
Total crashes 433,259 421,757 408,858 423,090 408,487 292,426 2,387,877
Freeway crashes 31,908 30,156 24,772 29,200 30,289 21,960 168,285
Possible wrong-way crashes 125 137 103 106 88 73 632
Confirmed wrong-way crashes 40 32 31 39 37 38 217
Percentage of freeway crashes 0.125 0.133 0.122 0.173 0.129 0.125 0.133
Percentage of total crashes 0.009 0.008 0.008 0.009 0.009 0.012 0.009
18
Data Analysis 19
The purpose of the data analysis was to investigate the characteristics of wrong-way driving 20
crashes. In this study, wrong-way crash data were analyzed from these three different 21
perspectives based on the three sub-databases: crash, person, and vehicle. For each of these sub-22
databases, different characteristics were separately considered and divided into discrete 23
categories for further analysis and to determine the role of that particular factor in the related 24
sub-database. These sub-databases are complementary and the information for the same crash 25
can be linked together from the three different sub-databases based on the crash identification 26
number. 27
28
RESULTS 29
Crash Characteristics 30
In the crash sub-database, each crash record consisted of 70 variables in either text or numerical 31
format. From this database, the researchers determined the temporal distribution, geographic 32
distribution, roadway characteristics, and some other crash characteristics such as collision type 33
and crash severity of wrong-way crashes. 34
35
Temporal Distribution 36
The temporal distributions of wrong-way crashes include crash year, month, day, and hour. The 37
weather and light conditions were considered part of the temporal distribution since both are 38
highly correlated with the crash time (month and time of day). The annual wrong-way crash 39
frequency from 2004 through 2009 varied between 31 and 40, with an average frequency of 36. 40
TRB 2014 Annual Meeting Paper revised from original submittal.
Zhou, Zhao, Fries, and Pour Rouholamin
5
The monthly distribution variation, from 9 to 25, was greater than the annual distribution. The 1
crash database also indicated that about 28.6% of wrong-way crashes occurred on weekends 2
between 12 midnight and 5 a.m. Figures 1 and 2 illustrate the temporal distribution of wrong-3
way crashes. Approximately 80% of wrong-way crashes occurred when the road surface was dry, 4
under clear weather conditions, and during nighttime hours. 5
6
7
FIGURE 1 Distribution of Wrong-way Crashes on Different Days of the Week 8
9
10
FIGURE 2 Hourly Distribution of Wrong-way Crashes on Illinois Freeways 11
12
Geographical Distribution 13
The geographical distribution characteristics of wrong-way crashes were extracted from the 14
crash database, including county, city, and township. Although wrong-way crashes were reported 15
in 43 of 102 counties in Illinois, about 64% of wrong-way crashes were located in the following 16
four counties: Cook County (37.8%), St. Clair County (9.7%), Madison County (9.2%), and Will 17
County (6.5%). It should be noted that Cook County and Will County are in the Chicago 18
23
19
28 27 27
53
40
0
10
20
30
40
50
60
Mon Tue Wed Thu Fri Sat Sun
Number of WWD Crashes
Day of the Week
18 20 17
30
26
99
3222
646
222
75
10 10
6
10 9
0
5
10
15
20
25
30
35
Number of WWD Crashes
Hour of the Day
TRB 2014 Annual Meeting Paper revised from original submittal.
Zhou, Zhao, Fries, and Pour Rouholamin
6
metropolitan area, while St. Clair County and Madison County are in the St. Louis metropolitan 1
area. In 35 other counties, the reported wrong-way crashes averaged less than four per county 2
over the six-year period. Approximately 25.8% of wrong-way crashes were reported in the city 3
of Chicago. 4
5
Roadway Characteristics 6
The crash database was then examined to determine the roadway characteristics of wrong-way 7
crashes, specifically the route number, route type, traffic control device, road surface condition, 8
road defects, intersection, and work zone. 9
Wrong-way crashes occurred on 30 different routes. Approximately 60% of them 10
happened on five routes: I-55 (19.4%), I-94 (14.7%), I-57 (9.7%), I-74 (7.8%), and I-64 (7.4%). 11
Ninety-five percent of wrong-way crashes occurred on roadways where no traffic control device 12
malfunctions were reported. Furthermore, construction/maintenance zone was noted in 13
approximately 7% of wrong-way crashes. Less than 3% of wrong-way crashes were related to a 14
work zone, and 5% of wrong-way crashes were reported to be related to a specific intersection. 15
Limited information on possible road design/traffic control problems was included in the crash 16
reports. 17
18
Crash Characteristics 19
The researchers reviewed crash characteristics, including the number of vehicles involved, 20
collision type, and crash severity. Table 2 summarizes the number of crashes by type and 21
severity in Illinois from 2004 to 2009. As can be seen, approximately 67% of wrong-way crashes 22
were found to involve multiple vehicles and resulted in head-on (45%) or sideswipe opposite 23
direction crashes (22%). The collision types for single-vehicle wrong-way crashes (14%) were 24
mainly fixed object ones. Altogether, the crash characteristics of multi- and single-vehicle 25
crashes suggested that wrong-way crashes are more severe than other crash types. Most wrong-26
way crashes involved two or three vehicles, including the at-fault wrong-way vehicle, which 27
collided with other vehicle(s) traveling in the correct direction(s). Findings indicated that the 28
crash severity levels were directly related to collision types. Ninety-seven percent of fatal crashes 29
were head-on crashes or opposite direction sideswipe crashes. The collision types for A-injury 30
(incapacitating injuries) crashes were also mainly head-on (71%), opposite direction sideswipe 31
(11%) and fixed object (9%). Almost 60% of head-on crashes caused fatalities or incapacitating 32
crashes, while only 17% of opposite direction sideswipe crashes resulted in one or more fatalities 33
and/or A-injuries. A significant proportion of wrong-way vehicles fled the crash sites after the 34
incident (17.5%). 35
The collision types were also related to the number of vehicles involved in wrong-way 36
crashes. For example, rear-end crashes caused by wrong-way drivers were usually between two 37
or more vehicles traveling in the correct directions that collided while avoiding wrong-way 38
vehicles. Most of the rear-end crashes were not severe. In addition, collisions for single-vehicle 39
wrong-way crashes were mainly fixed object. The collisions involving multiple-vehicles were 40
frequently either head-on crashes or opposite direction sideswipe crashes. There were no fatal 41
crashes for single-vehicle wrong-way crashes; however, many single-vehicle wrong-way crashes 42
resulted in A- or B-injuries (non-incapacitating injuries). 43
44
TRB 2014 Annual Meeting Paper revised from original submittal.
Zhou, Zhao, Fries, and Pour Rouholamin
7
Driver Characteristics 1
To analyze the characteristics of wrong-way drivers, the person database for the 203 crash 2
records was used. There were 32 variables in the person file. Eleven of them were duplicates in 3
different formats (code or text). A total of 660 persons were involved in the 217 wrong-way 4
crashes, among which 6.7% were killed, 38% were injured, and 56% incurred no injuries. For 5
those injured, 43% and 46% were A-injuries and B-injuries, respectively. Only about 10% of 6
injuries were classified as possible injuries (C-injury). Compared with other crash types, wrong-7
way crashes were more severe in which large proportions resulted in severe injuries and/or 8
fatalities. 9
10
TABLE 2 Crash Severity and Collision Type for Wrong-way Crashes (2004-2009) 11
Collision Type Crash Severity Level Total (#/%)
Fatal A-Injury B-Injury C-Injury No Injuries
Head-on 27 32 14 6 20 99/45.6
Sideswipe (opposite direction) 3 5 8 2 29 47/21.6
Fixed Object 0 4 7 1 19 31/14.3
Rear End 0 0 1 1 6 8/3.7
Sideswipe (same direction) 0 2 1 0 5 8/3.7
Overturned 0 2 2 0 2 6/2.8
Other Non-Collision 0 0 2 1 2 5/2.3
Other Object 0 0 1 0 4 5/2.3
Turning 0 0 1 0 3 4/1.8
Angle 1 0 0 0 2 3/1.4
Parked Motor Vehicle 0 0 0 0 1 1/0.5
Total 31 45 37 11 93
217/100
Percentage (%) 14.3 20.7 17 5.1 42.9
12
Driver Demographic Information 13
Driver demographic information refers to date of birth, age, sex, and state of residence (as 14
indicated on one’s driver’s license). The drivers were classified into different age groups based 15
on the NHTSA criteria, which classify drivers under age 25 as young drivers and those 65 years 16
and above as older drivers. The study results showed that younger drivers and older drivers were 17
proportionally overrepresented in all crash types. With respect to the sex of wrong-way drivers, 18
the database revealed that males represented nearly 68%, particularly those in the age groups of 19
21–24, 25–34, and greater than 65. Female wrong-way drivers were most prevalent in the 35–44 20
age groups. Demographic information indicated that most (77%) wrong-way drivers were 21
licensed in the State of Illinois. The state with the next largest frequency (6%) was Missouri. 22
Sixteen percent of wrong-way crashes included no residence information for the drivers. 23
24
Driver Physical Condition 25
The driver’s physical condition, including the apparent observed condition of the driver, driver’s 26
Blood Alcohol Concentration (BAC) test result, and driver’s vision were analyzed to investigate 27
the possible impact of DUI on wrong-way crashes. The illegal BAC limits in Illinois are 0% for 28
school bus drivers and drivers under the age of 21, 0.04% for commercial driver’s license 29
holders, and 0.08% for drivers aged 21 and over. A large proportion of wrong-way drivers were 30
TRB 2014 Annual Meeting Paper revised from original submittal.
Zhou, Zhao, Fries, and Pour Rouholamin
8
found to be DUI: 50% by alcohol and nearly 5% by other drugs. However, the actual percentage 1
is higher because many drivers refused to take the test or were tested with no results. Eighty 2
percent of the drivers completing a BAC test had a level greater than 0.1%. Most of the DUI 3
drivers were in the age range of 21–54, and almost no senior wrong-way drivers were driving 4
under the influence. Driver vision was not reported as a possible reason for wrong-way crashes: 5
nearly 99% of reports note that driver’s vision was not obscured or the information was unknown. 6
Only about 17% of wrong-way drivers were in normal physical condition. 7
8
Driver Injury Severity 9
An analysis was conducted to identify the factors related to wrong-way driver injury severity. 10
Figure 3 shows the percentage of wrong-way crashes at different severity levels. Seat belt use, 11
airbag deployment, driver age, sex, condition, and ejection or extrication were analyzed. More 12
than 70% of wrong-way drivers were using their seat belts when wrong-way crashes occurred. 13
Less than 7% of wrong-way drivers who used a seat belt were killed in wrong-way crashes; 14
however, the wrong-way driving fatality rate was raised to more than 30% when seat belts were 15
not used. Even though wrong-way crashes were more severe than most other crash types, more 16
than 50% of wrong-way drivers who used a seat belt were not injured in the crashes. Airbag 17
deployment for wrong-way vehicles was investigated as well. However, for more than 50% of 18
wrong-way vehicles, the airbag deployment was unknown. Less than 10% of wrong-way 19
vehicles’ airbags were deployed from either the front, side, or combined. Additionally, the 20
person database revealed that nearly 10% of wrong-way drivers were ejected or 21
trapped/extricated as a result of the collision. 22
23
24
FIGURE 3 Injury Severity Level for Wrong-way Drivers 25
26
Table 3 illustrates the apparent relationship between driver severity level and driver 27
condition. Approximately 80% of wrong-way drivers killed in the crashes were impaired by 28
alcohol or drugs. The DUI percentage was also relatively high among A-injured wrong-way 29
drivers. The BAC for 65% of wrong-way drivers killed in the crashes was higher than 0.1%, and 30
25% of them were completely ejected or trapped/extricated in the crashes. On the contrary, less 31
than 10% of wrong-way drivers under normal physical conditions were killed in wrong-way 32
crashes. The comparison between age, sex, and injury severity signified several trends. First, 33
wrong-way drivers who were killed or incapacitated in wrong-way crashes were mainly in the 34
Fatality,
20
A-Injury,
41
B-Injury,
33
C-Injury,
7
No
Injuries,
102
TRB 2014 Annual Meeting Paper revised from original submittal.
Zhou, Zhao, Fries, and Pour Rouholamin
9
age group 21–45. Next, although the total number of fatalities and A-injuries contributed by 1
older drivers was not as high as those of the 21–45 age group, the percentage of older wrong-way 2
drivers who ended up as fatalities was much higher than the other age groups. 3
4
TABLE 3 Driver Condition and Driver Severity Level 5
Driver Condition Driver Severity Level Total
Fatality A-Injury B-Injury C-Injury No Injuries
Alcohol Impaired 11 24 16 3 47 101
Other/Unknown 1 6 7 1 26 41
Normal 3 5 6 1 19 34
Drug Impaired 5 0 1 1 2 9
Illness 0 4 1 0 3 8
Had Been Drinking 0 2 1 1 3 7
Asleep/Fainted 0 0 1 0 1 2
Fatigued 0 0 0 0 1 1
Total 20 41 33 7 102 203
Percentage 9.9 20.2 16.3 3.4 50.2 ─
6
Vehicle Characteristics 7
Vehicle characteristics including vehicle type, use, defects, commercial vehicle indicator, and 8
number of occupants were analyzed. Results, which are summarized in Table 4, indicate that no 9
wrong-way vehicles were commercial vehicles. Nearly 70% of wrong-way vehicles were 10
passenger cars, and less than 2% of wrong-way crashes involved tractors. Approximately 90% of 11
wrong-way vehicles were used for personal purposes, and about 85.2% of wrong-way vehicles 12
had a single occupant. No vehicle defects were reported for more than 98% of wrong-way 13
crashes. 14
15
TABLE 4 Vehicle Type for At-fault Drivers 16
Vehicle Type Crash Frequenc
y
Percent (%)
Passenger 139 68.5%
Pickup 26 12.8%
SUV 18 8.9%
Van/Mini-Van 12 5.9%
Unknown 4 2.0%
Tractor with Semi-Trailer 2 1.0%
Motorcycle (over 150cc) 1 0.5%
Tractor without Semi-Trailer 1 0.5%
Total 203 100%
17
Vehicle Operation 18
More than 80% of wrong-way vehicles collided with other vehicles and/or fixed objects with the 19
front or the left and right front quarter panels, resulting in head-on collisions. Although 20
information on the event, including location, was not recorded in the database for more than half 21
TRB 2014 Annual Meeting Paper revised from original submittal.
Zhou, Zhao, Fries, and Pour Rouholamin
10
of wrong-way cases, the researchers reviewed the common events and locations for those 1
available. The majority of known events for wrong-way crashes were either “motor vehicle in 2
transport” or “ran off roadway”. The most frequently reported locations for the crash events were 3
on roadway pavement. 4
5
Collision Results 6
The vehicle database included information about the collision result. Researchers found that 7
more than 80% of wrong-way vehicles that crashed were towed away, and approximately 5% of 8
wrong-way vehicles caught fire after the crash. The findings also indicated that no wrong-way 9
vehicles spilled hazardous material during or after the crash. The number of vehicles involved in 10
wrong-way crashes ranged from one to six (Table 5). Sixty-three percent of wrong-way crashes 11
involved two vehicles compared with 18 percent for one vehicle or 13 percent for three vehicles. 12
14.3 percent of wrong-way crashes were also found to be fatal crashes. 13
14
TABLE 5 Number of Vehicles Involved and Crash Severity 15
Crash Severity Number of Vehicles Total
1 2 3 4 5 6
Fatal Crash 0 20 10 1 0 0 31
A Injury Crash 6 24 10 4 0 1 45
B Injury Crash 9 22 3 2 1 0 37
C Injury Crash 2 8 0 1 0 0 11
No Injuries 23 63 6 1 0 0 93
Total 40 137 29 9 1 1 217
Percentage (%) 18.4 63.1 13.4 4.1 0.5 0.5 ─
16
Wrong-way Characteristics 17
Based on the reported wrong-way entry points for the crashes studied, researchers can compare 18
predicted entry points and length of wrong-way travel. The correlation of these parameters is 19
then used to propose a calibrated method for predicting wrong-way entry points for wrong-way 20
crashes where little is known. 21
22
Wrong-way Entry Points 23
Identifying wrong-way entry points can help develop proper countermeasures to combat wrong-24
way driving at a specific interchange area. However, information on wrong-way entry points was 25
usually unavailable from the crash database. To obtain wrong-way entry point’s information, the 26
narrative description of the crash report hard copies were reviewed case by case, and the crash 27
locations were examined using aerial photographs. Some vehicles began driving the wrong way 28
after they crossed the median, made a U-turn on the freeway, or tried to leave the freeway from 29
an entrance ramp and these crashes were excluded from the analysis. For nearly 20% of wrong-30
way crashes, wrong-way entry points were recorded in the crash report hard copies. For crashes 31
without recorded wrong-way entry points, the first and second possible wrong-way entry points 32
at interchanges were evaluated. These possible entry points would be the nearest first and second 33
off-ramps on the freeway if the vehicle drives in the correct direction. When selecting the wrong-34
way collision site as the starting point, researchers determined the first and second wrong-way 35
entry points by searching the first and second exit from the freeway when driving from the 36
TRB 2014 Annual Meeting Paper revised from original submittal.
Zhou, Zhao, Fries, and Pour Rouholamin
11
collision site in the correct direction (downstream). As shown in Table 6, the total number of 1
recorded entry points was 47, and the numbers of first and second possible entry points were 147 2
and 146, respectively. Concerning the recorded, first, second, and total wrong-way entry points, 3
compressed diamond (26%, 30%, 30%, 29%) and diamond interchanges (34%, 27%, 26%, 27%) 4
were the top two interchange types. However, when considering the exposure (total number of 5
each type of interchange in IL), wrong-way crash rate (number of wrong-way entries per 100 6
interchanges per year) can be calculated for all the interchange types (Table 6). Accordingly, 7
compressed diamond, SPUI, partial cloverleaf, and freeway feeder are the top four interchange 8
types for the potential wrong-way entries. Note that due to the frequency of diamond 9
interchanges in IL, the number of wrong-way entries is not overrepresented. 10
11
TABLE 6 Interchange Types for Wrong-way Crash Entry Points 12
Interchange Type Recorded 1st Estimated
Entry Point
2nd Estimated
Entry Point
Total No. of
Interchanges in IL
WW Crash
Rate Rank
# % # % # % # % % per year
Compressed Diamond 12 25.53 44 29.93 44 30.14 56 7.64 13.39 1
Diamond 16 34.04 39 26.53 38 26.03 308 42.02 2.44 6
Partial Cloverleaf 5 10.64 28 19.05 23 15.75 79 10.78 5.22 3
Cloverleaf 3 6.38 12 8.16 12 8.22 59 8.05 3.39 5
Rest Area 1 2.13 9 6.12 6 4.11 64 8.73 1.82 6
Freeway Feeder 5 10.64 3 2.04 6 4.11 30 4.09 4.44 4
Modified Diamond 3 6.38 4 2.72 4 2.74 61 8.32 1.64 6
Semi-Directional 0 0.00 3 2.04 4 2.74 19 2.59 2.19 6
SPUI 1 2.13 2 1.36 3 2.05 8 1.09 5.73 2
Trumpet 0 0.00 2 1.36 4 2.74 25 3.41 1.33 7
Directional 1 2.13 1 0.68 2 1.37 24 3.27 1.39 7
Total 47 100.00 147 100.00 146 100.00 733 100.00 3.57 ─
13
Wrong-way Driving Distance 14
A comparison of driving distance between recorded and predicted entry points was conducted to 15
see whether the characteristics of entry points recorded in crash reports had similar 16
characteristics to those predicted by the researchers. The cumulative wrong-way driving distance 17
distribution for recorded, first, and second entry points was plotted (Figure 4). Results indicated 18
that wrong-way driving distances for recorded and estimated first entry points were very close. 19
For example, the average driving distance for wrong-way crashes was 1.2 miles when an entry 20
point was known/recorded and 1.54 miles when it was estimated as the first upstream 21
interchange. On the contrary, the average distance for the second estimated entry point was 3.6 22
miles. The maximum wrong-way driving distances for recorded, estimated first and second 23
wrong-way entry points were 6.4 miles, 13 miles, and 17.6 miles, respectively. 24
The differences between the estimated and recorded entry points led the researchers to 25
conduct a statistical analysis. The findings demonstrated that the recorded and 1
st
estimated entry 26
point were not different at the 95% significance level. On the contrary, the findings suggested 27
that the 2
nd
estimated entry point was significantly different from both other categories at the 95% 28
level, as illustrated in Figure 5. Additionally, the researchers conducted a correlation analysis of 29
TRB 2014 Annual Meeting Paper revised from original submittal.
Zhou, Zhao, Fries, and Pour Rouholamin
12
the data from Figure 4 and found that the recorded entry points had a 0.978 correlation with the 1
1
st
estimated entry points and a 0.886 correlation with 2
nd
estimated entry points. These findings 2
suggest that when there is no information available about the entry point of a wrong-way driver, 3
the 1
st
estimated entry point is statistically no different than the true entry point in length. When 4
conducting safety audits for crashes with unknown entry points, these findings suggest there is 5
value in the evaluation of both the 1
st
and 2
nd
estimated entry points. 6
7
8
FIGURE 4 Cumulative Distributions for Wrong-way Driving Distance 9
10
11
FIGURE 5 95% Confidence Intervals of Distance (mi) Between Entry Point and Crash 12
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Cumulative Percentage
Wrong-way Driving Distance (mi)
Recorded
1st Estimated Entry Point
2nd Estimated Entry Point
0.59
1.13
1.67
0.95
1.54
2.12
2.57
3.57
4.57
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
4.50
5.00
Miles from Entry Point to Crash
Entry Point
Recorded 1st Estimated 2nd Estimated
TRB 2014 Annual Meeting Paper revised from original submittal.
Zhou, Zhao, Fries, and Pour Rouholamin
13
The recorded, 1
st
, and 2
nd
estimated wrong-way entry points were used to rank the top ten 1
locations based on the summation of weighted recorded, 1
st
, and 2
nd
estimated wrong-way entry 2
frequencies. These weighting factors were identified through discussions with safety engineers at 3
the state Department of Transportation. A weight of 1.0 was assigned to the recorded entry point 4
to represent this point is definitely an entry point. A weight of 0.5 was assigned to the 1
st
5
estimated entry point representing 50% chance of being an entry point, and 0.25 was assigned to 6
the 2
nd
estimated entry point indicating 25% chance of being an entry point. 7
This method was applied to a case study in Illinois. Altogether there were 265 wrong-8
way entry points identified for all crashes, of which 12 percent were recorded, 35 percent 9
accounted for the 1
st
estimated, and 37 percent for the 2
nd
estimated. Wrong-way entry was a 10
sparse event, and the entry frequencies for most points were relatively low, varying from one to 11
five. Among the 265 entry points, approximately 76 percent experienced one wrong-way entry 12
and 34 percent experienced two to five entries. Based on weighted wrong-way entry points, the 13
top ten locations were identified for field review. These locations included four compressed 14
diamond, three partial cloverleaf interchanges, one diamond, one directional, and one Single-15
Point Urban Interchange (SPUI). Comparing this list of interchanges with those in Table 6, 16
indicates that including first and second likely entry points provides needed sensitivity to the 17
selection process. For example, other studies implicated partial cloverleaf interchanges in wrong-18
way driving entries, but without the weighting process there would not have been selected in this 19
case study. 20
21
CONCLUSION 22
Six-years of crash data from IDOT were collected for identifying wrong-way crashes. Out of 632 23
possible wrong-way crashes identified from the crash database, 217 wrong-way crashes were 24
verified by reviewing the hardcopies of those crash reports. General statistical characteristics of 25
wrong-way crashes on Illinois freeways were presented in this paper. 26
The study findings indicated that approximately 28.6% of wrong-way crashes occurred 27
on weekends from midnight to 5 a.m. About 64 percent of wrong-way crashes were located in 28
four counties adjacent to the urban areas of Chicago and East St. Louis, Illinois. Most (67%) of 29
the multiple-vehicle wrong-way crashes were head-on and sideswipe opposite direction crashes. 30
Road surface condition did not significantly contribute more to wrong-way crashes than other 31
freeway crashes; however, a large proportion of wrong-way drivers were DUI, among which 32
over 50 percent were confirmed to be impaired by alcohol. This data confirms that wrong-way 33
crashes have similar characteristics as those studied in other states and during previous decades. 34
The key contribution of this study was a new method developed to identify locations for 35
field review, based on wrong-way crash locations. The authors propose a straightforward method 36
for identifying sites for field review to identify possible causes. In this method, transportation 37
practitioners can use weighted wrong-way entry points to rank the entry locations for further 38
field inspection. Because most wrong-way driving crashes (78%) have unknown entry points, 39
this method and the findings in this paper could provide insight into their origin and help 40
accurately identify common wrong-way entry locations. 41
Future work should focus towards identifying the effectiveness of wrong-way prevention 42
countermeasures. In particular, this did not aim at finding empirical evidence to support the 43
effectiveness of countermeasures used by DOTs. 44
TRB 2014 Annual Meeting Paper revised from original submittal.
Zhou, Zhao, Fries, and Pour Rouholamin
14
ACKNOWLEDGEMENT 1
This study was sponsored and supported by the Illinois Department of Transportation (IDOT) 2
and Illinois Center for Transportation (ICT). Authors would also like to extend their sincere 3
thanks to the Technical Review Panel (TRP) members for all their feedback and inputs. 4
5
REFERENCES 6
1. Scaramuzza, G., and M. Cavegn. Wrong-way drivers: extent-interventions. The European 7
Transport Conference, Leeuwenhorst Conference Centre, The Netherlands, 2007. 8
2. National Highway Traffic Safety Administration, Fatality Analysis Reporting System (FARS) 9
Encyclopedia, Washington, D.C., http://www-fars.nhtsa.dot.gov/main/index.aspx. Accessed 10
May 5, 2013. 11
3. Zhou, H., J. Zhao, R. Fries, M. R. Gahrooei, L. Wang, B. Vaughn, K. Bahaaldin, and B. 12
Ayyalasomayajula. Investigation of Contributing Factors Regarding Wrong-way Driving on 13
Freeways. Publication FHWA-ICT-12-010. Illinois Center for Transpiration, 2012. 14
4. Copelan, J. E. Prevention of Wrong-way Accidents on Freeways. Publication FHWA/CA-TE-15
89-2. California Department of Transportation, 1989. 16
5. Cooner S. A., A. S. Cothron, and S. E. Ranft. Countermeasures for Wrong-way Movement on 17
Freeway: Guidelines and Recommendation Practices. Publication FHWA/TX-04/4128-2. 18
Texas Transportation Institute, 2004. 19
6. Cooner S. A., A. S. Cothron, and S. E. Ranft. Countermeasures for Wrong-way Movement on 20
Freeway: Overview of Project Activities and Findings. Publication FHWA/TX-04/4128-1. 21
Texas Transportation Institute , 2004. 22
7. Lathrop, S. L, T. B. Dick, and K. B. Nolte. Fatal Wrong-Way Collisions on New Mexico’s 23
Interstate Highways, 1990-2004. Journal of Forensic Sciences, Vol. 55, 2010, pp. 432-437. 24
8. Braam, A. C. Wrong-way Crashes: Statewide Study of Wrong-way Crashes on Freeways in 25
North Carolina. North Carolina Department of Transportation, July 2006. 26
https://connect.ncdot.gov/resources/safety/Documents/Crash%20Data%20and%20Informatio27
n/WrongWayCrash.pdf. Accessed May 12, 2013. 28
9. North Texas Tollway Authority (NTTA). Keeping NTTA roadways safe: Wrong-way Driver 29
Task Force Staff Analysis. North Texas Tollway Authority, September 2009. 30
https://www.ntta.org/newsresources/safeinfo/wrongway/Documents/WWDAnalysisAUG20131
1.pdf. Accessed May 7, 2013. 32
10. Cooner, S. A., and S. E. Ranft. Wrong-way Driving on Freeways: Problems, Issues and 33
Countermeasures. 2008 Annual Meeting of Transportation Research Board, Washington, 34
D.C., 2008. 35
11. Institute of Traffic Accident Research and Data Analysis (ITARDA). Highway Accidents 36
Involving Dangerous Wrong-way Traveling. 2002. 37
http://www.itarda.or.jp/itardainfomation/english/info36/36top.html, Accessed April 23, 2013. 38
12. SWOV Fact sheet: Wrong-way Driving. Leidschendam, the Netherlands, August 2009. 39
http://www.swov.nl/rapport/Factsheets/UK/FS_Wrong_way_driving.pdf, Accessed May 3, 40
2013. 41
TRB 2014 Annual Meeting Paper revised from original submittal.
