Conference PaperPDF Available

OVERVIEW OF SAFETY COUNTERMEASURES FOR ROADWAY DEPARTURE CRASHES

Authors:

Abstract and Figures

A roadway departure (RwD) crash happens when a vehicle leaves the traveled way (crossing an edgeline or a centerline). These crashes, comprising run-off-road (ROR) and cross-median head-on collisions, tend to be more severe than other crash types. In 2013, RwD crashes accounted for 56 percent of all motor vehicle traffic fatalities. Inattention or fatigue, an avoidance maneuver, or traveling too fast are the common reasons a driver may leave the travel lane. Roadway and roadside geometric design features (e.g., lane and shoulder widths, horizontal curvatures, sideslope, and clear zones) play a significant role in whether or not human error results in a crash. To achieve the Federal Highway Administration’s (FHWA’s) Toward Zero Deaths vision, many safety countermeasures have recently been implemented by state departments of transportation and local agencies to mitigate RwD crashes. This paper presents a summary of various case examples to provide transportation practitioners with a good understanding of the effectiveness of RwD safety countermeasures.
Content may be subject to copyright.
OVERVIEW OF SAFETY COUNTERMEASURES FOR ROADWAY DEPARTURE
CRASHES
Mohammad Jalayer, Corresponding Author
Ph.D. Candidate, Dept. of Civil Engineering
Auburn University, Auburn, AL 36849-5337
Tel: 312-351-4730; Email: jalayer@auburn.edu
Huaguo Zhou
Associate Professor, Dept. of Civil Engineering
Auburn University, Auburn, AL 36849-5337
Tel: 334-844-1239; Email: zhouhugo@auburn.edu
Cathy Satterfield
Federal Highway Administration (FHWA), Office of Safety
4749 Lincoln Mall Drive, Suite 600, Matteson, IL 60443
Tel: 708-283-3552; Email: cathy.satterfield@dot.gov
Word count = 3,972 words text + 4 tables/figures x 250 words (each) = 4,972 words
Submission Date: July 31, 2015
A Paper Submitted for Presentation at the 95th Transportation Research Board Annual Meeting
Jalayer, Zhou, Satterfield  2
ABSTRACT 1
A roadway departure (RwD) crash happens when a vehicle leaves the traveled way (crossing an 2
edgeline or a centerline). These crashes, comprising run-off-road (ROR) and cross-median head-3
on collisions, tend to be more severe than other crash types. In 2013, RwD crashes accounted for 4
56 percent of all motor vehicle traffic fatalities. Inattention or fatigue, an avoidance maneuver, or 5
traveling too fast are the common reasons a driver may leave the travel lane. Roadway and roadside 6
geometric design features (e.g., lane and shoulder widths, horizontal curvatures, sideslope, and 7
clear zones) play a significant role in whether or not human error results in a crash. To achieve the 8
Federal Highway Administration’s (FHWA’s) Toward Zero Deaths vision, many safety 9
countermeasures have recently been implemented by state departments of transportation and local 10
agencies to mitigate RwD crashes. This paper presents a summary of various case examples to 11
provide transportation practitioners with a good understanding of the effectiveness of RwD safety 12
countermeasures. 13
14 Keywords: Roadway Departure Crash, Safety Countermeasures, Run-off-road Crash 15
Jalayer, Zhou, Satterfield  3
INTRODUCTION
1
A roadway departure (RwD) crash is defined by the Federal Highway Administration (FHWA) as
2
“A crash in which a vehicle crosses an edge line, a centerline, or otherwise leaves the traveled
3
way.” These crashes, comprising of run-off-road (ROR) and cross-median/cross-centerline head-
4
on collisions, tend to be more severe than other crash types. The reasons for ROR events are varied
5
and include a driver attempting to avoid a vehicle, an object, or an animal in the travel lane;
6
inattentive driving due to distraction, fatigue, sleep, or drugs; the effect of weather on pavement
7
conditions; and traveling too fast through a curve or down a grade. There are also a number of
8
roadway design factors that can increase the probability that a driver error will result in an ROR
9
crash (e.g., travel lanes that are too narrow, substandard curves, and unforgiving roadsides) (1).
10
Most head-on crashes are similar to ROR crashes—in both cases, the vehicle strays from its travel
11
lane (2).
12
RwD crashes constitute one of the most severe types of crashes. According to the FHWA,
13
in 2013, 56 percent of all motor vehicle fatalities involved a RwD. Figure 1 depicts the percentage
14
of total RwD fatal crashes across the United States, categorized by the first event in the crash.
15
According to a query of seven years of crash data (20072013) from the Fatality Analysis
16
Reporting System (FARS) database, an average of 57 percent of motor vehicle traffic fatalities
17
occurred each year due to RwD in the United States (3). The distribution of this number differs
18
between states (see Figure 2). In addition, the majority of RwD crashes occurred during the
19
nighttime and inclement weather conditions (e.g., fog, snow).
20
21
22
23
FIGURE 1 Percentage of fatal motor crashes in the United States in 2013 (4).
24
Jalayer, Zhou, Satterfield  4
1
FIGURE 2 Average percentage of RwD fatalities in each state (2007-2013) (4).
2
State RwD State RwD State RwD
Vermont 79.0% Alabama 64.0% California 49.0%
Wyoming 78.0% Kansas 64.0% Michigan 49.0%
West Virginia 75.0% Nebraska 63.0% Illinois 49.0%
Montana 73.0% Wisconsin 63.0% Delaware 48.0%
Arkansas 72.0% Washington 63.0% Nevada 48.0%
Maine 71.0% Alaska 63.0% Arizona 46.0%
Kentucky 69.0% Oklahoma 63.0% New Jersey 46.0%
Tennessee 69.0% South Dakota 62.0% New Mexico 44.0%
Idaho 69.0% Pennsylvania 62.0% Florida 43.0%
New Hamps hire 69.0% South Dakota 62.0% Dist. of Columbia 29.0%
Virginia 68.0% Rhode Island 62.0%
North Dakota 66.0% South Carolina 61.0%
Missouri 65.0% North Carolina 61.0%
Oregon 65.0% Louisiana 61.0%
Ohio 61.0%
Iowa 60.0%
Colorado 60.0%
Connecticut 58.0%
Indiana 57.0%
Massachusetts 57.0%
Georgia 56.0%
Hawaii 55.0%
Texas 54.0%
Mississippi 54.0%
Utah 52.0%
Minnesota 52.0%
Maryland 50.0%
Average Annual Roadway Departure Fatalities (2007-2013)
Group 1 (65% and Higher) Group 2 (Between 50-65%) Group 3 (Below 50%)
Jalayer, Zhou, Satterfield  5
Several strategies to reduce the number of ROR crashes have been identified by the
1
American Association of State Highway and Transportation Officials (AASHTO) including (5):
2
3
Pavement edgeline installation
4
Centerline and shoulder rumble strip installation
5
Pavement marking enhancement
6
Shoulder drop-offs elimination
7
Safer slopes design
8
Object removal/relocation within the clear zone
9
Object delineation using retroreflective tape
10
Barrier design improvement
11
Horizontal curve geometric improvement
12
Skid-resistant roadway surface provision
13
14
Not only are most of these strategies low-cost countermeasures but they can also be
15
implemented systematically. This paper provides a brief easy-to-read overview of cost-effective
16
improvements for preventing vehicle departures from roadways. It is a summary of a recent
17
publication by the American Traffic Safety Services Association (ATSSA), which can help
18
transportation agencies better understand the effectiveness, and prioritize the implementation of
19
each RwD safety countermeasure.
20
21
ROADWAY DEPARTURE SAFETY COUNTERMEASURES
22
Countermeasure implementation case studies for mitigating RwD crashes were developed based
23
upon a comprehensive literature review and input from state and local agencies. RwD safety
24
countermeasures were divided into three major categories: signs, pavement safety, and roadside
25
design. Figure 3 illustrates the 14 RwD safety countermeasures discussed in this paper.
26
27
a) Chevron b) Dynamic Curve Warning Sign
Jalayer, Zhou, Satterfield  6
c) Advisory Ramp Speed Sign/Chevrons d) High Friction Surface Treatment
e) 20-foot Raised Pavement Markers f) Edge Line Pavement Markings
g) Safety Edge h) Milled Centerline Rumble Strips
Jalayer, Zhou, Satterfield  7
i) Shoulder Rumble Strips j) Cable Barrier
k) Guardrail l) Wider Shoulder
m) Sign with Breakaway Supports n) Clear Zone Improvements
1
FIGURE 3 RwD crash safety countermeasures (6).
2
Table 1 lists the 14 countermeasures, the evaluation method used, the results obtained, and
3
the relevant contact agencies. As shown in the table, the percentage reduction of the total number
4
of RwD crashes varies between 23 and 91, depending on the safety countermeasure. Similarly, for
5
the total number of ROR crashes, the reduction ranges from 22.1 percent to 61.6 percent. Most
6
agencies used a simple before-and-after evaluation method. Few determined benefit-cost (B/C)
7
ratios because significant crash number reductions and the relatively low-cost countermeasures
8
often resulted in fairly high B/C ratios.
9
Jalayer, Zhou, Satterfield  8
TABLE 1. Results from the 14 Case Studies 10
11
Safety
Countermeasure
Safety Evaluation
Method Results Benefit-cost
(B/C) Ratio
State Agency
Implementation
Implementation
Time
Signs
Chevron Empirical Bayes
(EB)
Total ROR crashes: -22.1 %
Total crashes during dark
condition: -24.5 %
8.0 WSDOT 1994-2006
Dynamic Curve Warning
Systems (DCWSs) ---
2.6 mph reduction in mean speed
76 % of vehicle slowed down --- ODOT 2002
Advanced Curve Warning
and Advisory Speed Sign
Simple before-and-
after
Before: 1 fatality per year; after:
0 fatality --- KYTC 2006
Pavement
Safety
High Friction Surface
Treatments (HFSTs)
Simple before-and-
after Total RwD crashes in wet
weather: -91.0 %
Total RwD crashes in dry
weather: -78.0 %
24 KYTC 2010
Raised Pavement Markers
(RPMs)
Simple before-and-
after
Total RwD crashes: -86.0 %
Total injuries: -94.0 %
--- ALDOT 2009
Edge Line Pavement
Markings
Simple before-and-
after
Total RwD crashes: -23.0 %
Total severe RwD crashes:
-38.0 %
--- MoDOT 2009
Safety Edge
Simple before-and-
after
Total crashes: -5.7 %
Two-lane
highways with
paved shoulder:
3.8 to 43.6
Two-lane
highways with
unpaved
shoulder: 2.8 to
62.8
GDOT and
INDOT 2005
Centerline Rumble Strips
(CLRS)
Simple before-and-
after
Total crashes: -33.0 %
Total RwD crashes: -31.0 %
Total head-one crashes: -35.0 %
Total opposite-direction
sideswipe crashes: -46.0 %
--- MDOT 2008-2010
Jalayer, Zhou, Satterfield  9
Shoulder Rumble Strips
(SRS)
Simple before-and-
after
Total ROTRR crashes: -47.0 to
61.6 %
Total severe ROTRR crashes:
-15.3 to 66.6 %
50 WSDOT 2000
Roadside
Design
Cable Barrier Simple before-and-
after
Before: 19 fatal crashes;
after: 0 fatal crash --- MnDOT 2004-2008
Guardrail Simple before-and-
after
Total severity and RwD index: -
16.6 to 36.7 % --- NCDOT 1997-2010
Shoulder Widening Simple before-and-
after
Total severity and RwD index: -
43.7 to 69.2 % --- NCDOT 2002-2011
Breakaway Supports for
Signs and Lighting --- --- --- --- ---
Clear Zone Improvements Simple before-and-
after Total crashes: -38.0 % --- Iowa DOT 2006
12
Jalayer, Zhou, Satterfield  10
Signs 1
According to the FARS database, more than 83 percent of the total RwD fatal motor vehicle 2
crashes occur at horizontal curves (3). Enhancing curve delineation with signs is typically 3
considered to be a low-cost safety improvement. These signs alert drivers to changes in road 4
alignment and provide information on the actions to be taken. For example, a sign may encourage 5
drivers to reduce their speeds. When placed and maintained appropriately, curve signage may 6
reduce the frequency and severity of RwD crashes. 7
8 Chevrons 9
According to the Manual on Uniform Traffic Control Devices (MUTCD), chevrons and/or one-10
direction large (Figure 3a) arrows shall be used where the difference between speed limit and the 11
advisory speed is 15 mph or more. It is important to ensure that these signs are placed and aimed 12
properly (7). The Washington State Department of Transportation (WSDOT) conducted a safety 13
evaluation analysis of chevron signs for 139 treated curves on rural two-lane roads. Empirical 14
Bayes (EB) analysis results demonstrated that chevrons along horizontal curves decreased the total 15
number of lane departures and crashes of all types during dark conditions by up to 22.1 and 24.5 16
percent, respectively. According to cost analysis results, chevrons are also a very cost-effective 17
countermeasure, with a benefit-cost (B/C) ratio exceeding 8:1 (6). 18
19 Dynamic Curve Warning Systems 20
Dynamic curve warning systems (DCWSs) detect the speed of approaching vehicles and are 21
programmed to provide drivers exceeding a certain speed threshold with a message, flashing light-22
emitting diodes (LEDs), or a display of their speed (Figure 3b). Results from a national safety 23
study indicate that, two years after installation, a 2.0 mph mean speed reduction occurred at the 24
beginning of the curve (8). The Oregon Department of Transportation (ODOT) installed a DCWS 25
system in advance of a curve on Interstate 5 near Myrtle Creek in Douglas County. The system 26
consists of a dynamic message sign, a 45-mph advisory speed sign, a controller unit, a radar unit, 27
and computer software. The analysis results showed that 76 percent of drivers slowed down 28
following the system’s installation, with a 2.6 mph reduction in mean speed for passenger cars (6). 29
30 Advanced Curve Warning and Advisory Speed Sign 31
Curve or turn warning signs are placed in advance of curves to alert drivers of what lies ahead on 32
their route (Figure 3c). Properly installed curve warning signs have been proven to improve safety 33
for horizontal curves. The cost for most commonly used curve warning signs with advisory speed 34
plates ranges from $500 to $700 per sign (9). The Kentucky Transportation Cabinet (KYTC) 35
installed an LED-enhanced curve warning sign on KY 82 in Estill County. Since its installation in 36
2006, no fatalities have been recorded, despite a crash history of one fatality per year for three 37
consecutive years prior to the installation of the sign (6). 38
39
Pavement Safety 40
Pavement safety countermeasures can also make significant contributions to reducing the number 41
of RwD crashes. Insufficient friction between the tire and pavement surface, poor visibility during 42
nighttime hours, and pavement drop-off edge are factors that may contribute to a vehicle leaving 43
the traveled way. 44
45
46
Jalayer, Zhou, Satterfield  11
High Friction Surface Treatments 1
High friction surface treatments (HFSTs) consist of a thin layer of durable aggregates (typically 2
calcined bauxite) that are highly resistant to polishing (10) (Figure 3d). The aggregate is bonded 3
to asphalt, concrete, or other pavement surfaces using polymer binders. HFST is not meant to 4
change the pavement’s structural performance. Rather, HFST provides greater friction, allowing 5
motorists to maintain better control in dry and wet road conditions, resulting in reduced numbers 6
of RwD crashes. According to the FHWA Every Day Counts (EDC) 2012 Initiatives, a B/C ratio 7
of about 24:1 can be achieved by implementing pavement friction treatments (11). The KYTC 8
launched a 3-year HFST program to enhance friction on horizontal curves at 75 locations statewide 9
in 2010. The safety analysis results confirm that the total number of RwD crashes at the treated 10
sites dropped by 91/78 percent in wet/dry weather conditions (6). 11
12 Raised Pavement Markers 13
Raised pavement markers (RPMs) are often used by transportation agencies as delineation 14
treatments to improve nighttime visibility, particularly in wet pavement conditions (Figure 3e). 15
According to the AASHTO’s Strategic Highway Safety Plan (SHSP), RPMs are considered to be 16
an effective, low-cost strategy for mitigating RwD crashes (12). Assisted by the FHWA and the 17
Alabama Department of Transportation (ALDOT), Mobile County in Alabama systematically 18
applied RPMs along 10 rural roadways with the highest number of RwD crashes. In this project, 19
RPMs were installed with 80-foot spacing in tangent sections of roadways, 40-foot spacing 20
between the advanced warning curve sign and the beginning of the curve, and 20-foot spacing 21
through the curve. Crash analysis results reveal an average annual decrease of about 86 percent for 22
RwD crashes and about a 94-percent reduction in injuries (6). 23
24 Edge Line Pavement Markings 25
Edge line pavement markings (Figure 3f) distinguish travel lanes from the adjacent shoulders to 26
delineate the travel path. According to the MUTCD, the edge line markings on the right edge of 27
the roadway shall be white. In addition, the normal width of edge line markings is 4 to 6 inches 28
and wide edge line markings are to be at least twice the width of a normal line (7). From 2009 to 29
2012, the Missouri Department of Transportation (MoDOT) initiated a program to install edge line 30
marking on eligible high risk rural roads (HRRRs). First, MoDOT performed a safety evaluation 31
of implemented countermeasures on 73 high-risk roadway segments. Based on the safety analysis 32
results, the total number of RwD crashes and severe RwD crashes decreased by 23 to 38 percent 33
following the installation of edge line markings (6). 34
35 Safety Edge 36
As determined by the FHWA in 2012, the Safety Edge is one of nine proven safety 37
countermeasures (Figure 3g). This strategy mitigates the vertical elevation difference by sloping 38
the edge of the pavement to 30 degrees during paving or resurfacing projects. A Safety Edge is 39
installed using one of several commercially available devices that can be attached to the hot-mix 40
asphalt (HMA) paver (13), and is also highly cost-effective. The added cost of resurfacing with 41
this treatment was determined to be very small, because the asphalt must simply be reformed to 42
create the Safety Edge. The Midwest Research Institute (MRI) conducted a safety evaluation of 43
the Safety Edge at 261 treated sites (685 miles) in Georgia and 148 sites (514 miles) in Indiana. 44
The evaluation results showed a 5.7 percent reduction in total crashes after the implementation of 45
the Safety Edge. Additionally, the B/C ratio for two-lane highways with paved shoulders ranged 46
Jalayer, Zhou, Satterfield  12
from 3.8 to 43.6 for Georgia and from 3.9 to 30.6 for Indiana. For two-lane highways with unpaved 1
shoulders, the B/C ratio ranged from 3.7 to 62.8 for Georgia and from 2.8 to 12.8 for Indiana (6). 2
3 Centerline Rumble Strips 4
Centerline rumble strips (CLRS) are a longitudinal safety feature that can be installed at or near 5
the centerline of undivided roadways (Figure 3h). CLRSs include a series of milled or raised 6
elements on the pavement (14). Tires rolling over rumble strips generate noise and vibration which 7
alert a distracted or drowsy driver to make a safe steering correction. The Michigan Department 8
of Transportation (MDOT) initiated a CLRS installation program during the period from 2008 to 9
2010. Approximately 5,400 miles of non-freeway roadways were included in this program. The 10
study results proved that the implementation of rumble strips resulted in a significant reduction in 11
both center line and edge line encroachments in tangent sections and through curves (6). More 12
specifically, after CLRS installation, the number of center line encroachments to the left side 13
within the curves dropped by 87 percent, and there was a 33 percent reduction in all crash types. 14
Additionally, the number of opposite-direction sideswipe collisions, multi-vehicle head-on 15
crashes, and single-vehicle RwD crashes decreased by 46, 35, and 31 percent, respectively (6). 16
17 Shoulder Rumble Strips 18
Shoulder rumble strips (SRS) are commonly installed in paved shoulders that are adjacent to the 19
travel lane (Figure 3i). Like CLRS, SRS provide acoustical and vibrational warnings to drivers 20
who are straying from their travel lane. According to survey results from 50 state DOTs, the B/C 21
ratio for SRSs was estimated to be approximately 50:1 (15). The WSDOT investigated the 22
possibility of applying SRS on undivided highways. To date, WSDOT has installed over 260 miles 23
of a mix of milled and raised SRS on its rural two-lane undivided highways. In early 2013, the 24
WSDOT undertook a review of historical crash data over the nine years from 2002 to 2010. The 25
study examined a total of 190 roadway miles with SRS in 45 segments, covering all geographic 26
areas of the state (6). In cases where SRS had been added during or after CLRS installation, the 27
results showed that run-off-road to the right side (ROTRR) crash rates were reduced by 47.061.6 28
percent for crashes of all severity types, and by 15.366.6 percent for fatal and serious injury 29
crashes. 30
31
Roadside Design 32
The probability of the severity of ROR crashes depends on the roadside features, including 33
sideslope, fixed-object density, offset to fixed objects, and shoulder width (16-21). Collision with 34
a fixed object has been identified as the primary harmful event in ROR crashes (22). A recent 35
inquiry of the FARS database revealed that 7,416 people perished in crashes involving roadside 36
fixed objects in 2012, accounting for 22 percent of the total fatalities for that year (3). Some 37
practical countermeasures to enhance roadside safety include roadway cross-section 38
improvements, hazard removal or modification, and delineation. These countermeasures have been 39
utilized in all area types (i.e., rural, suburban, and urban) to keep vehicles in travel lanes and to 40
reduce potential collisions with roadside objects, such as trees, signs, and utility poles (1). 41
42 Cable Barrier 43
A barrier is a device designed to stop or redirect errant vehicles to prevent a more serious crash. 44
Although barriers cannot reduce the total number of crashes, the benefits of cable barriers are that 45
they tend to minimize the severity of injuries by absorbing the impact of the crash and have safer 46
Jalayer, Zhou, Satterfield  13
consequence compared to vehicles striking the shielded obstacles. Flexible barriers, made from 1
wire rope strung between posts (Figure 3j), are the most forgiving type of barriers and the best 2
option for minimizing injuries to vehicle occupants (23). A number of high-tension cable barrier 3
systems are available, which remain functional after a crash and may not require immediate repairs. 4
In 20042008, the Minnesota Department of Transportation (MnDOT) installed cable barriers at 5
31 segments along approximately 150 freeway miles to reduce the number of fatalities and severe 6
injuries caused by cross-median crashes. The safety evaluation results revealed that the number of 7
fatal cross-median crashes and serious injury cross-median crashes after cable barrier installation 8
dropped from 19 to 0 and 8 to 6, respectively (6). 9
10 Guardrail 11
Guardrails (Figure 3k) are the most common and widely used type of barrier and can be effective 12
in reducing: 13
reportable RwD crashes, 14
vehicles from hitting fixed objects, and 15
vehicles from going over steep embankments. 16
The most common guardrail system used in the United States is the metal beam guardrail, 17
which consists of W-shaped metal beam rail elements fastened to wood or galvanized steel posts. 18
Guardrails have a low life-cycle cost since they often remain functional without immediate need 19
of repair (12). The North Carolina Department of Transportation (NCDOT) evaluated the results 20
of spot safety and hazard elimination projects of 14 divisions in the state. Using a before-after 21
analysis at the three treatment sites, the results showed that the percentage reduction in the total 22
Severity Index and RwD Severity Index range from 16.6 percent to 36.7 percent. In this study, 23
crash severity index was defined as being equal to the total number of equivalent property damage 24
only (PDO) crashes (76.8 for “K=Fatal” and “A= Incapacitating injury” crashes, and 8.4 for 25
“B=Non-Incapacitating injury” and “C=Possible injury” crashes) divided by the total number of 26
crashes (6). 27 28 Shoulder Widening 29
Roadway shoulders, when used as a safety feature, can improve road safety not only by allowing 30
drivers to recover in a stable, clear recovery area, but also by providing drivers with more space to 31
maneuver to avoid crashes. In addition, a wider shoulder improves stopping sight distance (SSD) 32
on horizontal curves and provides better bicycle accommodation (Figure 3l). Shoulder width can 33
vary between 2 feet for minor rural roads and 12 feet for major roads. It can also be widened both 34
inside and outside curves (24). For low-volume roads (less than 1,000 vehicles per day) with 35
narrow pavement width (less than 12 feet), it is more effective to consider narrower lanes with a 36
wider shoulder (25). Based on a before-after analysis of three treatment sites, the NCDOT showed 37
reductions in the total Severity Index and RwD Severity Index ranging from 43.7 percent to 69.2 38
percent (6). 39
40 Breakaway Supports for Signs and Lighting 41
Breakaway supports (Figure 3m) refer to various devices designed and constructed to break or 42
yield when they are hit by a vehicle (26). It is not always feasible to maintain object-free roadside 43
clear zones (the total roadside border area starting at the edge of the traveled way); however, crash 44
severity can be diminished by using breakaway supports for roadside objects. The 2009 MUTCD 45
mandates that post-mounted roadside sign supports in the clear zone be breakaway, yielding, or 46
Jalayer, Zhou, Satterfield  14
shielded (7). In phone interviews with traffic and safety engineers from several state DOTs 1
regarding the safety effects of breakaway supports, most agencies reported that this 2
countermeasure has been proven to be effective in reducing the severity of RwD crashes and that 3
no evaluation has been deemed necessary. 4 5 Clear Zone Improvements 6
A clear zone is defined by the 2011 Roadside Design Guide as The unobstructed, traversable 7 area provided beyond the edge of the through traveled way for the recovery of errant vehicles8
(27). This area includes shoulders, bike lanes, and auxiliary lanes, excepting those auxiliary lanes 9
that function as through lanes (Figure 3n). Clear zone distances are most affected by traffic volume, 10
speed, roadside slope, and curvature (27). In 2006, the Iowa Department of Transportation (Iowa 11
DOT) initiated a program to mitigate RwD crashes, mainly focusing on the removal/relocation of 12
hazards (e.g., trees, telephone poles, mailboxes) within the clear zone area and shielding or 13
delineating objects, if achieving the first option was not feasible. The safety evaluation results 14
showed that the number of total crashes dropped by up to 38 percent (6). 15
16
SUMMARY AND CONCLUSIONS 17
An investigation of 14 real-world case studies has provided an overview of current safety 18
countermeasures practices for RwD crashes. These case study examples fall into three major 19
categories: signs (i.e., chevrons, dynamic curve warning systems, and advance curve warning and 20
advisory speed signs), pavement safety (high friction surface treatments, raised pavement markers, 21
edge line pavement markings, safety edge, centerline rumble strips, and shoulder rumble strips), 22
and roadside design (cable barrier, guardrail, breakaway supports for signs and lighting, clear zone 23
improvements, and shoulder widening). The results of this study identify pavement safety as the 24
most effective countermeasure for reducing total RwD-crash frequency and severity. 25
26
ACKNOWLEDGEMENT 27
This study was sponsored and supported by the American Traffic Safety Services Association 28
(ATSSA). The authors extend their sincere thanks to the project panel members for their 29
comments/inputs. 30
31
REFERENCES 32
1. Neuman, T. R., R. Pfefer, K. L. Slack, K. K. Hardy, F. Council, H. McGee, L. Prothe, and 33
K. Eccles. A Guide for Addressing Run-Off-Road Collisions. Publication NCHRP Report 34
500, Vol. 6, Transportation Research Board, Washington, D.C., 2003. 35
2. Neuman, T. R., R. Pfefer, K. L. Slack, K. K. Hardy, F. Council, H. McGee, L. Prothe, and 36
K. Eccles. A Guide for Addressing Head-On Collisions. Publication. Publication NCHRP 37
Report 500, Vol. 4, Transportation Research Board, Washington, D.C., 2003. 38
3. National Highway Traffic Safety Administration (NHTSA). Fatality Analysis Reporting 39 System (FARS). www.nhtsa.gov/FARS. Accessed May 1, 2015. 40
4. Federal Highway Administration (FHWA). Roadway Departure Safety, Washington, D.C., 41
2014. 42
5. American Association of State Highway and Transportation Officials (AASHTO). Driving 43 Down Lane Departure Crashes: A National Priority, Washington, D.C., 2008. 44
6. American Traffic Safety Services Association (ATSSA). Preventing Vehicle Departures 45 from Roadways, Virginia, 2015. 46
Jalayer, Zhou, Satterfield  15
7. Federal Highway Administration (FHWA). Manual on Uniform Traffic Control Devices 1 (MUTCD), Washington, D.C., 2009. 2
8. Hallmark, S. L., Y. Qui, N. Hawkins, and O. Smadi. Crash Modification Factors for 3 Dynamic Speed Feedback Signs on Rural Curves. Center for Transportation Research and 4
Education, Iowa State University, Ames, Iowa, 2014. 5
9. Oregon Department of Transportation (ODOT). Systemic Safety Measures-Updated Curve 6 Warning Signs. www.oregon.gov/ODOT/HWY/TRAFFIC-7
ROADWAY/docs/pdf/UpdatedCurveWarningSigns.pdf. Accessed May 1, 2015. 8
10. American Traffic Safety Services Association (ATSSA). High Friction Surface 9 Treatments, Virginia, 2014. 10
11. Federal Highway Administration (FHWA). Every Day Counts (EDC)-High Friction 11 Surface Treatment, Washington, D.C., 2012. 12
12. American Traffic Safety Services Association (ATSSA). Cost Effective Local Road Safety 13 Planning and Implementation, Virginia, 2011. 14
13. Federal Highway Administration (FHWA). The Safety Edge-A Pavement Edge Drop-off 15 Treatment. Publication FHWA-SA-10-034. FHWA, U.S. Department of Transportation, 16
Washington, D.C., 2012. 17
14. Federal Highway Administration (FHWA). Technical Advisory: Shoulder And Edge Line 18 Rumble Strips. Publication T5040.39. FHWA, U.S. Department of Transportation, 19
Washington, D.C., 2011. 20
15. Oklahoma Department of Transportation (OKLADOT). Shoulder Treatments-Rumble 21 Strips (Milled SRS, Rolled SRS). www.okladot.state.ok.us/oshsp/pdfs/ld-22
shouldertreatments.pdf. Accessed May 1, 2015. 23
16. Jalayer, M., J. Gong, H. Zhou, and M. Grinter. Evaluation of Remote Sensing Technologies 24
for Collecting Roadside Feature Data to Support Highway Safety Manual Implementation. 25 Journal of Transportation Safety and Security, Vol 7. No. 4, 2015, pp. 345-357. 26
17. Gong, J., H. Zhou, C. Gordon, and M. Jalayer. Mobile Terrestrial Laser Scanning for 27
Highway Inventory Data Collection. Presented at International Conference on Computing 28
in Civil Engineering, Clearwater Beach, FL., 2012, pp. 545–552. 29
18. Jalayer, M., H. Zhou, J. Gong, S. Hu, and M. Grinter. A Comprehensive Assessment of 30
Highway Inventory Data Collection Methods. Journal of Transportation Research Forum. 31
Vol. 53, No. 2, 2014, pp. 73-92. 32
19. Jalayer, M., Hu, S., Zhou, H., and Turochy R. E. Evaluation of Geo-tagged Photo and 33
Video Logging Methods to Collect Geospatial Highway Inventory Data. Journal of Papers 34 in Applied Geography. Vol. 1, No. 1, 2015, pp. 50-58. 35
20. Jalayer, M., Zhou, H., Gong, J., Hu, S., and Grinter, M. A Comprehensive Assessment of 36
Highway Inventory Data Collection Methods to Support Highway Safety Manual. The 37
94rd Annual Meeting of the Transportation Research Board, Washington D.C., 2015. 38
21. Jalayer, M., and Zhou, H. A Sensitivity Analysis of Crash Prediction Models Input in the 39
Highway Safety Manual. The 2013 ITE Midwest District Conference, Milwaukee, WI, 40
2013. 41
22. Noyce, D. A., C. Martinez, and M. Chitturi. The Operational and Safety Impacts of Run-42 off-road Crashes in Wisconsin: Tree, Fence, and Pole Hits. Wisconsin Department of 43
Transportation (WISDOT), Madison, Wisconsin, 2008. 44
23. International Road Assessment Program (iRAP). iRAP Safety Assessment and 45 Recommended Countermeasures. 46
Jalayer, Zhou, Satterfield  16
www.eurorap.org/media/134139/irap_mari_el_report_english_.pdf. Accessed May 1, 1
2015. 2
24. American Association of State Highway and Transportation Officials (AASHTO). A 3 Policy on Geometric Design of Highways and Streets. 6th Edition, Washington, D.C., 2011. 4
25. Federal Highway Administration (FHWA). Manual for Selecting Safety Improvements on 5 High Risk Rural Roads, Washington, D.C., 2014. 6
26. Federal Highway Administration (FHWA). Breakaway Features for Sign Supports, Utility 7 Poles and Other Roadside Features, Washington, D.C., 2012. 8
27. American Association of State Highway and Transportation Officials (AASHTO). 9 Roadside Design Guide. 4th Edition, Washington, D.C., 2011. 10
... One of the lateral driver behaviors that can be highly associated with the run-off-road crashes is lane-keeping ability. A recent study revealed that run-off-road crashes contribute to an average of 57% of motor vehicle traffic fatalities occurred in each year, where a major portion of these crashes occurred at nighttime and inclement weather conditions (Jalayer et al., 2015). Therefore, it is worth investigating driver lane-keeping performance in inclement weather (foggy weather in this study) considering the contribution of poor lane keeping in run-off-road crashes. ...
Article
Alongside human factors, contextual factors are believed to have an ongoing and complex impact on driving outcomes. However, how and to what extent the components of context influence driving outcomes (e.g. rule violations, crash, stress, fatigue) are far beyond full understanding. The purpose of this study is to explore the effects of weather condition, lighting condition and traffic density on driving outcomes. Thirty-six volunteers were enrolled to participate into a driving simulator-based experiment. Each participant was required to complete twelve trials of simulated driving under different sets of scenarios. Driving outcome was measured by five dependent variables: frequency of speeding, frequency of lane deviations, number of correct sign recognition, completion time and workload. The results showed the frequency of speeding was significantly affected by weather condition, lighting condition and traffic density. Lighting condition had a significant effect on number of correct sign recognition. Weather condition, lighting condition and traffic density had significant effects on task completion time. Weather condition and lighting condition had significant effects on driver’s workload. The implications of the results could help traffic safety professionals better understand the risk factors that may lead to human errors during driving. Practically, countermeasures could be inspired and developed to mitigate the adverse impacts brought by driving context to minimum.
Thesis
Driving in foggy weather conditions has been recognized as a major safety concern for many years. Driver behavior and performance can be negatively affected by foggy weather conditions due to limited visibility and shorter available perception-reaction time. In addition, random and unusual patterns of fog affect driver behavior greatly. A number of previous studies focused on driver performance and behavior in simulated environments. However, very few studies have examined the impact of foggy weather conditions on specific driver behavior in naturalistic settings. The second Strategic Highway Research Program (SHRP2) has conducted the largest Naturalistic Driving Study (NDS) between 2010 and 2013 on six US states to observe drivers performance and their interactions with roadway features, traffic, and other environmental conditions. The study conducted in this thesis utilized the SHRP2 NDS dataset to evaluate driver lane-keeping behavior in clear and foggy weather conditions. A total of 62 drivers involved in 124 trips in fog with their corresponding 248 matching trips in clear weather were selected for investigating lane-keeping behavior. Preliminary descriptive analysis was performed and a lane-keeping model was developed using ordered logistic regression approach to achieve the study goals. Individual variables such as visibility, traffic conditions, occurrence of lane-changing maneuver, driver marital status, geometric characteristics, among other variables, as well as some interaction terms (i.e., weather and gender, surface condition and driving experience, speed limit and mileage last year) have been found to significantly affect lane-keeping ability. An important finding of this study illustrated that affected visibility caused by foggy weather conditions decreases lane-keeping ability significantly. More specifically, drivers in affected visibility conditions showed 1.37 times higher Standard Deviation of Lane Position (SDLP) in comparison with drivers who were driving in unaffected visibility conditions. The outcome of this research may provide a better understanding of driver lane-keeping behavior and their perception of foggy weather conditions. This thesis also provided valuable insights into lane-changing characteristics based on driver behavior in fog and clear weather conditions. While a few studies focused on lane-changing maneuvers based on driver type, the impact of adverse weather conditions (especially in fog) was not addressed. This thesis examined lane-changing maneuvers in fog and clear weather conditions using the SHRP2 NDS dataset. A total of 125 drivers involved in 214 trips in fog with their corresponding 214 trips in clear weather were selected for analyzing the lane-changing characteristics. These participants performed 92 lane changes in heavy fog, 445 in distant fog, and 1,163 in clear weather conditions. The study tested several hypotheses to identify significant differences in number of lane-changing events per mile and lane-changing durations in fog and clear weather in different traffic conditions. In addition, different distributions of lane-changing durations were fitted to identify common trends. Using K-means cluster analysis technique and based on lane-changing behaviors, drivers were classified into two categories, conservative and aggressive. It was found that in heavy fog the mean lane-changing durations were significantly higher than clear weather in mixed-flow conditions. The cluster analysis results revealed that both conservative and aggressive drivers in heavy fog conditions had longer lane-changing durations than in clear weather. The comparison between the SHRP2 administrated survey questionnaires and the cluster analysis suggested that drivers’ responses related to foggy weather were more consistent with survey questionnaires compared to their responses in clear weather during free-flow conditions. The findings of this study have several practical implications. The result of lane-keeping behavior might be used to improve Lane Departure Warning (LDW) systems algorithm considering affected visibility by fog. The outcomes of lane-changing analysis could be used to classify drivers in real-time based on their lane-changing behaviors in a connected vehicle (CV) environment. The results might also be used in microsimulation model calibration and validation related to lane change in reduced visibility due to fog and various traffic conditions.
Article
Full-text available
For many years, state departments of transportation (DOTs) and local agencies have collected and maintained highway inventory data (HID) to assist the decision makers at different levels. In light of the implementation of the recently published Highway Safety Manual (HSM) in 2010, many state DOTs have sought to tailor the various safety measures and functions to evaluate the safety in their jurisdictions. Insufficient HSM-required HID in many current DOTs’ databases, however, necessitates the collection of the absent features. To obtain these data, various techniques for different purposes have been used, including field inventory, photo and video log, integrated Global Positioning System/geographic information systems (GPS/GIS) mapping systems, aerial photography, satellite imagery, terrestrial laser scanners, airborne light detection and ranging (LiDAR), and mobile LiDAR. Among many data collection methods, the photo and video log is widely employed by DOTs due to its simplicity and low cost. Therefore, the focus of this article, which is a timely and needed research effort, is to evaluate the capability of the photo and video logging method to collect HID for supporting HSM implementation through a comprehensive literature review, a nationwide survey, and a field trial. The results of this study demonstrate that the photo and video log can provide worthy and relevant HSM data sets with acceptable accuracy.
Article
Roadside feature data are critical inputs to highway safety models as described in the Highway Safety Manual (HSM). Collecting safety-related roadside feature data is an important step for HSM implementation. Many states’ department of transportations (DOTs) routinely collect data on roadside objects using a variety of sensing methods, which often incur in significant costs. At present, it is unknown which of these data collection methods or any combination of them is capable of efficiently collecting safety-related roadside feature data while minimizing costs and safety concerns. This research is designed to identify required roadside feature data for various types of facilities in the HSM and to characterize the capabilities of existing remote sensing methods (e.g., Mobile LiDAR) to collect those required data. In order to accomplish this objective, tasks such as literature reviews, a nation-wide survey, and large-scale field trial are performed in this research. The findings of this research suggest that either the mobile LiDAR or the combination of the video/photo log method with the aerial imagery method is capable of collecting required HSM-related roadside information. However, due to the high data reduction effort, the current mobile LiDAR method needs significant improvement in the data processing and in the feature extraction stage.
Prothe, and 33 K. Eccles. A Guide for Addressing Run-Off-Road Collisions
  • T R Neuman
  • R Pfefer
  • K L Slack
  • K K Hardy
  • F Council
  • H Mcgee
Neuman, T. R., R. Pfefer, K. L. Slack, K. K. Hardy, F. Council, H. McGee, L. Prothe, and 33 K. Eccles. A Guide for Addressing Run-Off-Road Collisions. Publication NCHRP Report 34 500, Vol. 6, Transportation Research Board, Washington, D.C., 2003. 35 2.
38 3. National Highway Traffic Safety Administration (NHTSA) Fatality Analysis Reporting 39
Report 500, Vol. 4, Transportation Research Board, Washington, D.C., 2003. 38 3. National Highway Traffic Safety Administration (NHTSA). Fatality Analysis Reporting 39
Publication FHWA-SA-10-034
  • Treatment
Treatment. Publication FHWA-SA-10-034. FHWA, U.S. Department of Transportation, 16
Federal Highway Administration (FHWA) Technical Advisory: Shoulder And Edge Line 18
  • D C Washington
Washington, D.C., 2012. 17 14. Federal Highway Administration (FHWA). Technical Advisory: Shoulder And Edge Line 18
Oklahoma Department of Transportation (OKLADOT). Shoulder Treatments-Rumble 21
  • D C Washington
Washington, D.C., 2011. 20 15. Oklahoma Department of Transportation (OKLADOT). Shoulder Treatments-Rumble 21
Mobile Terrestrial Laser Scanning for 27
  • J Gong
  • H Zhou
  • C Gordon
  • M Jalayer
Gong, J., H. Zhou, C. Gordon, and M. Jalayer. Mobile Terrestrial Laser Scanning for 27