Prepared by:
Steven L. Johnson, Ph.D., PE, CPE and Naveen Pawar
Department of Industrial Engineering
Funded by:
Mack-Blackwell Rural
November
2005
The researchers would like to
express sincere appreciation to the participating companies that provided
valuable advice and guidance during the conduct of this effort. Although many
individuals provided important input, the authors would like to specifically
thank Grant DuCote, Greer Woodruff and Henry Pianalto of JB Hunt Transport
Services; Shannon Lively, Jim McFarland and Mark Bradley of ABF Freight
Systems; and Rick Foster and Mark Helms
of Wal-Mart. We would also like to thank the many contributors from other
trucking organizations and the original equipment manufacturers that provided
valuable information that was not available from public sources. Lastly, we
would like to acknowledge the many truck drivers who offered their time and
opinions at the truck stops and rest areas. This research would not have been
possible without the contributions of the stakeholders.
1. Introduction
2. Review and Analysis of Literature
2.1
Setting Speed Limits Based on the 85th Percentile ……….……………………
5
2.2
Effects of Speed Limits on the Distribution of Traffic Speed ..
2.2.1 Effects of Posted Limits on Mean Speed and Speed Variance…
2.3
Effects of Speed Limits on Rural
2.3.1 General Trends in Highway Safety………………………………...…… 15
2.3.2 Methodological Issues Contributing to
Different Study Results .….... 17
2.3.3 Cause and Impact of Speed Variation …….
2.3.4 Effects of Speed on Individual Vehicle
Risk…..
2.3.5 Effects of Speed on Crash Severity ...………………………………….. 28
2.3.6 International Studies of the Safety Impact
of Speed Limits
2.3.7 Studies of Speed Limit Changes in the
2.3.7.1 Studies Prior to 1987 …………………………….…………..…. . 37
2.3.7.2 Impact of the 1987 Speed Limit Increase….………………….. . 38
2.3.7.3 Impact of the 1995 Speed Limit Increase
…………………….. . 45
2.3.8 The Effect of Differential Speed Limits on
Safety .…1
2.3.9 Cause and Impact of Truck Accidents
2.4
Effects of Speed on Driver Fatigue
2.5
Effects of Speed and Weight on Braking Distance……………………….…... 63
2.6
Effects of Speed on Operational Costs………………………….…………….... 65
2.6.1 Effects of Speed on Fuel Costs
2.6.2 Effects of Speed on Tire Costs
2.6.3 Effects of Speed on Maintenance Costs
2.7 Effects of Speed on Pollution
3. Research Methodology……………………………………………………….……..…... . 75
3.1
Measurement of Traffic Speeds on Highways with Different Limits……..….. 7
3.2 Computer Simulation Evaluation of Speed Differentials on Interactions
3.3 Assessment of Speed Limiters Use on Heavy
Trucks
3.4
Survey of Truck Drivers’ Opinions
3.5
Survey of Carrier Fleet Safety and Maintenance Personnel
3.6
Survey of Equipment Manufacturers of Trucks, Engines and Tires
3.7
Comparison of Fleet Experience in States with Different Speed Limits…..... 78
3.8 Financial Cost-Benefit Analysis of Operating Speeds ……………..……..….. 79
4. Analyses and Results…………………………………………………………………..…. 80
4.1 Traffic
Speed Measurement under Different Speed Limits Configurations… . 80
4.1.1
4.1.2
4.1.3
4.1.4
4.1.5 Summary of Speed Data from Different
Speed Configurations…….. 86
4.1.5.1 Speed Differentials and Compliance
4.1.5.2 Posted Speed Limits and Mean Speeds and
Differentials….. . 89
4.1.5.3 Posted Speed Limits and Speed Variance
4.1.5.4 Speed Differentials and Clustered Congestion……………….. 93
4.2
Impact of Speed Differentials on the Number of Vehicle Interactions…….…. 95
4.3
Speed Limiter Use on Heavy Trucks
4.3.1 Driver Category and Speed Limiter Settings………………………….1 00
4.3.2 Distribution of Speed
Limiter Setting……………………………….….101
4.3.3 Driver Experience and Speed
Limiter Setting………………………... 101
4.3.4 Fleet Size and Speed Limiter
Setting…………………………………. 101
4.4
Opinions of Truck Drivers
4.4.1 Characteristics of Vehicles and
Routes…………………….………… 103
4.4.2 Truck Driver’s Opinion of
the Effects of Vehicle Interactions……….105
4.4.2 Effects of Vehicle Interactions …………………………………..……. 104
4.4.3 Effects of Speed Differentials at On-Ramps
and Off-Ramps………. 108
4.4.4 Effects of Speed and Speed
Differentials on Driver Fatigue……….. 109
4.4.5 Effects of Speed Limits on Driver
Retention……………………..….. 110
4.4.6 Effects of Speed and Speed Differentials on Operating
Costs…….. 110
4.4.7 Comparison of Owner-Operator and Company
Driver Opinions…... 111
4.5 Opinions
of Carrier Fleet Safety and Maintenance Management………..…. 114
4.6 Opinions
of Original Equipment Manufacturers…………………………….........
4.6.1 Opinions of Engine Manufacturers
4.6.2 Opinions of Tire Manufacturers
4.7
Comparison of Fleet Experience in States with Different Speed Limits
4.7.1 Selection of Accident Data
4.7.2 Analyzing Accident Data by State Speed
Limits.
4.8
Financial Cost-Benefit Analysis of Operating Speeds……..........................
......
5. Summary………….……………………………………………………………………….. 1
5.1
Research on Truck Speed Effects
on Traffic Flow and Safety………..…..... 122
5.1.1 Impact of Speed Limits on Traffic Speed
…………………………..... 122
5.1.2 Impact of Speed Limits on Rural
5.1.3 Causes and Impact of Speed Variance
……………………………… 124
5.1.4 Impact of Speed on Crash Severity
…………………………….…..... 126
5.1.5 Impact of Differential Speed Limits on
Highway Safety …….……..... 126
5.1.6 Effect of Speed on Driver Fatigue
………………….……………….... 128
5.2
Effect of Speed on Operation Costs ………………………………………..... 128
5.2.1 Effect of Speed on Fuel Efficiency
………..………………………….. 128
5.2.2 Effect of Speed on Roadway Wear
………………………………….. 129
5.2.3 Effect of Speed on Tire Costs
………………………………………... 129
5.2.4 Effect of Speed on Engine Life and Routine
Maintenance Costs ..... 130
5.3
Financial Cost-Benefit Analysis of Operating Speeds …………………….... 130
5.4
Conclusions …………………………………………………………….………. 131
6. References …………………………………………………………………………...... 132
Appendices………………………………………………………………………….………. 146
A. Speed
Limits Before 55 mph NMSL in 1974 (Source: Atkinson, 1996)…...........
B. 1987
Speed Limit Increase ……………………………………………….............
C. 1995
Speed Limit Increase ……………………………………..………..….........
D. Rural
Interstate Speed Limits ….........................................................................
E. Summary
of Speed Data at Individual Sites……………………………...….. 150
F. Truck
Driver Survey ……………….………………………………...………... 152
G. Safety
Manager Survey ……………….…………………………...………..... 154
H. Maintenance
Manager Survey………………..……………...……………...... 156
I.
Survey
Statistics ……………………………………………………………….. 158
1.
Introduction
The setting of speed
limits has been controversial since the first speed limits were set in 1901. Other
than during the period of the National Maximum Speed Limit policy between 1973
and 1994, setting speed limits has historically been the responsibility of the
states. Posted speed limits on
In addition to
state-regulated maximum speed limits, traffic flow is affected by the fact that
most commercial truck fleets and many owner-operators have speed limiters on
their vehicles. These limiters result in speed differentials between many
trucks and automobiles, even if the posted limits are not different. The primary
reasons that trucking companies use speed limiters include safety and a
reduction in operating costs associated with fuel efficiency. The potential financial
benefits of increasing per-truck revenues versus the additional costs
associated with higher speeds are discussed. The objective is to provide
information for both regulatory agencies and commercial trucking operations in
the decision process of setting maximum truck speeds on rural interstate
highways.
The initial portion
of the report reviews the research and applications literature related to the
factors that are affected by vehicle speed. The empirical studies that have
addressed the effect of changes in highway speed limits on traffic flow and the
distribution of vehicle speeds are discussed. Understanding the causes of
highway accidents that involve trucks is important in order to evaluate the
effect of speed on highway safety. The causes of single and multiple vehicle
accidents involving trucks were reviewed. The extensive literature that has
dealt with the safety impact of increasing and decreasing speed limits at both
the national and state levels is critically reviewed. In particular, the
results of safety studies after the 1974 decrease in national speed limits to
55 mph and the subsequent increases in 1987 and 1995 are evaluated. The
methodological issues that help explain the many different conclusions drawn
from this body of research are presented.
The effects of both
absolute speed and differential speed limits are discussed in the context of
traffic flow and speed variation. Whether being due to state regulated limits
or company policies, the difference in speed between heavy trucks and
automobiles results in more speed variance. The research literature that
discusses the impact of speed variance on highway safety is presented.
There is a
relationship between vehicle speed and the amount of time required to cover a
particular distance. This is important for motorists, although it is
particularly important for commercial transport operations. The effect of
driving time has been an issue that has received a significant amount of
attention from the trucking industry, governmental agencies, and the general
public in the context of truck driver “hours of service.” The research
literature that addresses the effect of driving time and driving speed is
discussed with respect to driver fatigue.
In addition to the
safety implications, the operational costs associated with truck speeds are
important in a benefit/cost analysis. The research and applications literature
that pertains to the costs of direct costs such as fuel, tires, and maintenance
are discussed. In addition, the research that addresses the indirect costs such
as emissions and road wear are presented.
The next portion of
this research effort collected data in an attempt to fill some of the holes
that were observed in the literature. For example, although there was a very
large amount of research on speed limits, virtually none of the studies had
recognized the impact of speed limiters on heavy commercial trucks. Even the
studies that specifically analyzed increases in traffic speed when posted
limits were increased (e.g., 1987 and 1995) did not account for the fact that
the majority of heavy trucks, which often make up a large portion of the
traffic on interstate highways, could not increase their speed.
To address this
issue, empirical data were collected under four different speed limit
configurations. Data were collected on Interstate I-44 where the speed limit is
70 mph for both automobiles and trucks. The Cherokee Turnpike in
The empirical distributions
for truck and automobile speeds that were observed at two of the locations (
As previously
stated, even in states that have uniform speed limits on the rural interstates,
there is still a difference in the speeds of automobiles and heavy trucks due
to company speed limitation policies. The next portion of the study collected
data on the use of speed limiters by commercial trucking operations. The data
were collected from 236 drivers at truck stops in seven different states (AR,
IL, MO, OK, NM, AZ, TX). These drivers represented the full spectrum of
owner-operators operating under their own authority, lease/contract drivers,
and employees of large trucking fleets. The distribution of settings used by
these different groups provides an important reference point for understanding
how truck speed limiters affect traffic flow under different speed limit
configurations.
Surveys were
completed by the 236 drivers that addressed their opinions about speed limits
on rural interstates and speed differentials between automobiles and heavy
trucks. The surveys addressed perceived safety issues, as well as the drivers’
judgments about the effect of truck speed on operational costs (fuel, tires,
etc.) and psychosocial factors (driver fatigue, stress, and driver retention). The
specific effects on drivers of speed differentials, whether due to posted limits
or company policies, were documented.
In addition to
collecting opinion data from the drivers, the opinion of commercial fleet
management personnel were obtained through a combination of surveys, on-site
visits, and communications at professional and trade meetings. In particular,
the opinions of fleet safety and maintenance managers were collected, along
with any data that the fleets had that pertained to the effects of truck speed
and speed differentials. These opinions were then contrasted with the information
that was obtained from the literature review and the opinions of the truck
drivers.
The last group surveyed represented the
original equipment manufacturers of the components that could be affected by
the vehicle speed. In particular,
manufacturers of commercial trucks, engines, and tires were surveyed with
respect to the effect of truck speed on their products. These communications included both technical
sales personnel and engineers in the various companies’ technical and research
centers.
Data from
participating companies were used to conduct an analysis of “virtual”
differential speed limits between automobiles and heavy trucks. The companies
with fixed maximum speeds that were limited to either 62 or 65 mph operated in
different states with different maximum speed limits for automobiles (65, 70 or
75 mph). By comparing the accident data from these different situations, the
impact of a “virtual” speed differential
between the fleets’ trucks and the automobiles was analyzed.
The last section of
the report addresses the financial benefit-cost relationships associated with
higher truck speeds. There is a trade-off between the benefits of increased
company revenue that could be attainable with higher truck speeds and the
increased operational costs incurred at higher truck speeds.
The issue of speed
differentials between automobiles and heavy trucks is a complex combination of
the impact on safety and financial considerations for both the truck drivers
and the commercial trucking organizations. This report addresses the currently
available published information, as well as the opinions of the various
stakeholders with respect to the benefits and costs of limiting heavy truck
speeds to below the traffic speed. This
information is important for both public policy and company policy related to
setting speed limits on rural interstate highways.
The objective of this
effort was to investigate the costs and benefits related to speed differentials
between heavy trucks and other vehicles on rural interstate highways. Truck
speeds are limited by a combination of state regulated speed limits and company
policies that limit truck speed with electronic control units on the trucks’ engines.
Both the effect of absolute speed and the speed of trucks relative to the other
vehicles in the traffic flow are important to understand the impact of heavy
truck speed policies. The initial phase of the effort involved a comprehensive
review of the research and applications literature that pertains to the topic.
The first part of
the literature review addresses the standard methods used to set posted speed
limits and the impact of speed limits on the speed distributions of both heavy
trucks and other traffic. The next section reviews the literature that has
documented how speed limits and speed limit changes affect accident and
fatality rates in the United States and internationally. The extensive number
of studies that have investigated the safety impact of increases and decreases
in speed limits has been reviewed. The last part of that section specifically
addresses the causes and impact of heavy truck accidents and the impact of
speed differentials between trucks and automobiles. The research literature
pertaining to the relationship of vehicle speed and driver fatigue is
discussed.
The last sections
of the literature review address the research and applications literature on
the operational impact of speed. In particular, the effects of truck speed on
fuel consumption, tire costs, and maintenance costs are discussed.
2.1 Setting Speed Limits
Based on the 85th Percentile
The geometric features of the roadway, such as
horizontal and vertical alignment, sight distance, and cross-section determine
the highway design speed. The original definition of design speed, coined by the
American Association of State Highway and Transportation Officials (AASHTO) in
1938, was “the maximum approximately uniform speed which probably will be
adopted by the faster group of drivers but not, necessarily, by the small
percentage of reckless ones” (Krammes, Fitzpatrick, Blaschke, and Fambro,
1996). AASHTO’s current definition of design speed is “the maximum safe speeds
that can be maintained over a specified section of highway when conditions are
so favorable that the design features of the highway govern” (AASHTO,
2001). This is the maximum speed prudent
drivers would choose when environmental conditions are very good and traffic
volumes are low. Subject to the constraints of environmental quality, economics,
aesthetics, and social impacts, AASHTO recommends higher design speeds to
promote safety, mobility, and efficiency.
Design speed is highly sensitive to certain highway design features like
curvature, sight distance, and roadside elements.
When speed limits are set based on design speed, the
posted speed limit is generally lower than the design speed because it is known
that some drivers will tend to drive faster and also that the road conditions
are sometimes poorer than were used in the design standards (Persaud, Parker,
Knowles and Wilde, 1997). However, according to Abraham and Abdulhai (2001), a
speed limit that is set using this as a basis will often appear unrealistic to
drivers since the limit is for an entire highway segment, even though is often
reflects relatively few elements.
According to AASHTO (2001) posted speed limits are
usually set to approximate the 85th percentile speed of traffic. For
many rural highways, it is a common practice to establish the speed limit near
the 85th percentile speed.
The term “85th percentile speed” is the speed at or below
which 85% of drivers travel in free-flow conditions at representative locations
on the highway or roadway section (National Research Council, 1998). The 85th percentile speed is determined
through spot speed studies of “free flowing” traffic (i.e., traffic unimpeded
by other vehicles) (Krammes, Fitzpatrick, Blaschke and Fambro, 1996). According
to AASHTO (2001) the 85th percentile speed is usually within the
“pace” or the 10 mph speed range used by most drivers. In general, the speed
limits for rural interstates are set below the 85th percentile speed
limits. Harkey, Robertson and Davis (1989) collected data from urban and rural
highways in
The 85th percentile speed for a
distribution of speed observations is shown in Figure 1. In most cases, the difference between the
85th percentile speed and the average speed provides a good approximation of speed
standard deviation, which is another important factor that relates to the
speed-safety relationship.
The distribution of
traffic speeds on any particular highway is affected by the posted speed limits
and the enforcement of the limits. The observed 85th percentile
speed on a highway with a 65 mph speed limit will be different than the 85th
percentile speed on a highway with a 75 mph speed limit, even if they are both
rural interstates with identical geometries.
For this reason,
although it is discusses many times in the context of setting speed limits on
rural interstate highways (Governors Highway Safety Association, 2005), the
concept of “design speed as defined by the 85th percentile” does not
appear to make apply. Safety, efficiency,
and economics have played a significant part in the process of setting limits.
This is shown by the large differences in speed limits set on similar highways
in different states.
The first speed
limit in the
The National
Highway Traffic Safety Administration (1992) analyzed the speed data available
from 18 of the 40 states that increased the automobile speed limits from 55 mph
to 65 mph in 1987. The average speed of automobiles increased from 60.4 mph in
1986 to 64 mph in 1990. It was concluded that the increase in the speed limit
significantly increased the average traffic speed. However, another way of looking
at the same statistics is that the average driver’s speed exceeded the posted
speed limit by 5.4 mph in 1986, while in 1990 the average speed was actually 1
mph below the posted speed limit.
Freedman and
Esterlitz (1990) measured the effect of increased speed limits on the traffic
speed in Virginia and found that a 10 mph increase for automobiles speed limit,
from 55 mph to 65 mph (leaving truck limits at 55 mph), resulted in an increase
in the average speed of automobiles of 2.8 mph (63.1 to 65.9 mph) within one
month of implementation. Later, as drivers “adapted” to the new speed limits,
the average speed gradually increased, reaching 66.9 mph after one year. The
authors contended that the percentage of automobiles “over speeding” (traveling
above 65 mph) doubled from 32% to 69%. Again, however, another way of
presenting the statistics is that the compliance rate increased and the average
speed was reduced from 8 mph above the speed limit to only 0.9 mph above the
speed limit. The conclusion as to the
effect of a speed limit on “speeding” depends upon the definition. The average speed observed in this study was
significantly lower than the average speed observed by the National Highway
Traffic Safety Administration (1992) because the former study considered only
the automobile speed, whereas the latter study included heavy trucks.
Godwin (1992) found
that an increase of 10 mph (55 mph to 65 mph) increased the average traffic
speed by 3 mph (60.2 mph in 1986 to 63.2 mph in 1988). In the same period, the
average speed in states that maintained the 55 mph speed limit increased by 0.9
mph (58.7 mph to 59.6 mph). Nakao (1989)
found similar results for automobile speed data. A 10 mph speed limit increase (55 mph to 65
mph) resulted in 2.5 mph increase in average speed (62.4 mph to 64.9 mph) from
April 1987 to September 1987. However,
the observed speed change might have been greater if the data were collected later,
when the drivers had “adapted” to the new speed limits. Any increase in the speed
limit is followed by a “transition” period and then by “adaptation.” During the
initial “transition” period, the drivers’ speed does not increase suddenly to
the new higher speed, although it does increase gradually. After the transition period, they become
“adapted” to the new higher speed limits and travel at the higher speeds. Ledolter and Chan (1996) found that after the
1987 increase in the speed limit in
McKnight, Klein,
and Tippetts, (1989) analyzed nationwide data from 1983-1988 and found that the
number of drivers spotted “speeding” increased by 48% for the states which had
increased their maximum speed limit to 65 mph; whereas the number of drivers
observed “speeding” increased by only 18% in the states that retained the 55
mph maximum speed limit. However, an important point to be noted is that the
definition of “speeding” in this study was “anyone traveling at speeds higher
than 65 mph” Obviously the number of people traveling above 65 mph in a 65 mph
speed limit state will be much higher compared to the number in a state with a 55
mph speed limit. It will be observed that in many of the studies discussed, the
researchers defined speeding as the percentage of drivers who exceeded 65 mph
because it is widely assumed that high speeds are the primary contributors to
fatal accidents. This definition of speeding does not consider the design speed
of the highways, which is a major factor in determining the effects of traffic
speed. Many of the highways included in
the studies have design speeds that far exceed 65 mph.
Agent, Pigman, and Webber
(1998) conducted a study to evaluate the effect of speed limits in
Similar results
were obtained by Parker (1992); however, his study was limited to only rural
and urban highways that were not limited access. Parker collected speed and
accident data from 100 sites in 22 states before and after speed limits were
altered. The average change in any of
the percentile speeds (i.e., 90th, 80th, etc.) at the
experimental sites was less than 1.5 mph, regardless of whether the speed limit
was raised or lowered. This indicates that
distribution of speed remains relatively constant and that the average speed of
traffic generally follows an increasing trend, irrespective of the change in
posted speed limits. The authors concluded that speed limits that are set close
to the 85th percentile speed had a beneficial effect on the drivers’ tendency
to comply with the posted speed limits. It was concluded that lowering and raising
the speed limits has relatively little effect on the traffic speed and that drivers
travel according to the traffic conditions.
Binkowski, Maleck,
Taylor, and Czewski (1998) studied the 1996 increase in speed limits for automobiles
from 65 mph to 70 mph in
Najjar, Stokes, Russell, Ali, and Zhang (2000) studied the results of the
1996 increase in maximum speed limit from 65 mph to 70 mph in
Borsje (1995)
studied the effects of having different speed limits on different highways within
the same highway category (referred to by the authors as differentiated speed
limits) in the
In addition to the
direct impact on traffic speed resulting from increases in posted limits on
highways, there are two indirect effects on traffic: speed spillover and
traffic diversion. Speed spillover results
when an increase in the speed limit on one highway increases the average
traffic speed on other highways that have not had an increased limit.
McKnight and Klein
(1990) studied the nationwide impact of increasing the speed limits on rural
interstate highways to 65 mph. It was found that for the states that raised
speed limits to 65 mph, speeding on rural interstates and on non-rural interstates
(highways still posted at 55 mph) increased by 48% and 9%, respectively.
Whereas, in states that maintained the 55 mph limit on rural interstates,
speeding increased by 18% and 37% on rural interstates and non-rural
interstates, respectively. It is important to note that “speeding” was defined
as the percentage of drivers who exceeded 65 mph for both the 55 mph and 65 mph
highways.
Nakao (1989)
analyzed the 1987 speed data from
Mace and Heckard
(1991) collected data between 1986 and 1988 from Illinois, Ohio, Texas and
Alabama and found that the average traffic speed for states that increased
their speed limit from 55 mph to 65 mph increased by 4 mph; whereas, on roads
still having a 55 mph posted speed limit in these states, the average speed
increased by only 0.8 mph. This study does not support a spillover effect.
A “traffic
diversion” effect occurs when an increase in the speed limits on certain
highways leads to an increase in traffic on the interstates that have a higher
speed limit and a reduction of traffic on highways with lower speed limits.
Lave and Elias (1994) observed the national traffic volumes before and after
the 1987 speed limit increase. They observed that there was a 73% greater increase
in vehicle miles traveled on the higher speed interstates compared to the
statewide value. The non-interstate vehicle miles traveled decreased by 11%. These
values illustrate that the speed limit increase resulted in traffic shifting from
lower speed limit roads to higher speed limit roads.
Comparing the
results of these studies indicates that the increase in speed limits does
appear to increase the average speed and the 85th percentile.
However, the magnitude of these increases has been found to vary significantly in
different studies. One of the reasons for the differences is the time duration
over which the studies were conducted. For example, the increase in average
traffic speed observed by Ledolter and Chan (1996) was much higher than the
increase observed by Nakao (1989). On possible reason for this difference is
that Nakao took only six months of data into consideration (during the
“transition” period), while Ledolter and Chan measured the speed increases over
10 years (when the drivers had adapted to the higher speed limits). Other factors,
such as the geography of different states, that affects the highway design
speeds and traffic volumes could account for the large differences in results
of the different studies. Borsje (1995) and Davis (1998) concluded that
enforcement can have an even greater effect on traffic speed than the posted
limits. The level from which the speed limit increased, whether it was raised
from 55 mph to 65 mph or from 65 mph to 75 mph, also caused differences in the magnitude
of increases observed by the different studies.
One very important factor
that most of the researchers failed to address, and may not even have realized,
is that the speed of most of the commercial heavy trucks are restricted to
below the posted speed limits by speed limiters, due to company policies. This obviously had a large effect on the
magnitude of traffic speed increases when posted speed limits were raised,
particularly for highways that have a relatively high proportion of heavy
trucks.
As previously
discussed, speed differentials between automobiles and heavy trucks occur due
to two primary factors. First, many states impose lower posted speed limits on
heavy trucks. These regulatory differentials range from 5 mph to 15 mph. The
second factor that results in speed differentials between automobile and heavy
trucks is the speed policy that is employed by commercial trucking companies.
Many companies use speed limiters on the truck engines to restrict the maximum
speed. These devices are becoming increasingly sophisticated in both their
ability to control speed and record the speed that is driven. The literature
discussed in this section relates to the effect of posted speed limits in that
there is virtually no literature that addresses the effect of company speed
policies on traffic speed in general, or truck highway speed, in particular. The
notation will characterize speed limits in the format: 70/65 for differential
limits of 70 mph for automobiles and 65 mph for trucks.
Mace and Heckard
(1991) collected data between 1986 and 1988 in Illinois, Ohio, Texas and
Alabama and found that the automobile speeds were 3.5 mph faster than truck
speeds on interstates with a uniform 65 mph speed limit; whereas automobile
speeds were 6 mph more than truck speeds on interstates with different speed
limits of 65 mph for automobiles and 55 mph for trucks. Therefore, a 10 mph speed differential
resulted in a change of 2.5 mph in the average speed difference between automobiles
and trucks.
Baum, Esterlitz, Zador
and Penny (1991) collected data in
Pfeffer, Stenzel,
and Lee (1991) conducted a time series analysis to study the impact of
differential speed limits for automobiles and trucks in
In 1994, Harkey and
Mera examined the impact of differential speed limit on average speed based on
data from 11 states, all having the same speed limit for automobiles but
different limits for trucks. The states were divided into three groups based on
their speed limits: 65/65, 65/60 and 65/55 mph. The mean speeds for automobiles
under these limits were 67.6, 67.8 and 67.4 mph, respectively, which were not statistically
different. However, the average truck speeds in these states were 63.8, 63.6
and 61.1 mph, respectively, for the 65, 60 and 55 mph truck limits. The average
truck speed in 65/55 states was significantly less than for the 65/65 and 65/60
mph states. According to this study, a speed differential of 5 mph (from 60 to
65) did not have a significant impact on the trucks’ speed and a 10 mph speed
differential decreased truck speed less than 3 mph. Furthermore, the percentage
of automobiles traveling above the speed limit by more than 10 mph was
significantly lower in the 65/55 mph (63.8%) states compared to the 65/65 mph
and 65/60 mph states (68.7 and 66.6% respectively). Even though the automobile speed
limit was uniform across all the states, it appears that the slower trucks in
the 65/55 mph speed limit states had the effect of reducing the average speed
of the automobiles. The non-compliance rate for trucks was much larger in the
65/55 and 65/60 speed limit group (89.4 and 76.5%, respectively) compared to
that in 65/65 group (35.6%).
Garber and Gadiraju
(1991) conducted a study in which they increased the speed limits from 55/55 to
65/55 on test sites and retained the uniform 55 mph speed limit on control
sites in
In the
Freedman and
Williams (1992) collected data from 11 northeastern states to estimate the
effect of differential speed limits on the mean speeds and 85th
percentile speeds. Six of these states had remained at 55/55 mph, three had
increased to 65/65 mph and two employed differential speed limits of 65/55 mph.
It was found that the average speed of automobiles in the states with 65 mph
speed limit was 2 to 5 mph faster than those with 55 mph limits. For trucks, the
mean speeds were 3 to 7 mph faster in states with a 65 mph speed limit than in
states with 55 mph limits. The results indicated that the average truck speed was
more sensitive to the posted speed limit than was the average automobile speed.
This could have been due to the fact that the compliance rate of trucks was higher
than the compliance rate of automobiles. For automobiles, there was no
significant difference in the average speed or the 85th percentile
speed in the 65/55 mph speed limit states (67.7 and 72.2 mph) compared to the
65/65 mph speed limit states (66.7 and 72.1 mph). However, the average and the
85th percentile automobile speeds for the 55/55 mph states were
significantly lower (63.0 and 68.7 mph). The results indicated that the lower
truck speed in differential speed limit states did not have any significant
effect on the average speed of automobiles. The mean and the 85th
percentile speeds of trucks were also not significantly different for states
with 65/55 mph speed limit (61.6 and 66.3 mph, respectively) compared to those
for the 55/55 mph limit states (60.2 and 65.3 mph, respectively). However, the
mean and the 85th percentile truck speed for the 65/65 speed limit
states were significantly higher (65.0 and 69.8 mph). The conclusion was that lower
speed limits for trucks did reduce the average and the 85th
percentile truck speeds. These results were in contrast to the opinions
expressed by Ganote (1997), who believed that a differential speed limit does
not really succeed in lowering truck speeds because the drivers takes into
account the prevailing road conditions.
Most of the studies
have concluded that a 10 mph posted speed differential does not produce a 10
mph difference in the average speed of the automobiles and trucks. In addition,
even under uniform speed limits, the average speed of trucks is 3 mph to 4 mph
slower than the average speed of the automobiles. It was also observed by
Harkey and Mera (1994) that the average speeds of automobiles and trucks are
similar in 65/65 mph and 65/60 mph states, indicating that a speed differential
of 5 mph does not have any significant impact on the truck speed.
The literature available on impact of speed limits on accidents and
fatalities is reviewed in this section. It has been indicated in literature
that vehicle speed is only one of the factors that affect the probability and
type of accidents. The type of roadway and
the design speed of the highway are also important factors affecting the number
and type of accidents. Preston (1996) studied the accident records of
|
Accident Type |
Rural |
Urban |
|
Rear End |
12.90 % |
50.50 % |
|
Sideswipe |
7.30 % |
17.40 % |
|
Right Angle |
8.40 % |
2.40 % |
|
Head On |
1.50 % |
0.80 % |
|
Ran Off Road |
33.70 % |
17.80 % |
|
Hit Deer |
25.10 % |
0.40 % |
|
Other |
11.10 % |
10.70 % |
Figure 2
illustrates the amount of variation in the number of highway fatalities over
the last 40 years. To evaluate the effect of speed limits on highway safety, it
is important to consider the amount of exposure experienced by drivers in terms
of the vehicle miles traveled. Figure 3 illustrates that, although speed limits
have increased, the fatality rate (fatalities per 100 million miles traveled)
has been consistently improving. This is the result of improved safety
characteristics of both vehicles and roadways.
Figure 2. Trends in
National Fatalities
(Source.
Figure
4 shows both the number of fatalities on rural interstates and the vehicle
miles traveled. The trend in fatalities is upward; however, the trend in
vehicle miles traveled is also increasing.
Figure 5 illustrates that the trend in the fatality rate on rural
interstates was actually improving during that period.
65+ mph 55 mph 65 mph
65+ mph 65 mph 55 mph
Over the past 40
years, the relationship between highway speed limits and safety has received an
extraordinary amount of attention in both the research and popular
literature. There have often been
conflicting conclusions reported in this literature. Some studies have found
positive effects of higher speed limits, some found very negative effects and
many have not found there to be a relationship. There are a number of reasons
for these differences. It is apparent from a cursory review of the literature
that much of the public comment and even a significant amount of the research
is biased by the entities conducting the research. In addition, there are
serious methodological issues that need to be considered when interpreting the
research presented in the following sections.
The first
explanation for the differing results from different studies is simply the
natural variation that affects accident rates.
Figure 6 indicates the amount of variation in the number of fatalities
on rural roads in
The
results of speed limit studies can also be affected by the states or regions
compared. Figure 7 shows data from a
study by Ashenfelter and Greenstone (2004). They documented the fatality rates
for the states that adopted the 65 mph limits versus the states that retained
the 55 mph limits. It is apparent that the states that increased e speed limits
had a higher fatality rate both before and after the speed limit increase.
If the studies compared the two groups
after the change, without correcting for this difference, the results would not
represent the actual effect of the speed limit increases. The time frame that is selected for the analysis can also significantly
affect the interpretation of the research results. Some of the studies compare the before-and-after
accident data to evaluate the effect of the speed limit increase. Notice in Figure 7 that there was a
significant drop in the fatality rate in 1989 for the states that maintained
the
55 mph limit.
Subsequently, in 1990 and 1991 the fatality rate increased. By comparison,
the fatality
rates for the states that increased their speed limits decreased steadily from 1989 to 1992. If the relative rate of each group was used
in the analysis and the study compared 1986 to 1989 the conclusion could have been
that there was a large increase in the relative fatality rates for the states
that increased their limits. However, if the study had compared the data from
1986 to 1991, the conclusion could have been that there was no effect of the
increase in speed limits.
Another aspect of the time frame
issue is the adaptation that occurs when a speed limit is changed. There is
inertia to traffic speed when the limits are changed. The average speed and the
85th percentile speed do not change very much initially. In
particular, when limits are changed, a few drivers will adapt rapidly, moving at
new speed limit or even faster; whereas most drivers will increase their speed
gradually as they become more comfortable with the increased speed. The result
is that there is initially an increase in the speed variance among vehicles.
The negative effect of speed variance is potentially confounded with the effect
of the speed limit increase.
As previously discussed, the amount and severity of enforcement also has
a large effect on traffic speed behavior.
If enforcement was relatively lax when the speed limits were lower and
became more strict with new, higher limits, the actual effect of the change on
traffic speed might be minimal. In this case, the impact of increased “posted”
speed limits might have no effect on traffic behavior and, therefore, accident
rates.
The effect of having highway types with very different design speeds is
also important to the interpretation of the speed limit studies. The current
study is focused on rural interstates. Most of the research combined all
highways, some with low design speeds and others (i.e., rural interstates) with
design speeds that are significantly above the posted speed limits. Even for
the studies that specifically address the speed limits on interstate highways,
most do not differentiate between urban and rural interstates. It is often
difficult to extrapolate the results of these studies to rural interstates, in
particular.
The fact that trucks have limiters that often do not allow them to go
the posted speed limit also has an effect on the interpretation of speed limit
studies. When limits were increased from 65 mph in 1995, many, if not most, of
the commercial heavy trucks on the interstate highways were restricted to a speed
of 62 or 65 mph. As previously
discussed, this is the reason that the average vehicle speed generally
increases much less than the amount of the increase in the posted speed limits.
The volume of heavy trucks on the highway can have an effect on the traffic
speed. This issue has not been addressed in studies that have investigated the
safety impact of speed limits changes.
The archival databases that many studies have used for their analyses
include only fatalities and do not include accidents that do not involve a
fatality. The effectiveness of passive safety systems (i.e., seat belts, air
bags, etc.) have improved the “crash worthiness” of vehicles that are involved
in an accident. The result is that the relationship between fatalities and
total accidents changes as a function of time. This is particularly the case
for speed limit studies. The simple physics of higher speed accidents could
have a proportionately larger impact on fatalities than on the number of
accidents. Studies that only address fatalities
can come to very different conclusions about the safety implications of speed
limits compared to studies that include non-fatal accidents.
The last methodological issue that makes the interpretation and
comparison of studies in this area difficult is the use of the number of fatalities
or accidents, rather than the fatality or accident rates. As previously
discussed, studies that simply look at the number of fatalities or accidents,
without considering the vehicle miles traveled, can come to different
conclusions than those that include vehicle miles traveled. This again is
particularly the case for speed limit studies. There is an inverse relationship
between speed and exposure time on the highway. That is, for a given mileage
driven, a driver (truck or automobile) is exposed to the potential of a
collision longer at lower speed limits.
The objective of this section was to introduce some of the
methodological issues that limit the interpretability of much of the vast
amount of research literature on the relationship between safety and posted speed
limits. In particular, many of these issues make it difficult to extrapolate
the research findings to truck speeds on
rural interstates. As the safety research is reviewed in the following
sections, these methodological issues should be kept in mind.
2.3.3 Cause and Impact of Speed Variation
Although there has
been a debate as to the impact of speed limits on accidents, one aspect on
which most of the research is consistent is that speed variance can have a
significant impact on the probability of accidents. There are four primary
methods of calculating speed variance reported in the literature: (a) the standard
deviation of the individual vehicle speeds, (b) the difference between the 85th
percentile speed and the median speed (50th percentile), (c) the
difference between the 85th percentile speed and the mean speed, and
(d) the difference between the 85th percentile and the 15th
percentile speed. However, for the data analysis section of this report, only
the first two of the above four methods were used to calculate the speed
variance.
It has been widely
acknowledged that an increase in speed variance is often associated with an increase
in the probability of accidents. According to the National Research Council
(1998), the narrower the speed distribution (e.g., less spread between the
average speed and the 85th percentile speed), the greater the safety
benefits.
Garber and Gadiraju
(1988 and 1989) found that the level of safety on any highway is related to the
characteristics of the traffic stream and the geometry of the highway. It was
found that the major factor that affected speed variance was the difference
between the posted speed limit and the design speed of the highway. Speed
variance was observed to be the lowest when the posted speed limit was 6 to 12
mph lower than the design speed of the highway. The accident rates were
observed to increase with increasing speed variance for all classes of roads.
For average speeds up to 70 mph, speed variance decreased with increased
average speed. The authors also concluded that the accident rates on a highway
do not necessarily increase with an increase in average speed.
Lave (1985)
collected nationwide average speed and 85th percentile speed data
for 6 different types of highways (rural and urban interstates, arterials and
collectors) from 48 states for 1981 and 1982. Speed dispersion was calculated
as the difference between 85th percentile speed and the mean speed. Speed,
by itself, was not found to have a significant effect on fatality rates.
However, when using speed variance as the metric, 10 out of 12 road types indicated
a statistically significant positive relationship. This result indicated that it is not absolute
speed, but the speed variance that increases fatality rates. It was also
observed that, speed variance decreased with increases in the average speed. A
series of responses to Lave’s models by Levy and Asch (1989), Fowles and Loeb
(1989) and Synder (1989) confirmed the negative effect of speed variance, but also
suggested that there is also an impact of average speed on fatality rates. One
common potential drawback in all of the above models is that the speed data and
accident data do not belong to the same highway types. The fatality data for
all road types were combined and then used with interstate average speeds in
their models. Therefore, the results must be interpreted with care (Monsere,
Newgard, Dill, Rufolo, Wemple, Bertini and Miliken, C., 2004).
Graber and Gadiraju
(1991) studied the impact of a speed limit increase on speed variance in
Baxter (1999) and
Addis (1999) also held a similar opinion of the relationship between speed and
safety. According to Baxter, accidents will increase only if speed increases
beyond the design speed of the highway; whereas, if the posted speed remains
below the design speed of the highway, there will not be a significant increase
in accidents as speed limit increases. Addis (1999), also stressed, although
with no data to support his claim, that speed variance has a significant effect
on the fatality rate and that speed, alone, has no effect on fatality rate.
Garber and Ehrhart
(2000) conducted a study of traffic speed, traffic flow and geometric characteristics
on the crash rates for
A study conducted
by Rajbhandari and Daniel (2002) examined the effects of increase in speed
limits from 55 mph to 65 mph in New Jersey in 1998. The data were collected
from 1997-2000. The increase in speed limit to 65 mph caused more speed
variance between automobiles and trucks and increased the accidents that involved
trucks by 19% (772 per year to 919 per year). There was also a 27% increase in
total accidents in the same period.
Fitzgerald (1989)
studied the increase in the speed limit of trucks from 80 kph to 90 kph, while
retaining the 100 kph speed limits for automobiles in
Liu (1998) examined
accident data from 1969 -1995 in
Godwin (1992)
studied the impact of a 1987 speed limit increase on the speed variance. The
standard deviation of traffic speed increased by 0.8 mph (6.1 to 6.9 mph) for
the states that retained the 55 mph speed limit. For the states that increased
their speed limits, the standard deviation increased by only 0.2 mph (6 to 6.2
mph). Similar conclusions were drawn by Binkowski, Maleck, Taylor and Czewski
(1998), who studied the 1996 increase in speed limits for automobiles from 65
mph to 70 mph in
Pfeffer, Stenzel
and Lee (1991) conducted a time series analysis to examine the impact of
differential speed limits on speed variance in
Freedman and
Esterlitz (1990) measured the effect of increased speed limits on traffic speed
and found that in Virginia, the 10 mph speed limit increase from 55/55 to 65/55
mph, had no significant effect on the standard deviation (a measure of speed variance)
of automobiles and trucks, even after one year of speed limit change.
To analyze the
impact of the increase in speed limits on the speed distribution of vehicles,
Nakao (1989) compared
Zlatoper (1991)
analyzed nationwide data in 1987 and found average speed, speed variance, and
traffic volume to be directly related to accidents. Other factors, such as
spending on highway police and safety, income levels , inspection laws, and
seat belt laws were found to be inversely related to the number of accidents.
Radwan and
Sinha (1978) studied the effect of the decrease in speed limit from 70 mph to
55 mph on truck crashes in
Agent, Pigman and
Webber (1998) conducted a study to evaluate the impact of increasing speed
limits from 55 mph to 65 mph on rural interstates in
Harkey and Mera (1994) examined the impact of
differential speed limits on traffic speed variance based on an investigation
of speed data from 12 states (26 sites) divided into four different speed limit
groups (65/65, 65/60, 65/55 and 55/55 mph). The variance of truck speeds was
higher than for automobile speeds when the truck speed limit was higher. Due to
the speed limiters on trucks, not all trucks could travel at the higher speeds,
resulting in more speed variance for trucks. They found differences in truck
speed variance for ten of thirteen pair-wise comparisons between uniform and
differential speed limit sites. No significant differences were found in the automobile
speed variances at the sites.
From the studies
reviewed it appears that differential speed limits increased the amount of speed
variance among vehicles because trucks travel at lower speeds than the automobiles.
When considering automobiles and trucks individually, different results were observed.
Increases in the speed limits decreased the speed variance among automobiles. However,
due to the presence of speed limiters on trucks, most of the trucks can not
travel at speeds above 70 mph. Therefore, if the speed limit for trucks is
raised to 75 mph the speed variance among trucks increases. Regarding the
impact of speed variance on traffic safety, most of the studies have agreed
that increases in speed variance increases the probability of accidents.
2.3.4 Effects
of Speed on Individual Vehicle Risk
In the previous
sections, the effect of traffic speed and speed limits on traffic safety was
discussed. This section focuses on the role of an individual vehicle’s speed on
the probability of being involved in an accident. It has been argued that an increase
in speed will increase the probability of accidents if the number of
interactions with other vehicles increases. Similarly, if a vehicle moves
slower than the traffic speed, the number of interactions will also increase.
Solomon (1964) conducted a comprehensive study on crashes and how other
roadway, driver, and vehicle characteristics affect the probability of being
involved in a crash. Approximately 600 miles of rural two-lane and four-lane
highways were studied using a spot speed sampling procedure. Interviews with
290,000 drivers were collected over a two-year time period. The travel speed
prior to the crash was collected from 10,000 crash records, as reported by the police
or by the driver. The estimated travel speeds from the accident records were
compared to the speeds measured at representative sites within each study
section. The comparisons indicated that vehicles involved in crashes were
over-represented in both high and low speed categories within the speed
distribution. The crash involvement rate was represented by a U-shaped curve as
a function of the amount of deviation from the average speed. The
accident-involvement, injury, and property damage rates were found to be
highest at speeds significantly below the average traffic speed. The accident
rates were least at the average traffic speed and increased with increasing
speed above the average traffic speed (Figure 8).
Cirillo (1968) also
conducted a study that addressed speed variation. Two thousand vehicles
involved in daytime crashes on interstate highways were analyzed. The data
represented a U-shaped curve similar to the Solomon data. The analysis took
into consideration only the crashes that involved two or more vehicles (rear
end, same direction sideswipe or angle collisions). Data were collected on
rural and urban section of interstate highways from twenty state highway
departments. The type of collision was controlled since the focus was on how
the differences in speeds of vehicles in the same traffic stream contributed to
crashes. The U-shaped curve obtained by Cirillo is shown Figure 8. According to
the Insurance Institute of Highway Safety (1991), one of the main concerns
regarding the validity of the results obtained by Cirillo is that only two-
vehicle accidents were considered while single vehicle crashes were not
included.
To address the
average speed of sections of highway not directly related to the crash location,
the Research Triangle Institute (1970) used a combination of trained on-scene
crash investigators and a system of automated continuous speed monitoring
sensors embedded in the roadway pavement to measure the speed of crash-involved
vehicles and their traffic speeds at the time and location of the crash. Data
were collected on 114 crashes involving 216 vehicles on state highways in
A study by Munden
(1967) conducted on the rural main roads in the
Fildes and Lee
(1993) studied the issues associated with speed and traffic safety in
Another Australian
study, conducted by Kloeden, studied the relationship between free traveling
speed and the risk of involvement in a casualty crash on rural highways with posted
speed limits of 80 kph or greater. A total of 83 crash cases were investigated.
The representative speed (average control speed) was obtained by measuring the
speeds of 830 control passenger vehicles that matched the 83 crash cases by
location, direction of travel, time of day, and day of week. The risk of
involvement in a casualty crash was found to increase more than exponentially
with increasing speed above the mean traffic speed (see Figure 9). Unlike the
results of the studies by Solomon and Cirillo, the traveling speeds below the
mean traffic speed were associated with a lower risk of being involved in a
casualty crash. The crash risk doubled with each 3 mph increase above the speed
limit. One of the possible reasons for the different results obtained by
Kloeden, compared to Solomon or Cirillo is that Kloeden studied the risk of
involvement in casualty crashes; whereas Solomon and Cirillo studied the risk
of involvement in any crash, irrespective of its severity. As the travel speed
increases, the accident severity increases.
Garber and Ehrhart
(2000) found that, as the mean speed increased, the crash rate decreased
slightly until the mean speed reached the posted speed limit of 65 mph, and
then the rate began to increase. The crash rate also increased as the mean
speed increased beyond the speed limit. It was noted that the changes in crash
rates were not necessarily caused by any one independent factor. The changes
were a result of the combined effects of independent factors like speed,
standard deviation, and traffic flow.
Hauer (1971)
performed theoretical analysis of “overtaking.” The study demonstrated that the
number of vehicle interactions, in terms of passing or being passed, is a
U-shaped curve with a minimum at the median speed. The increased risk of crash
involvement was a result of potential conflicts created when a faster vehicle
passes a slower vehicle. The relative overtaking rates for a vehicle as a
function of deviation from mean speed on a 100-kph road is shown in Figure 9.
Harkey, Robertson
and Davis (1989) studied the relationship between speed and accidents on non-55
mph urban roads in
Coffman, Stuster
and Warren (1998) conducted a literature review of all American and
international research to analyze the relationship between speed and accidents.
It was concluded that the crash risk is lowest near the average speed of
traffic and increases for vehicles traveling much faster or slower than
traffic. Finch, Kompter, Lockwood and Maycook (1994) collected international speed
and accident data and performed a regression analysis to study the relationship
between speed and accidents. Their results indicated that the probability of
being involved in an accident was represented by a U-shaped curve as a function
of speed.
2.3.5 Effects of Speed on Crash Severity
The research
literature presents a clear relationship between vehicle speed and the severity
of injury resulting from a crash, when a crash does occur. In a crash, the basic
physics of motion explains this relationship. A vehicle occupant continues in
motion at the pre-crash speed for a short time after impact, until collision
with another surface within or outside the vehicle occurs and completely halts
the motion of the person (Evans, 1991). Seat belts and airbags provide some
protection; however, greater vehicular speed upon impact usually results in
faster motion of an occupant into the vehicle surroundings and a higher chance
of serious injury or death. The
relationship between travel speed and the severity of injuries sustained in a
crash was examined more than 40 years ago by Solomon (1964) who reported an
increase in crash severity with increasing vehicle speeds on rural roads. After
analyzing 10,000 crashes, Solomon observed that crash severity increased
rapidly at speeds in excess of 60 mph, and that the probability of fatal
injuries increased sharply above 70 mph.
The
impact of vehicle speed on the severity of an accident has been significantly
affected by the improvements in automobile and truck crash worthiness. Passive systems, such as seat belts and air
bags, have decreased the severity of highway accidents. Increasingly, active
safety systems, such as lane departure, collision avoidance, and vehicle
stability systems are improving highway safety for both automobiles and heavy
trucks. The improvements in crash worthiness over time have, to some extent, made
the direct relationship between speed and crash severity more difficult to
interpret.
2.3.6 International Studies of the Safety Impact
of Speed Limits
There has been a significant
amount of international research conducted on the issue of the impact of speed
limits on accidents and fatalities. However, as demonstrated by the wide
disparity in rural speed limits in different countries, there is currently no
consensus on the relationship between speed limits and safety. Table 2
summarizes the maximum speed limit in different countries and the accident and
fatality rates in those countries (Source: International Road Traffic and Accident Database, 2004).
|
|
Fatalities per 100,000 pop. |
Injury accidents |
Fatalities per 100 million vehicle
km |
|
Probability of fatality |
|||
|
Country |
Total |
per 100,000 pop. |
per 100 million vehicle km |
All roads |
Motor-ways |
Speed |
Based on VMT |
Based on pop. |
|
|
8.8 |
|
|
0.9 |
|
110 |
|
|
|
|
11.9 |
537 |
55 |
1.23 |
0.72 |
130 |
1.31 |
2.22 |
|
|
14.5 |
462 |
52 |
1.63 |
0.62 |
120 |
1.19 |
3.14 |
|
|
8.9 |
496 |
50 |
0.9 |
0 |
113 |
0.00 |
1.79 |
|
Czech Rep. |
14 |
260 |
62 |
3.31 |
1.22 |
110 |
1.97 |
5.38 |
|
|
8.6 |
133 |
15 |
0.92 |
0.49 |
110 |
3.27 |
6.47 |
|
|
8 |
119 |
13 |
0.85 |
0.41 |
120 |
3.15 |
6.72 |
|
|
12.9 |
178 |
19 |
1.36 |
0.45 |
130 |
2.37 |
7.25 |
|
|
8.3 |
439 |
59 |
1.11 |
0.41 |
130 |
0.69 |
1.89 |
|
|
19.3 |
218 |
30 |
2.67 |
0 |
100 |
0.00 |
8.85 |
|
|
14 |
193 |
|
0 |
1 |
120 |
|
7.25 |
|
|
10.1 |
301 |
41 |
1.6 |
0 |
70 |
0.00 |
3.36 |
|
|
9.6 |
169 |
18 |
1.09 |
0.74 |
89 |
4.11 |
5.68 |
|
|
11.1 |
366 |
|
0 |
0.99 |
130 |
|
3.03 |
|
|
7.5 |
735 |
120 |
1.27 |
0.46 |
100 |
0.38 |
1.02 |
|
Luxemburg |
14 |
174 |
|
0 |
0 |
120 |
|
8.05 |
|
|
6.1 |
208 |
30 |
0.85 |
0.17 |
120 |
0.57 |
2.93 |
|
|
10.3 |
258 |
21 |
1.24 |
0 |
100 |
0.00 |
3.99 |
|
|
6.9 |
192 |
25 |
0.83 |
0 |
90 |
0.00 |
3.59 |
|
|
15.3 |
140 |
|
0 |
0 |
110 |
|
10.93 |
|
|
21 |
505 |
|
0 |
1.51 |
120 |
|
4.16 |
|
|
14.9 |
485 |
74 |
2.28 |
0 |
100 |
0.00 |
3.07 |
|
Slovak Rep. |
11.3 |
146 |
59 |
4.69 |
0 |
130 |
0.00 |
7.74 |
|
|
13.7 |
523 |
83 |
2.17 |
0.99 |
130 |
1.19 |
2.62 |
|
|
13.2 |
244 |
|
0 |
0 |
120 |
|
5.41 |
|
|
6 |
178 |
23 |
0.83 |
0.25 |
110 |
1.09 |
3.37 |
|
|
7.1 |
326 |
39 |
0.84 |
0.37 |
120 |
0.95 |
2.18 |
|
|
5.6 |
80 |
105 |
7.3 |
5.01 |
90 |
4.77 |
7.00 |
|
|
6.1 |
386 |
52 |
0.75 |
0.21 |
113 |
0.40 |
1.58 |
|
|
14.9 |
682 |
46 |
0.94 |
0.52 |
113 |
1.13 |
2.18 |
Nilsson (1977)
studied the impact of having different speed limits on different highways
within the same highway category in
Another study by Nilsson
(1990) analyzed the impact of a reduction in speed limits from 110 kph to 90
kph (68.35, 55.93 mph) on motorways in the summer of 1989 in
Cameron, Newstead
and Vulcan (1994) conducted a study in
In 2003, Cameron
performed a total cost benefit analysis of the impact of increasing or decreasing
speed limits on the overall economic costs. The author concluded that if the
speed limits were raised to 130 kph (80.78 mph) from the speed limit of 110 kph
for automobiles and 100 kph for trucks, the vehicle operating costs would
increase by 7.2% and the crash costs would increase by 89.4%. Whereas the time
savings, due to higher speed limits, would decrease the time cost for the
public by 16.9%. Overall, the total economic cost was estimated to increase by
2.2%, from $288.8 million to $295.25 million. It was also observed that having
a uniform speed limit of 110 kph for automobiles and trucks could reduce the
overall cost. However, the optimum speed differed substantially by vehicle type
and it was estimated that a speed limit of 120 kph (74.57 mph) for automobiles
and 95 kph (59.03 mph) for trucks would minimize the economic costs.
Fieldwick (1987) conducted
a global study to estimate the effect of speed limits on road casualties using
1984 accident data. The data collected from 20 European countries and the
Elvik and Vaa
(2004) analyzed the results of many studies conducted worldwide to assess the
impact of changes in speed limits on the number of accidents and on the average
traffic speed. Based on a meta-analysis, it was concluded that increases in the
speed limits from levels less than or equal to 90 kph (55.93 mph) to levels
above 90 kph were associated with increases in the number of accidents for all
levels of severity. The fatal accidents increased by 21% while the injury and
property damage accidents increased by 17% and 16%, respectively. The increase corresponded
to an average increase of 17.4 kph (10.81 mph) above 90 kph, which resulted in
an increase of the mean traffic speed of 4.9 kph (3.04 mph). The reduction in
speed limits from the range of 115-110 kph (71.46-68.35 mph) to the range of
97-88 kph (60.28-54.68 mph) was associated with a 54% reduction in the number
of fatal accidents and a 6% reduction of injury accidents.
Donald (1998)
investigated the possible impact of increasing the speed limits on rural roads
in
Sliogeris (1992)
conducted a study to analyze the impact of imposition and removal of 110-kph
speed limits in
Many speed limit
experiments were conducted from 1962 to 1978 in
Even though there
are no mandatory speed limits on the Autobahn in
Johansson (1996)
studied the reduction of speed limits from 110 kph to 90 kph (68.35 to 62.14
mph) on Swedish motorways and other major highways. Monthly automobile accident
data of these affected highways were collected and a Poisson time series
analysis was used to determine the effect of reduced speed limits on
fatalities, injuries, and vehicle damage. Ninety months of “before” data and 30
months of “after” data were used in the analysis. The results indicated no
statistically significant effect on fatal or injury crashes, although the minor
injury and vehicle damage crashes were reduced significantly.
Coesel and Rietveld
(1998) investigated the social costs and benefits of reducing the highway speed
limits in the
With the establishment of the 55 mph National Maximum Speed Limit in
1974, the primary aim of the new rule was partially achieved by reducing the
fuel consumption by approximately 2.9%. This is partially due to the reduced
speeds and partially due to a reduction in distances traveled by motorists.
Before the National Maximum Speed Limit, most of the states had a 70 mph or
higher speed limits. Four states had a 60 mph speed limit, 5 states had a 65
mph speed limit, 30 states had a 70 mph speed limit, 9 states had a 75 mph
speed limit, and 2 states (Montana and Nevada) had no mandated speed limits
(see Appendix A for details). In 1987 congress enacted legislation allowing states to
increase speed limits on rural interstate highways from 55 mph to 65 mph (P.L.
100-17; P.L. 100-202). By the end of that year, 38 states had raised their
speed limits with and two additional states following in 1988 (see Appendix B for details). Of the forty states that raised their
limits, ten set differential speed limits for automobiles and heavy trucks. The
National Highway Designation Act of 1995 repealed the national maximum speed
limit and returned authority to the states to set speed limits. Twenty-nine
states increased their speed limit for automobiles to above 65 mph. As of 2004,
there were 11 states that had differential speed limits between automobiles and
trucks. Figure 10 and Figure 11 below
indicate the maximum speed limits for automobiles and trucks, respectively.
Figure 10 illustrates that most of the north eastern states have a 65 mph
maximum posted speed limit for automobiles, while many of the states in the
Due to the changes
in federal speed limit policies over the last 40 years, there has been an
abundance data available for studies of the impact of increasing highway speed
limits in the
65 mph 75 mph 70 mph
75 mph 60 mph 65 mph 55 mph 70 mph
Due to the changes
in federal speed limit policies over the last 40 years, there has been an
abundance of data available for studies of the impact of increasing highway
speed limits in the
Most of the research literature that
investigated the 1987 increases in speed limits concluded that the increased
speed limit from 55 mph to 65 mph on rural interstates led to an increase in
fatalities. However, the studies that found the largest effects frequently
analyzed only the number of fatalities and did not consider the effect of
vehicle miles traveled (i.e., fatality rate). Figure 12 summarizes some of the major studies
that analyzed the impact of the 1987 interstate highway speed limit increase on
the number of fatalities. These and
others will be discussed in detail in later sections.
Figure 13
summarizes the results obtained by studies that investigated the impact of the 1987
speed limit increase on safety at the individual state level. The label, “no
effect,” indicates that the particular study concluded that the increase in
rural interstate speed limits did not have a statistically significant impact
on fatalities. Figures 14 and 15 summarize some of the major studies that
analyzed the impact of the 1995 speed limit increase on safety at the national
and state levels, respectively.
From the four
graphs, is can be seen that the majority of the studies that found a difference
observed a negative effect of increased speed limits on the affected highways.
Among these studies, some concluded that increased interstate speed limits have
positive effects on highway safety, when observed statewide. In addition, the
many
1995 Increase in Speed Limit
methodological issues that were previously
mentioned make the interpretation of some of these results difficult. These
issues will be discussed in detail in the following sections.
When the National Maximum Speed Limit of 55
mph was established in 1974, the average vehicle speed dropped by 7.4 mph (65
mph to 57.6 mph); However, non-compliance with the new law was widespread
(Meier and Morgan, 1981). Meier and
Morgan analyzed the national fatality data from 1950 to 1980 and developed a
regression equation that linked fatalities with average vehicle speed.
According to the model, for every one mph increase in average speed, it was estimated
that an additional 1,206 people would be killed in traffic accidents, all other
things being equal. They disagreed with the opinion of other researchers that the
reduction in vehicle miles traveled and safety improvements could have been the
primary factors that led to the significant reduction in fatalities rather than
the decrease in speed limits. After conducting a regression analysis of traffic
fatalities including both vehicle miles traveled and the average vehicle speed,
they concluded that vehicle miles traveled did have a statistically reliable
effect on fatalities; although the average speed had a much more significant
impact. However, the “significance level” of the statistic used in this study does
not represent the relative impact of the two variables on the number of
fatalities. The authors also argued that the standardized regression
coefficients for miles traveled (0.57) was much lower than that of average
speed (0.89). This comparison would only make sense if the basic units being
compared were the same, which they were not.
Cerelli (1981) also analyzed the national
fatality data and estimated that increasing the national speed limit from 55 to
60 mph would result in an increase of 3,500 fatalities per year. A study by the
National Research Council (1984) found that lower speed limits had a
significant positive effect on fatalities. It was reasoned that the reduction
in vehicle miles traveled, improvement in vehicle safety, and improvement in
roadway and medical services could not explain all of the reductions in
fatalities, and claimed that lower and more uniform speeds were responsible for
saving some 3,000 to 5,000 lives in 1974.
Godwin and Kulash (1988) indicated that
highway travel declined by 1.5% between 1973 and 1974, and long-term
improvements in the rate of fatalities per mile driven averaged approximately
3%. The sudden drop in the fatality rate in 1974, which was measured to be approximately
15%, was still more than three times the combined effect of the two factors: 1)
decline in travel and 2) improvement in the fatality rate. Further, the
greatest decline in fatality rates occured on those roads where the speed limit
reduction was largest.
2.3.7.2 Impact of the 1987 Speed Limit Increase
The review of the studies
conducted to estimate the safety implications of the speed limit increase in
1987 from 55 to 65 is divided into two major categories: (1) studies conducted using
data from multiple states (mostly national level studies) and (2) studies
conducted at the individual state level. These two categories will be reviewed
separately. Within each category, the reviews will follow the pattern: the
studies that found a negative effect of increased speed limits are discussed first
and then the studies that found no effect or a positive effect are discussed.
The National Highway Traffic Safety
Administration (1992) estimated that in 1990, the 38 states that had increased
the speed limits to 65 mph in 1987 experienced a 30% increase in the number of fatalities
on rural interstate highways than what would have been expected if the limits
had not been raised. However, the study
concluded that, even though the number of fatalities had increased on rural
interstates with the implementation of the 65 mph speed limit, the interstates remained
the safest component of the national highway system. The fatality rate on rural
interstates was 1.3 fatalities per 100 million vehicle miles traveled in 1990
compared to 2.1 fatalities per 100 million vehicle miles traveled for the
nation as a whole.
A study conducted
by Advocates for Highway and Auto Safety (1995) agreed that the increase in
speed limits in 1987 caused 30% more fatalities on rural interstates among states that increased their speed limits. This
same study estimated that if the National Maximum Speed Limit was repealed, the
highway fatalities would increase by 6,400 every year at a cost of an
additional $19.3 billion every year. It is important to note that, after the
repeal of National Maximum Speed Limit, the annual fatalities increased by only
248, from 41,817 deaths in 1995 to 42,065 in 1996, which was 96% below the
projected values by Advocates for Highway and Auto Safety.
Balkin and Ord
(2001) studied the national fatality data from 1975 to 1988 to estimate the
effect of increasing speed limits. They found that 19 of the 40 states that
increased their speed limit in 1987 experienced a significant increase in fatal
crashes on their rural interstate highways; however, the exact impact of the speed
limit increase on highway fatalities was not provided. The lack of an impact of
increased speed limits on fatal crashes in 21 of 40 states weakens the argument
made by the authors that the increased speed limits had a significant negative
impact on highway safety.
The National
Highway Traffic Safety Administration (1989) analyzed accident data from 1975
to 1987 and found that the 38 states that increased their speed limits in 1987
experienced 16% more rural interstate fatalities than expected. However, note
that a large percentage (64%) of this increase resulted from only six states.
The actual increase in the number of fatalities on rural interstates was 19% for
the states that raised limits, and 7% for the states that retained lower speed
limit.
Baum, Lund and
Wells (1989) analyzed the fatality data from 1982 to 1987 for the 38 states that
increased their speed limits in 1987, and found that the rural interstate
fatalities in 1987 were 15% (confidence interval of 65 to 24%) more than
expected; whereas, the rural interstate fatalities for states that did not
increase their speed limits decreased by 6% (confidence interval of -23 to -13%)
in the same period. Baum,
Garber and Gadiraju
(1989) collected fatality data from 1976 to 1988 and performed separate time
series analyses for each of the 40 states that enacted a 65 mph speed limit.
The authors concluded that the fatalities increased by a median value of 15% on
rural interstates and 5% on non-rural interstates. However, it should be noted
that the increase in fatalities wase not uniform across all the states. Out of the 40 states, 28 experienced an
increase in the number of fatalities and 12 experienced a decrease in fatalities.
One of the differences between the Baum,
et al. study and the Garber and Gadiraju study is that the latter study used a
longer period of time. As previously discussed, the time frame can have a large
effect on the results observed.
According to McCarthy
(1994a) there are three main highway safety consequences from an increase in
rural interstate speed limits: (a) direct effect, (b) traffic diversion affect,
and (c) spillover effect. Since the direct and the traffic diversion effects
are likely to operate in the opposite direction of the spillover effects, the
overall impact on highway safety remains ambiguous, with a possible bias
towards improved highway safety. This bias reflects the induced shift of
traffic away from the most dangerous, rural non-interstate, roads toward rural
interstate highways that have traditionally been safer. McCarthy studied these
effects using
McKnight and Klein
(1990) studied the nationwide impact of increasing the speed limits to 65 mph on
rural interstate highways. Speed and accident data were collected from 1982 to
1988 for all 50 states. In the states that raised their speed limits to 65 mph,
the number of fatal accidents on rural interstates increased by 22% over
projections based on previous trends. There was no change in the number of fatal
accidents on non-rural interstates. In states that maintained the 55 mph limit on
rural interstates, there were significant 10% and 13% increases in fatal
accidents on rural interstates and non-rural interstates, respectively. States that
increased their speed limit to 65 mph did not experience an increase in fatal
accidents on non-rural interstates.
McKnight, Klein and
Tippetts (1989) collected nationwide data from 1983 to 1988 and found that, in
states that raised their speed limits to 65 mph, the fatal accidents on rural
interstates increased by 22% over projections based on the previous trends. There
was no significant increase in fatal accidents in the same states for non-interstate
highways that did not experience an increase in speed limit. The states that
did not increase their speed limit observed only a 10% increase in fatal
accidents (i.e., approximately 20 more fatal accidents). The non-rural
interstate highways from these 55 mph states observed a significant 12.7%
increase in fatal accidents (i.e., an increase of 295 fatal accidents)
indicating a “traffic diversion” effect.
Godwin (1992) analyzed national data from
1986 to 1988 and found that the fatalities on highways on which the speed limit
was increased to 65 mph were 15 - 25% higher than expected in 1988.
Furthermore, the fatality rates for rural interstate highways increased by 18% (1.4 to 1.7 per million vehicle
miles traveled) for 65 mph states, while the fatality rates for non-rural
interstate highways in the same states were 7% lower (2.7 to 2.5 per million vehicle miles
traveled). The fatality rates for rural interstates and non-rural interstates
for the 55 mph states remained the same. The traffic diversion phenomenon was
noted in that, in that the fatalities and fatality rates went up on rural interstates
for the 65 mph speed limit states, the fatality rates went down for non-rural
interstates in the 65 mph speed limit states.
Some researchers have found that an
increase or decrease of the speed limit makes only an initial negative impact
on safety which later decays. Chang, Carter and Chen (1993) concluded that the
increased speed limit had a significant “initial” impact on highway fatalities
at the nationwide level; however, the impact decayed after approximately a year
of “transition” period.
Wilmot and Khanal
(1999) surveyed the literature and came to the conclusion that speed affects
the severity of accidents but not the probability of accidents on rural
interstates. The statewide fatalities can be reduced by having higher speed
limits on rural interstate (which are the safest roads in the system, having the
highest design speeds) while maintaining lower speed limits on other more
dangerous highways that have lower design speeds.
Lave
and Elias (1994) found that during the period of 1986 to 1988, the statewide
fatality rates in states that increased their limits to 65 mph, decreased more
than the states that retained the limit of 55 mph. The actual reduction in
fatality rates observed in the 65 mph states was 6.15%; whereas, the 55 mph
speed limit states had a reduction of 2.62%. The authors suggested that the three
main reasons for this decrease were (1) state highway patrols were allowed to
shift resources from speed enforcement on the interstates to other safety
activities and other highways, (2) higher speed limits attracted faster drivers
away from other, more dangerous roads (non-interstates), and (3) speed variance
among vehicles might have declined. Lave and Elias also observed a significant
increase in vehicle miles traveled in states that had increased their speed
limits.
The
following studies were conducted at the state level to investigate the effect of
the 1987 speed limit increases. Bamfield (1989) studied
McCarthy
(1994b) also analyzed accident data from
Wagenaar, Streff, and Schultz (1990)
collected data from 1978 through 1988 to evaluate the effect of the 1987 speed
limit increase in
|
Year |
Fatalities |
|
1984 |
1,531 |
|
1985 |
1,545 |
|
1986 |
1,605 |
|
1987 |
1,602 |
|
1988 |
1,708 |
|
1989 |
1,639 |
|
1990 |
1,571 |
Table 3 shows that the
number of fatalities increased by 106 from 1987 to 1988; however, in the next
year, which this study did not consider, the fatalities decreased and, by 1990, the number of fatalities was
below the level of the pre-increase years. This confirms what Chang, Carter and
Chen (1993) suggested that highway fatalities increase in the initial
“transition” period, but after the drivers
“adapt” to the higher speeds, the number of fatalities decreases.
Ledolter and Chan
(1996) studied the impact of the 1987 speed limit increase from 55 mph to 65
mph in
|
Year |
Fatality Rate |
|
1986 |
2.16 |
|
1987 |
2.36 |
|
1988 |
2.54 |
|
1989 |
2.28 |
|
1990 |
2.02 |
Brackett and Ball
(1990) studied a speed limit increase from 55 mph to 65 mph in
Pant, Adhami and
Niehaus (1991) studied the impact of the 1987 speed limit increase from 55 mph
to 65 mph in
Brown, Maghsoodloo
and McArdle (1990) studied the effect of the speed limit increase in
|
Year |
Fatalities |
Fatality Rate |
|
1986 |
1,081 |
3.18 |
|
1987 |
1,111 |
2.97 |
|
1988 |
1,024 |
2.58 |
|
1989 |
1,029 |
2.52 |
Rock (1995) studied the effect of the 1987
speed limit increase (from 55 to 65 mph) in
A Virginia study
was conducted by Jernigan, Lynn and Garber (1988) to investigate the issues
related to increasing speed limits on rural interstates to 65 mph. They
estimated that the increased speed limits would increase the traffic speed from
60 mph to 63 mph resulting in an annual increase of 6 to 18 fatalities and 171
to 405 injuries. However, increased speeds would reduced travel time by up to
1.3 million hours. The authors concluded that the economic benefits of raising
speed limit to 65 mph might outweigh the cost by a minimum of $3.8 million.
This study was
followed by another
Agent, Pigman and
Webber (1998) conducted a study to evaluate the speed limits in
Khan and Sinha
(2000) studied the impact of increasing speed limits in
Upchurch (1989) in
|
Year |
Fatalities |
Fatality Rate |
|
1983 |
675 |
3.28 |
|
1984 |
869 |
4.15 |
|
1985 |
893 |
4.14 |
|
1986 |
1,007 |
4.44 |
|
1987 |
939 |
2.96 |
|
1988 |
944 |
2.76 |
|
1989 |
879 |
2.52 |
McCarthy (1988)
studied the effects of increased speed limits in
2.3.7.3 Impact of the 1995 Speed Limit Increase
Following December
1995 repeal of the National Maximum Speed Limit, many states raised the maximum
interstate speed limit to 70 or 75 mph. Twenty-nine states increased their
speed limit for automobiles to speeds above 65 mph (See Appendix C for details). There have been a number of studies that
have estimated the impact of these increased speeds on the number of crashes
and fatalities. As in the previous section, the following review will be
divided into two major categories: (1) studies conducted using data from multiple
states (mostly national level studies), and (2) studies conducted at the individual
state level. Within each of the two categories, the review of studies will
first present the studies that found a negative effect of increased speed
limits, followed by the studies that found no effect or a positive effect.
Balkin and Ord
(2001) studied the national fatality data from 1975 to 1998 to estimate the
effect of higher speed limits. Ten of the 36 states that increased their rural
interstate speed limits in 1995 experienced a significant increase in fatal
crashes on those highways. However, data pertaining to the exact impact of the speed
limit increase on highway fatalities was not provided. The lack of an impact resulting
from increased speed limits on fatal crashes in 21 of 40 states, weakens the
argument made by the authors that the increased speed limits have a negative
impact on highway safety.
Farmer, Richard,
and Lund, (1997) collected fatality rates and data on the number of fatal crashes
from 1990 through 1997 for 12 states that increased their speed limits above 65
mph (study group) and 18 states that retained their 65 mph maximum speed limit
(comparison group). It was estimated that there was a 12% increase in
fatalities and a 17% increase in fatality rates on interstate highways and
freeways for the 12 states that increased their speed limits. There was also a significant
increase (p=.06) in fatalities and fatality rates on other roads associated
with speed limit increases, indicating the potential of a spillover effect. Considering
only the rural interstates, there was an 11% increase in fatalities due to
higher speed limits. One of the limitations of this research is that it was limited
to data from only 1995 and 1996. The year, 1996, was a transition year for most
states that had increased their speed limits.
Moore (1999)
observed that the results obtained by Farmer, Richard, and Lund would not be
consistent with an analyses that used the data for the following year because 8
out of the 12 states from the test group (states that increased their speed
limit after 1995) experienced a drop in fatality rates in 1997. Many states
that either increased or maintained their speed limits were omitted from the
study. Another fact that is problematic for the interpretation of the results
of this study was that the fatality rate increased more than the fatality count
for the test group. This would only be possible if the vehicle miles traveled
had decreased, which was not the case.
The National
Highway Traffic Safety Administration (1998) examined the effects of increased
speed limits above 65 mph. Based on the fatality data from 1991 to 1995, NHTSA
predicted the number of fatalities in 1996. When the predicted number of
fatalities was compared to the actual numbers, it was observed that the group
of states that increased their speed limit above 65 mph experienced a 9%
increase (350 more fatalities) in fatalities on rural interstate highways in
1996 than were predicted by the model.
This study did not appear to account for changes in traffic volume.
A national level
study conducted by Patterson, Frith, Povey and Keall (2002) modeled changes in
rural interstate fatalities considering the changes in speed limit along with
fatality count data and vehicle miles traveled from 1992 to 1999. Compared to
the states that did not raise their limits, it was estimated that there was an increase
of 35% in rural interstate fatalities (confidence interval of 6% to 72%) for the
states that raised their speed limits to 70 mph. There was a 38% increase in rural
interstate fatalities (confidence interval of 8% to 78%) for the states that
raised their speed limits to 75 mph. Again, these estimates were based on prediction
models and,even though there was no significant increase in statewide fatality
rates in states that increased their speed limits, there was a significant
decrease in statewide fatality rates (19% decrease) in states that retained the
65 mph speed limit.
Srinivasan (2002)
reviewed the literature on the research work done on examining the impact of
increased speed limits and concluded that the increased speed limits in 1995
had increased the probability of fatal accidents, although the impact of speed
on total accidents and speed dispersion was unclear.
Langlotz (1999)
compared the changes in overall fatality rates following the 1995 speed limit
increase between two groups: states that raised speed limits and a control
group that did not. Langlotz compared the fatality rates of 1995 with the
fatality rates of 1997. The fatality rates were studied in addition to fatality
counts to account for the increase in miles traveled on highways. The state-wide
effect was taken into consideration as well as the effect on interstates in
order to address the potential traffic diversion from local roads to faster
interstate highways. The fatality rates in states that raised their limit decreased
by 5.00%; whereas, while the states that did not increase their limits
experienced a fatality rate decrease of 5.38%.
The difference between groups of 0.38% was statistically insignificant.
Of the 33 states that raised limits, 10 experienced an increase in fatality
rate, and 23 experienced a decrease in fatality rate. Of the 15 states in the control
group, five experienced an increase in fatality rate and 10 experienced a
decrease. No significant change in statewide fatalities was observed for either
the test group or the control group.
Lave (1997) found
that, after the 1995 increase in speed limits, fatalities did not increase
nationally by the 10% to 14%, as expected by the opponents of higher speed
limits. Instead, it decreased by 0.7% during a time when the vehicle miles
traveled increased by 1.8%. However, the
data available from NHTSA indicates that national fatalities had increased by
0.6% (from 41,817 to 42,065) between 1995 and 1996, and not decreased by 0.7%,
as claimed by Lave. Although Congress gave permission to raise speed limits in
November 1995, it took some states a period of time to adopt the new
legislation. Only half of those that implemented the changes had done so by May
1996. To understand the effect of the 1995 speed limit increase, it is more
meaningful to analyze changes in fatality data between 1996 and 1997, rather
than comparing 1995 and 1996. From the NHTSA database, fatalities between 1996
and 1997 decreased by 0.12% (from 42,065 to 42,013).
After analyzing the
fatality and accident data from all of the states,
A number of studies
were conducted to examine the safety implications of increased speed limits for
individual states. Those studies will be discussed next.
Banasiak (1997)
observed that the interstate traffic speeds increased after the 1996 speed
limit increase from 65 mph to 75 mph.
However, the number of fatalities decreased during the seven month period
following the speed limit increase. Fatalities in those seven months decreased
from 44 to 40 fatalities, compared to the same seven months in 1995.
Najjar, Stokes,
Russell, Ali and Zhang (2000) studied the results of the 1996 change in the maximum
speed limit from 65 mph to 70 mph on rural interstates in
A
Dornsife (2001)
studied the impact of changes in speed limits on accidents in
The Iowa Highway
Safety Management System Task Force (2002) studied the impact of increased
speed limits from 55 mph to 65 mph on rural expressways and freeways in
Raju, Souleyrette
and Maze (1998) studied the accident data from 1980 to 1996 in Iowa to estimate
the safety effect of an increase in rural interstates speed limit from 55 to 65
mph. Increased speed limits were found to be increasing the annual fatal
accidents on rural interstates by approximately 60% on rural highways (16 fatal
accidents more per year).
Bartle, Baldwin,
Johnston and King (2003) examined the increase in the number of fatalities on
Some studies investigated
the relationship between 85th percentile speed, design speed and
highway safety. Parker (1992) collected speed and accident data from 22 states
at 100 sites (non-limited access rural and urban highways) before and after the
speed limits were altered. “Before” and “after” data were also collected at
comparison sites where speed limits were not changed to control for the time
effects. It was found that the average posted speed limits were set at the 45th
percentile speed or below the average speed of traffic. Average speed average
limits were posted between 5 and 16 mph below the 85th percentile speed. Raising
the speed limits in the region of the 85th percentile speed had a
beneficial effect on drivers complying with the posted speed limit. This
results in a more uniform traffic flow, thus reducing speed variance and
improving highway safety. At the 58 experimental sites where speed limits were
lowered, accidents increased by 5.4%, although this was not statistically significant.
Accidents at the 41 experimental sites where speed limits were raised decreased
by a non-significant value of 6.7%. This could be explained by the fact that
after increasing the speed limit, the new posted speed would have become closer
to the design speed of the highway, and also closer to the 85th
percentile speed of the traffic, thus decreasing the accidents on these
highways.
Baxter (1999) expressed
the opinion that considering design speed is critical to evaluating the
relationship between speed and safety. The assertion is that accidents increase
only if speed increases beyond the design speed of the highway, and that if the
posted speed remains within the design speed of the highway, there will not be
a significant increase in accidents.
Most of the
national level studies conducted after the 1987 speed limit increase to 65 mph observed
an increase in the number of crash fatalities in the range of 10% to 20%.
However, the results of the studies conducted to estimate the impact of the 1995
increase in speed limits (from 65 mph to 70 or 75 mph) on crash fatalities have
varied from no effect to a 55% increase. After reviewing the studies, the
reasons for the difference in results obtained could be broadly classified into
two categories: (1) random fluctuation in the number of accidents, which
depends on many factors, including driving behavior, traffic conditions,
geographic conditions, weather, economic issues, enforcement etc. and (2)
research methodological issues, including the selection of different time
frames, etc. The increase in speed limits has been observed to result in very
different effects in individual states. The increase in speed limits appeared
to increase highway fatalities in some states, reduced them in others, and had no
detectable effect in the remainder. The global median effect was approximately
a 10% to 15% increase in fatalities. However, increases in speed limits were
also observed to be associated with increases in the vehicle miles traveled. Many
of the studies did not take into consideration vehicle miles traveled. When the increase in the miles traveled was considered,
the effects were much less pronounced or did not occur. Many of the studies
illustrated that increases in speed limits on rural interstates result in
traffic diversion, resulting in fewer fatalities on “less safe” highways, which
compensates for the increase in fatalities on interstates.
2.3.8 Effects
of Differential Speed Limits on Safety
The issue of
setting uniform or differential speed limits for automobiles and trucks has also
been controversial, particularly in recent years. The proponents of
differential speed limits contend that trucks have significantly different
operating characteristics than automobiles in terms of performance,
maneuverability, and braking and should, therefore, operate at lower speeds. The
opponents of differential speed limits on rural interstates contest the idea that
lower speeds for trucks improves safety because the amount of variation in vehicle
speed increases the probability of accidents.
The difference in
operational characteristics of automobiles and trucks will be reviewed in
detail in the next section. In this section, the impact of differential speed
limits on highway safety will be discussed. There have been many studies
conducted on this topic; however, the studies have been unable to provide consistent
information as to how differential speed limits affect safety. During the
period when the 55 mph National Maximum Speed Limit was in effect (from 1973 to 1987), there was no difference
in the posted limits for trucks and automobiles. Therefore, the studies on
differential speed limits were either conducted before 1973 or after 1987. In 1987, the states were faced with the
question of whether to set speed limits for all vehicles or to set differential
speed limits for automobiles and heavy trucks. Out of the 40 states that
increased their speed limits in 1987, ten states set differential speed limits
for automobiles and larger vehicles. Seven states adopted a 65/55 mph differential
speed limit, and three states had 65/60 mph differential speed limit for automobiles
and trucks, respectively. As of June, 2005, 11 states had differential speed
limit for automobiles and trucks (Figure 15).
These speed
differentials do not only apply to heavy trucks. Some states also include buses,
towing vehicles, etc. There have been extensive discussions in many states pertaining
to whether to initiate, retain or eliminate differential speed limits. For
example, the legislature of
Efforts have been
made on the international level to reduce the truck-involved accidents. The
European Commission passed a regulation that required speed limiters on all
trucks and buses. Speed governors with a maximum speed limit of 90 kph were
made mandatory for all trucks and buses to reduce the severity and number of
trucks involved in accidents.
The effect of
differential speed limits on safety is a controversial issue. Lower speed limits
for trucks help reduce truck-rear-ending-automobile accidents and the severity
of such accidents. The Federal Motor Carrier Safety Administration’s report of 2003
stakeholder forums indicated that the opinions of the participants were divided
on the issue of speed differentials. Some participants viewed speed differential
laws to be effective, while the industry representative and many enforcement
personnel viewed them as less safe, stating that it forces trucks to become
slower moving “obstacles” on the roads.
According to
Cirillo (2003), a former assistant administrator and chief safety officer for the Federal
Motor Carrier Safety Administration, traffic operating at or about the
same speed, regardless of the speed limit, is the safest traffic environment.
The author observed that the fatality rates and accident rates on interstate
highways are 2 to 5 times less than the non-interstate highways. Adherence to
differential speed limits creates an unsafe situation in which a significant
percentage of traffic is operating much slower than general traffic. Lower
truck speeds can also entice commercial traffic to use less safe non-interstate
facilities.
Spencer (2003), executive vice president of the Owner-Operator
Independent Drivers Association, stated that having a 10 mph speed differential
between automobiles and trucks increases safety concerns on highways because it forces
vehicles to be constantly in conflict with each other. Spencer’s concerns
include the problem that lane changing and passing are constantly required to
avoid crashes, which increases probability of accidents. Differential a speed
limit increases the number of bottleneck and leapfrog situations on highways.
Yuan and Garber
(2002) studied the impact of differential speed limits by comparing the
accident data of states having uniform speed limits (USL) with those having
differential speed limits (DSL). Speed and crash data during the 1990’s were
taken from four types of states. The four groups were: (a)
states that retained uniform speed limits, (b)
states that retained differential speed limits during 1990’s, (c) states that changed their speed
limits from uniform speed limits to differential speed limits during 1990’s,
and (d) states that changed their
speed limits from differential speed limits to uniform speed limits during
1990’s. The states that retained uniform speed limits experienced increases in
total crash rates and rear-end crash rates.
All of the groups experienced increases in the total number of truck-involved
crash rates. No significant increase was observed in truck-involved rear end
crash rate for the third group (uniform to differential limits). States that
changed from differential to uniform limits experienced an increase in total
crash rate.
A study by Harkey
and Mera (1994) examined the impact of differential speed limit on safety, based
on data from nine states. The states were divided into four groups based on
their speed limits: 65/65 (
|
Speed Limit |
Rear End |
Sideswipe |
Other |
|||
|
Auto-into-truck |
Truck-into-auto |
Auto-into-truck |
Truck-into-auto |
Auto-into-truck |
Truck-into-auto |
|
|
USL: 65/65 mph and 55/55 mph |
10.91 |
10.78 |
22.12 |
21.07 |
2.57 |
2.01 |
|
DSL: 65/55 mph and 65/60 mph |
13.7 |
6.86 |
21.52 |
14.96 |
2.07 |
0.99 |
A study was
conducted by Garber, Miller, Yuan and Sun (2003) to compare the safety impacts
of differential and uniform speed limits on rural interstate highways, using crash
data from six states for the period of 1991
to 1999. These states were divided into three groups based on the type of speed
limit employed: states that maintained uniform limits (Arizona, Missouri and
North Carolina), states that changed from uniform to differential limits
(Arkansas and Idaho) and one state that changed from differential to a uniform
limits (Virginia). Six types of crash rates were evaluated: total crashes,
fatal crashes, rear-end crashes, total truck-involved crashes, truck-involved
fatal crashes, and truck-involved rear end crashes. Using a before-and-after
comparison, it was observed that the crash rates increased over the ten-year
period, regardless of whether uniform or differential limits were employed.
There was no consistent trend in crash rates matching the changes in speed
limits. The authors concluded that
measurable variation within crash rates by year and by state might have
confounded the statistical tests employed.
A simulation study
was performed by Garber and Gadiraju (1990) to analyze the safety impact of
differential speed limits and the restriction of trucks in the right lane. It
was concluded that the implementation of differential speed limits, in addition
to lane restriction of trucks, increased the interactions between automobiles
and trucks and, therefore, the potential for accidents. The authors recommended
that, to reduce interactions, the best
speed strategy was a uniform 65/65 mph posted speed limit.
The simulation
study was followed by an empirical study, also by Graber and Gadiraju (1991).
In this study, three differential limit states (
Hall and Dickinson
(1974) obtained similar results when they analyzed accident data from 83 sites
in
Pfeffer, Stenzel
and Lee (1991) conducted a time series analysis to examine the safety impact of
differential speed limits in
Monsere, Newgard,
Dill, Rufolo, Wemple, Bertini and Miliken, C. (2004) examined the differential
speed limits in Oregon and concluded that except for travel time savings and
some economic development benefits, all other issues (like crashes,
enforcement, health, environment etc.) to be negatively impacted by the
proposed reduction of speed differentials fom 65/55 to 70/65 mph.
2.3.9 Cause and Impact of Truck Accidents
Even though the number of fatalities
associated with truck accidents has been fluctuating in the last few years
(Figure 16), the fatality rates have been steadily declining for the past
decade (Figure 17). In fact, the figure illustrates that the decline in the fatality
rate has been much higher in trucks compared to the decline in automobiles.
Figure 17. Fatality Rates (Automobiles vs. Trucks)
(Source:
Wislocki (2003) reported that the truck related fatalities
fell in 2002 for the fifth year in a row, dropping by 4.2% (5,111 fatalities in
2001 to 4,897 fatalities in 2002); whereas, the overall number of traffic
fatalities increased from 42,196 in 2001 to 42,815 in 2002.
From Figure 16, shows that there was a 9.6%
increase in truck accidents from 1995 to 1997. These were the two subsequent
years after the 1995 speed limit increase from 65 mph (National Highway
Designation Act). There could four possible reasons that could account for this
increase: (1) normal fluctuation in accident frequency, (2) greater number of
vehicle miles traveled, (3) higher truck speed, and (4) higher speed increase of
automobiles relative to trucks. Since most of the commercial fleets had speed
limiters on their trucks that were set at or below 65 mph, the increase in
posted speed limit beyond 65 mph would not have had a significant impact on the
average truck speed. As a result, the
speed of trucks became significantly lower than the average speed of
automobiles, thus increasing the speed variance between the two. As discussed
in a previous section, speed variance has been associated with an increase in
accidents. From the data available from the Federal Highway Administration’s website, the increase in vehicle miles traveled by
trucks, from 1995 to 1996 was just 1%. Therefore the increase in miles does not
completely account for the increased number of truck accidents. In addition,
the reduction of approximately 10% accidents, from 1997 to 2002 during which
speed limits did not change indicates that the increased speed variance between
automobiles and trucks in the initial period could have contributed to the increase
in truck accidents between 1995 and 1997.
Truck accidents are
a major concern for safety authorities because of the higher probability of involving
fatalities. According to Council,
Harkey, Nabors, Khattak and
and Mohamedshah (2003), in 1998, large trucks accounted for
7% of total miles traveled but were involved in 13% of all traffic fatalities
(5,374 out of 41,471). In these truck crashes, the automobile’s occupants were
much more likely to be killed (78% of the fatalities) or injured (76% of the
injuries) than the truck driver. It was found that an automobile driver’s
behavior was three times as likely to contribute to a fatal crash as the truck
driver’s behavior. Automobile drivers were solely responsible for 70% of fatal
crashes, compared to 16% for the truck drivers. A similar study conducted by
Carroll (2004) analyzed the interaction-critical incidents (incidents during
vehicle interactions which could possibly lead to accidents), and concluded
that 82.4% of these incidents were initiated by automobile drivers, while 17.6%
were initiated by truck drivers. Kostyniuk, Streff and Zakrajsek (2002)
conducted a study for the AAA Foundation for Traffic Safety of fatal, large
truck-passenger vehicle accidents between 1995 and 1998. The study found that,
when improper following or improper lane changes were a contributing factor in
an automobile-truck accident, the automobile driver was in error 75% of the
time and the large truck driver was in error 25% of the time.
Another reason
provided for having lower speed limits for trucks is to avoid accidents that
are caused by a loss of vehicle control at higher speeds. However, the research
indicates that most of the accidents are the result of human error and very few
are due to mechanical failure of vehicles. Treat (1977) performed a five-year
study that examined the cause of 2258 automobile accidents. Only 2.4% accidents
were caused solely due to mechanical fault and 4.7% were caused by
environmental factors. It was observed that human error was the sole factor in
57% and a contributing factor in 92.6% of the accidents. Of those accidents, 90%
involved perceptual error and only 10% response error.
Garber and Joshua (1989) studied the
characteristics of large truck crashes in
Blower and Campbell
(2002) studied the actions by automobile and truck drivers that lead to fatal
truck accidents. Truck driver fatigue was found to be responsible for 2.9% of
the fatal crashes. Truck driver action was found to be responsible for 21.8% of
the fatal crashes, while other vehicles were responsible for 59% of the fatal
crashes. The loss of control of truck was responsible for 5.8% of all fatal
crashes. Speed contributed to 2.4% of these crashes, road conditions to 1.7%
and vehicle failure to 0.7%. The number of fatigue-related accidents reported
in this study was much lower than was reported in the study by Garber and
Joshua (1989). However, when only the crashes caused by truck drivers were
considered, fatigue was found to be responsible for 13.3% of the fatal crashes.
The study by Kostyniuk,
Streff and Zakrajsek (2002) documented the frequency of unsafe driver actions
in fatal automobile-truck accident. They found that the top three factors for automobile
drivers involved in automobile-truck accidents were: (1) failure to keep within
the lane or running off the road (21%), (2) failure to yield right of way
(16%), and (3) driving too fast for road conditions or exceeding the speed
limit (12%). For truck drivers, the top three factors were: (1) failure to
yield the right of way (14%), (2) failure to keep within the lane or running
off the road (12%), and (3) driving too fast for the road conditions or
exceeding the speed limit (11%).The top three “unsafe actions” of automobile
drivers involved in automobile-automobile accidents were similar to the actions
of automobile drivers involved in an automobile-truck accident. The data
suggest that pre-crash driving actions of automobile drivers involved in fatal
crashes were not significantly affected by whether the crash involved another automobile
or a truck.
Thiriez, Radja and
Toth (2002) found that more than 70% of accidents that occurred in traffic
moving in the same direction were rear-end accidents. These were, followed by
sideswipe and forward impact crashes, which constituted 20% and 10%,
respectively.
Craft (2002) conducted
an analysis of trucks involved in fatal accidents that focused on rear-end
accidents. It was found that each year, approximately 400,000 trucks are
involved in motor vehicle crashes. Eighteen percent of the accidents involving
trucks are rear-ended crashes. Rear-end crashes can be further categorized into
automobile rear-ending a truck and truck rear-ending an automobile. Since the
operating characteristics (maneuverability and braking distance) and physical
features (weight) of automobiles and trucks are different, the two crash types
are quite different in their probability of occurrence and their severity.
Craft found that 50% more trucks rear-end automobiles than do automobiles
rear-end trucks (42,000 versus 28,000). However, there are 70% more fatal
accidents in which the automobile rear-ends the truck. Of the 271 fatal
accidents in which the truck hit the automobile, 58% took place on interstates,
while 40.5% of the 461 fatal accidents where the automobile hit the truck
occurred on interstates. The probability of a fatal accident, given that an
accident has occurred is 183% more for automobiles rear-ending trucks than for
trucks rear-ending automobiles.
Large trucks are much more likely to
be involved in fatal multiple vehicle crashes than automobiles. According to
Knipling, Waller, Peck, Pfefer, Neuman, Slack and Hardy (2004), 84% of all
crashes involving large trucks were multiple vehicle crashes, compared to 61%
for passenger vehicles. According to Craft (2002), 18% of all accidents where
the truck rear-ended an automobile involved three or more vehicles; whereas,
only 5% of the automobile rear-ending a truck accidents involved three or more
vehicles. For fatal accidents only, Craft observed that 46% of all accidents in
which a truck rear-ended an automobile involved three or more vehicles; whereas,
only 16% of the accidents in which an automobile rear-ended a truck involved three
or more vehicles.
Stuster (1999) performed an analysis of the
causes of fatal accidents and listed the 25 most frequent acts committed by
automobile drivers that can lead to accidents. The top 5 were: (1) driving
inattentively, (2) merging improperly into traffic and causing a truck to
maneuver or brake quickly, (3) failing to stop for a stop sign or light, (4)
failing to slow down in a construction zone, and (5) following too
closely.
Carroll (2004)
studied the incidents that led to 142 automobile-truck accidents. The top two
incidents attributable to automobile drivers were: (1) lane change without
sufficient gap and (2) entering the roadway without sufficient clearance. The
most frequent incident attributable to truck drivers was entering a roadway
without sufficient clearance. Posted differential speed limits and truck speed
limiters can increase the potential problems with drivers entering the roadway
at traffic speed.
To examine the
impact of differential speed limits on traffic fatalities involving trucks,
Neeley and Richardson (2004) analyzed nationwide fatality data from 1994 to
2000. They found that truck speed limits and the drunken driving laws were the
only laws that significantly reduced the fatalities in crashes involving large
trucks. The authors concluded that the difference between automobile and truck
speeds did not affect safety, and the enforcement also did not affect the
number of traffic fatalities.
Another
major factor that has been associated with truck accidents is speeding.
Speeding is the act of exceeding the posted speed
limit or driving too fast for existing conditions. The accident databases do
not differentiate between these two scenarios. When the percentage of truck
accidents caused by “speeding” trucks is reported, care must be taken to
understand that most of the trucks were not traveling at a rate above the
posted speed limits. Reviewers of the literature that discuss the frequency of
“speeding” related accidents and drawing conclusions about the posted limits
may be drawing invalid conclusions. It should also be noted that the definition
of “traveling too fast for conditions” is not clearly defined. Therefore,
studies that classify accidents using this category may report different
results based on their definition. In some of the reported research, “speeding”
is defined as the act of exceeding a certain speed (i.e., 65 or 70 or 75 mph)
which may not be the posted speed limit.
Bowie and Marie
(1994) analyzed the nationwide accident data and found that speeding was
involved in 12% of all police-reported crashes and 33% of all fatal crashes.
Speed was found to affect the single-vehicle accidents most, as up to 40% of
single vehicle accidents were due to high speed. According
to the National Highway Traffic Safety Administration (2004), speeding was a
contributing factor in 31% of all fatal crashes in 2003. Speeding and driving
while intoxicated (DWI) frequently occur together. In 2003, 28% of the
drivers who were involved in fatal crashes when speeding were driving under the
influence of alcohol. Advocates for Highway and Auto Safety (1995) reported
that speeding was a factor in 33% of all fatal crashes, and it was also reported
that 56% of the drivers in speed related fatal crashes were under the influence
of alcohol.
According
to Gruberg (1999), 22% of the accidents involving trucks in multi-vehicle fatal
accidents involved speeding by at least one of the drivers. Truck drivers were
found to be speeding in 6.7% of the occurrences compared to 14.9% for automobile
drivers. Speeding related multi-vehicle crashes most frequently result in
rear-end collision (34%), followed by head-on (27%), angle (25%) and run-off-road
(9%). According to Garber and Joshua (1989), who studied the characteristics of
large truck crashes (both single and multiple vehicle), 21% of the fatal
crashes caused by truck driver-related factors were associated with speeding.
Although there are
different opinions as to the effect of speed, as opposed to “speeding,” there
is a consensus as to the physics involved. Faster vehicles have less time to
respond (in seconds and distance) and the severity of accidents that occur at
higher speeds is greater.
2.4 Effect of Speed on Driver Fatigue
A significant amount of research literature has addressed the
relationship between fatigue and accidents. Driver fatigue can be categorized
in two main types: (1) physical and mental fatigue caused by physical and
mental stress and, (2) inattention caused by boredom. Fatigue causes several
problems for drivers, such as slower reactions and decisions, slower control
movements, hallucinations; decreased tolerance for other road users, poor lane
tracking and maintenance of headway speed, and loss of situational awareness.
Symptoms vary among drivers, but may include: yawning, poor concentration,
tired or sore eyes, restlessness, drowsiness, slow reactions, boredom, feeling
irritable, making fewer and larger steering corrections, missing road signs,
having difficulty in staying in the lane and micro sleeps. As fatigue can
decrease the ability of a driver to maintain a steady speed, it increases the
speed variation, which can increase the speed variance of the traffic flow
(Roads and Traffic Authority, 2004).
A study conducted by Sagberg (1999) showed that 4% of incidences of
micro sleeps can lead to a crash, most of which are running-off-the-road
crashes (3.5%). The most common consequences of fatigue were incidents such as
crossing the right edge line (42%), which occurred more frequently than
crossing either the centre line (16%) or the left edge line (4.6%). Williamson, Feyer, Friswell and Sadural
(2001) surveyed professional long distance heavy vehicle drivers in
According to the National Highway Transportation Safety Administration
(1994), truck driver fatigue is a contributing factor in as many as 30-40% of
all heavy truck crashes. In 1995, the National Transportation Safety Board
found that of 107 heavy truck crashes, fatigue was a prominent factor in 75% of
the run-off-the-road crashes, with 68% of long-haul drivers and 49% of short
haul drivers involved in fatigue-related
crashes (Advocates for Highway and Auto Safety, 2001).
Data from Western Australia that used proxy measures (such as: “drifted
off curve or straight”, “wrong side of road with no overtaking maneuver”, “where
speed or alcohol was not a factor”), indicated that approximately 30% of the rural
crashes could be attributed to fatigue (Office of Road Safety, Western
Australia, 2004). This estimate is much higher than the estimated 14% fatigue
related accidents reported by Garber and Joshua (1989) because the former study
considered only the rural roads, whereas the later study considered all
highways types. Most fatigue-related crashes occur on rural roads. One reason for this is that the average trip length is
likely to be longer on these roads and inattention and drowsiness are brought
on by the constant speeds and monotony. In 1998-2002, 79% of
fatigue-related fatal crashes in
In another Australian study, Fell (1987) reported that
commercial trucks have a higher involvement in fatigue-related accidents, compared
to their involvement in other types of accidents. Heavy trucks in
Although there has been a significant amount of research conducted on the
effect of driver fatigue on safety, there has been virtually no published
research that addresses the effect of operating speed on driver fatigue. One study
conducted by Jiao (2004) assessed the impact of operating speed on fatigue. Thirty
drivers were chosen for observation and were randomly divided into three groups,
driving at 40 kph (24.86 mph), 80 kph
(49.71 mph) and 120 kph (74.57 mph), respectively. All of the three groups were
asked to drive for 2 hours, without any break. After the completion of two
hours, the heart rate variability was measured. The group that drove the
fastest had the maximum change in heart rate variability and the slowest group
had the least change. Based on this physiological measure, the author concluded
that the higher speeds induce more fatigue on drivers compared to lower speeds.
However, these results are based on the amount of fatigue per unit of time, not
the amount of fatigue per mile. The fastest group drove for 240 km in 2 hours;
whereas the slowest group drove only 80 km in the two hours. If all of the
three groups of drivers had traveled for an equal number of kilometers, the results
might have been different.
Oron-Gilad, Ronen, Cassuto and
The driving performance was measured using five variables: the average
lane position, lane position variability, steering wheel variability, average
speed and speed variability, and the rate of off-road incidents. The average
speed during the “fun” speed driving section was measured to be 110 kph (68.35
mph). The between subject comparison of driving performance under the 90 kph and
60 kph showed that performance in 90 kph speed trial was significantly poorer
than in the 60 kph speed trial. The physiological measures indicated that the
driver was most relaxed and least stressful operating at 90 kph compared to
operating at 60 kph or the even higher “fun” speed. Based on the survey
results, motivation to continue driving was observed to be significantly lower
in ‘low’ speed condition compared to “legal” speed condition. It was concluded
that, although lowering the travel speed can yield better driving performance
and lower rates of off-road incidents, it can also cause a significantly lower
motivation to continue driving and a significantly higher level of stress.
Although fatigue has received a large amount of attention in the
literature, particularly in the context of “hours of service,” none of the available
research or applications literature addresses the relationship between vehicle
speed and fatigue. To the extent that the speed is increased, the travel time,
and possibly the amount of fatigue, is reduced. There is no empirical data
indicating that increased speed, within the normal driving range, increases
fatigue. To the extent that fatigue is related to driving time, rather than
distance, higher speeds could reduce fatigue on a per mile basis. However,
there is no published research or data to support or contest this hypothesis.
2.5 Effects
of Speed and Weight on Braking Distance
Speed affects the
handling, stopping and operating characteristics of vehicles. Due to the simple
physics related to their large size and weight, vehicle speed has a significant
effect on truck handling and dynamics. Among these, stopping distance has been
the most frequent reason for setting lower speeds for trucks. Since rear-end
accidents make up a high proportion of accidents involving trucks, some policy
makers suggest that trucks should operate at slower speeds so that the stopping
distance of trucks would be made more compatible to that of automobiles. Braking
distance consists of two primary components: (1) the distance traveled by the
driver from the time a hazard is perceived to when brakes are applied and (2) the
distance traveled while brakes are applied. Generally the braking distance
required to come to a complete halt increases as the speed increases. According
to the North Carolina Department of Motor Vehicles (NCDMV) when the speed is
doubled, the braking distance increases by four times and the vehicle will have
four times the destructive power in a crash.
The effect of speed on a truck’s operating characteristics is determined
by its size and the configuration. Brake technology has been improving rapidly
for heavy trucks. The Federal Motor Vehicle Safety Standards (FMVSS) have
required anti-lock braking systems (ABS) on new trucks and trailers since 1997.
According to Harwood (2003), roughly 43% of the trucking fleet is estimated to
have anti-lock brakes. The wide spread use of anti-lock brakes on today’s
trucks helps avoid wheel lock and jackknife conditions, thus considerably
improving stability during braking. Anti-lock brakes also help when stopping a
truck in adverse weather conditions like ice, snow, and rain; however, the
braking distance of the trucks remains longer than that of automobiles.
The stopping distances of automobiles and trucks are compared in Table 8
below. The Oregon Trucking Association
web page provided estimates of the stopping distance for 80,000 lb., loaded
tractor-trailers and mid-sized passenger automobiles traveling on a dry, level
road.
The table illustrates that, for a completely loaded truck on a level
roadway, up to 60% more distance is required to come to a complete halt at 65
mph compared to the distance required for 55 mph. With the improvements in
truck braking systems (e.g., air disc brakes, electronic braking systems,
etc.), the stopping distance for trucks has been significantly reduced.
Currently, pneumatically braked truck tractors are required to stop from 60 mph
in 355 ft. at gross vehicle weight rating; whereas the stopping distance
requirement for passenger automobiles is 216 ft. The introduction of all-disc
brakes and
|
Speed |
Reaction
Distance Automobiles |
Reaction
Distance Trucks |
Stopping
Distance Automobiles |
Stopping
Distance Trucks |
|
40
mph |
44' |
44' |
124' |
169' |
|
55
mph |
60' |
60' |
225' |
335' |
|
65
mph |
71' |
71' |
316' |
525' |
“all S-cam” brakes,
the new NHTSA law will potentially require reducing the stopping distance for
trucks by 30 %, to 249 ft., making the stopping distance for automobiles and
trucks more comparable. This will
decrease the importance of one of the most frequently stated reasons for
differential speed limits that require trucks to travel slower.
There is one advantage that a truck driver has over an automobile driver
which is the eye height. Since the truck drivers’ an eye height is much higher than
automobile drivers’ (8 feet as compared to 3.5 feet when seated, they can see
farther down the road and over other vehicles. Therefore, the truck drivers
have an advantage in response time to forward hazards. This also helps offset the
effect of the longer stopping distances for trucks.
Many references, particularly in the popular literature,
discuss the fact that truck speeds should be low because the weight of the
truck significantly increases the stopping distance. This is not supported by
data. The brakes, tires, springs, and shock absorbers on heavy vehicles are
designed to work best when the vehicle is fully loaded. Empty trucks require greater
stopping distances, because an empty vehicle has less traction. It can bounce
and lock up its wheels, giving much poorer braking (Commercial Drivers License
Study Guide). The data available from NHTSA also indicates that the difference
in stopping distance between a lightly loaded truck (335 ft.) and heavy loaded
truck (355 ft.) is just 20 ft. In near future, with the new NHTSA braking
requirement law, the stopping distance for both the lightly loaded and heavily
loaded truck will be reduced by 30 %, thus reducing the difference between them
to only 16 ft.
2.6 Effects of Speed on Operational Costs
In addition to different posted speed limits on rural interstates, speed
differentials between heavy trucks and other vehicles occur due to the fact
that most large commercial trucking fleets use speed limiters to restrict truck
speed. One of the primary reasons for the use of limiters is to reduce
operating costs. This section will summarize the of speed on fuel economy, tire
wear and other maintenance costs.
2.6.1 Effects of Speed on Fuel Costs
The National Minimum Speed Limit (NMSL) of 55 mph was introduced in 1973
for the primary purpose of saving fuel during an energy embargo. This rule was in
effect until 1987, when it was modified and until 1995, when it was repealed.
After the 1987 and 1995 legislation speed limits were raised in most of the
states. However, many trucking companies preferred the lower speed to reduce
operating speeds. Although there is the possibility of increasing their revenue
with higher speeds, based on increased miles traveled by each truck, companies have
chosen to operate at lower speeds because of the assumption that the increase
in fuel costs would outweigh the benefit of increased revenue.
Within the trucking
industry, there is a common “rule of thumb” that “each increase in vehicle
speed of 1 mph reduces the fuel efficiency by 0.1 mpg.” This rule of thumb was developed
by The Maintenance Council [now the Technology and Maintenance Council] of the
American Trucking Association (1996) in their study that addressed the effects
of truck speed on operational costs.
Fuel economy tests were
conducted in 1987 and were re-published again in 1996. The actual tests were conducted
in 1987. Two trucks operated simultaneously, one with a 55 mph maximum speed
limit and other with a 65 mph maximum speed limit. The “55 mph” truck had an
average speed of 50.1 mph over the complete trip; while the “65 mph” truck had
an average speed of 57.1 mph. After testing, the average fuel consumption values
were observed to be 5.46 mpg for the 65 mph condition and 6.44 mpg for the 55
mph condition. Therefore a loss of 0.98 mpg was observed, which was caused by a
10 mph increase in the maximum operating speed and a 7 mph increase in the average
speed. It was concluded that for every 1 mph increase in average speed, there
is a 0.14 mpg penalty on fuel economy. These results were once republished by
the same committee in 1996, the reasoning being that most of the committee
members thought that these results were still valid. Until today, the majority of the trucking
industry uses these estimates when making speed limiter decisions. There were
methodological issues involved with the 1987 study. The vehicles were
underpowered, compared to most trucks today and, therefore, were not as suitable
for operating at the higher speeds. During one of the test runs, the vehicle
operating at 65 mph was not able to maintain its speed, which creates questions
as to the validity of this study. It should be noted that the measurement of
fuel efficiency is very complex and the Technology and Maintenance Committee of
the American Trucking Association has recently formed a special group to study
the test procedures for measuring the fuel efficiency of trucks.
Broderick studied
the effect of speed on fuel consumption of heavy trucks (1975). The tests were
conducted at 50, 55 and 60 mph operating speeds on the
Efficiency losses
in heavy trucks include (a) aerodynamic drag, (b) grade resistance, (c) the
rolling resistance, and (d) engine accessory/drivetrain losses. Aerodynamic
drag has the largest effect at higher speeds (above 50 mph). There are a large
number of aerodynamic forces acting on a vehicle that depend upon the speed,
frontal area, and external shape of the vehicle. According to the U.S.
Department of Energy, at 70 mph, aerodynamic drag accounts for approximately
65% of the total energy loss for a typical heavy truck (Ang-Olson and Schroeer
(2003)). The authors state that, with no aerodynamic treatment, at 65 mph a
total of 264 horsepower is needed to overcome all of the forces acting on the
truck. Aerodynamic forces account for 145 hp (55%) of power demand, tire rolling
resistance accounts for 87 hp (33%), and miscellaneous forces account for 32 hp
(12%). At 65 mph, with full aerodynamic treatment, the horsepower required to
overcome aerodynamic forces can be reduced to 113 hp (22% reduction).
The use of roof-top
deflectors and fairings, cab-side extenders, gap seals, tapering rears of the
trailer, along with underside and trailer sidewall improvements reduce
aerodynamic drag (Cooper, 2003). Aerodynamic drag has been reduced by 40% in
the last 30 years. Starting with aerodynamic drag coefficient value of 1 in the
1970’s, today the value can be reduced to 0.7. Cab-over-engine designs further
lower drag to 0.5. If the tractor and the trailer could be integrated then this
value could be reduced to 0.4 (Muster, 2000).
A brochure
published by Cummins, Inc. (2003) listed recommendations for improving the fuel
economy of heavy trucks. The company brochure states that the “rule of thumb”
is for each 1 mph increase in speed above 55 mph the fuel economy decreases by
0.1 mpg. It was also indicated that tires have the largest effect on fuel
consumption below 50 mph, whereas aerodynamics is the most important factor
above 50 mph. There were also other factors
listed in the Cummins (2003) study that could improve the efficiency of trucks,
summarized in Figure 18. The table illustrates that “driver variability” is
almost twice the effect of vehicle speed.
According to
Deierlein (2000), the most important fuel economy variable was the driver, who
controls the idle time, vehicle speed, brake use, etc. The difference between a
“good” and a “bad” driver can be up to a 35% in fuel efficiency. Another very
important factor is the proper specification
and setup of the engine. An electronically controlled engine can save up to 15% over a manual
engine. Use of cruise control versus no cruise control can also improve fuel
economy by up to 6%. These percentages are very dependent upon the skill level
of the driver.
Considering the
importance of fuel consumption by commercial carriers to the national economy,
it is interesting that there is very little published research addressing the
effect of truck speed on fuel consumption. To further address the issue of
vehicle speed and fuel consumption the literature pertaining to automobiles will
be addressed.
The power-to-weight
ratio has been found to be an important factor in fuel consumption. For small power-to-weight automobiles, changing
speed from 55 mph to 65 mph increased fuel consumption by approximately 13%;
whereas for high power-to-weight automobiles the fuel consumption increases by only
9% (Bedard, 1996). This result has serious implications for the interpretation
of the Maintenance Council’s results that have previously been discussed. Tests
that are conducted with trucks that do not have sufficient power could give a
distorted view of the impact of speed..
The Transportation
Energy Data Book: Edition 24, published by US Department of Energy, illustrated
the relationship between speed and fuel efficiency for automobiles, as shown in
Figure19 below. According to the 1997 test results, the increase in speed from
55 mph to 65 mph results in a 9.7
% loss in fuel economy; however, for the
same
Figure 19. Effect of Speed on Fuel Consumption for Automobiles (Source: US Department of Energy, 2004)
increase in speed,
the 1987 test results indicated a 17.8% loss in fuel economy. These results
indicate that, with improvements in vehicles aerodynamics and engine
components, traveling at a higher speed has a less negative impact on fuel
consumption.
2.6.2 Effects
of Speed on Tire Costs
After aerodynamic
drag, the most significant factor that affects fuel consumption is tire rolling
resistance. The energy (fuel) required to move the vehicle is directly
proportional to the rolling resistance coefficient, which is influenced by the frictional
properties of road and tires.
According to Muster (2000), there is
a 1% truck fuel efficiency gain for every 2.6% reduction in the rolling
resistance coefficient. A study by Hall and Moreland (2001), found a reduction
in the rolling resistance of 10% can improve fuel efficiency by 0.5% to 1.5%
for automobiles and 1.5 to 3% for trucks. They also found that 5 to 15% of the
fuel is necessary to overcome rolling resistance for passenger automobiles and
15 to 30% for heavy trucks. There has been a significant effort to reduce the
tire rolling resistance. Reductions in rolling resistance of 50% have been
accomplished, relative to 1980 level. Muster
(2000) reports that the rolling resistance coefficient has been reduced
significantly by the introduction of radial tires (from 0.01 to 0.0054). It is
predicted that in the future, super single tires will be able to decrease the
rolling resistance coefficient to as low as low to 0.005.
Under-inflated
tires increase the rolling resistance coefficient. According to Farkhan (1999), tires represent
approximately 20% of the total maintenance costs. A 10 psi under inflation can
result in a 1% increase in fuel consumption, and 20% faster tire wear. A properly
inflated tire running at 65 mph will heat up to approximately 170 degrees and
5-psi under inflation can cause the tires to become up to 25 degrees hotter.
Under inflation also results in more flexing, thus limiting the number of
potential retreads from each casing.
A document published by Goodyear (2003) states that for every 1 mph
increase in operating speed over 55 mph, there is a reduction of 1% in tread
mileage. This means that operating at 75 mph instead of 55 mph would cost
trucks 20% in terms of tread life. Similar data are presented in a document
published by Bridgestone/Firestone Commercial Truck Tires (2004). This brochure
states that higher speeds reduce tire life by 10-30%. At higher speeds, the tires
are hotter, which can reduce casing life and retreadability. The maximum load
capacity at 75 mph decreases by up to 12%, from the maximum load capacity at 65
mph. However, load-carrying capacity decreases by only 4% when operating at 70
mph rather than at 65 mph.
The effect of speed on rolling resistance was explained in detail by
Hall and Moreland (2001). Different trends are observed as the speed increases.
They found that the phase lag angle for the composite material of tires
decreased with frequencies in the range of the rolling tire deformations. This results
in a decrease in rolling resistance as speed increases. They also found that
tire temperature increases with speed, which also reduces the rolling
resistance. However, these two positive effects are more than offset by the
increase in the tire deformation that occurs due to centrifugal force with
increasing speed. In addition, the aerodynamic drag, which is a component of
rolling resistance, also increases with the square of speed. Thus, the authors
concluded that an increase in speed results in an increase in the rolling
resistance.
2.6.3 Effects of Speed on Maintenance Costs
The Maintenance
Council of the American Trucking Association analyzed the effects of speed on
operation costs in a landmark study (1996). The operating speed was assumed to
affect the component durability. No detailed data were presented in this study
and most of the results obtained were based on the consensus among the
committee members. According to the document, an increase in operating speed
from 55 mph to 65 mph had the following effects:
(a)
10 to 15% decrease in miles-to-engine overhaul
(b)
oil consumption increase of 15%
(c)
shortened mileage between preventive maintenance intervals
(d)
decrease in effective tire casing life
(e)
reduction of up to 15% in brake lining life
With respect to comparing
the potential increase in productivity (due to more miles traveled) to the estimated increase in cost, The
Maintenance Council believed that it was nearly impossible to make a case for
sufficient productivity gains to offset the increased costs associated with
operating at speeds higher than 55 mph. No other published data related to
maintenance costs, engine life, and operating speed were found by this study.
2.7 Effects
of Speed on Pollution
The transportation
sector is the dominant source of fuel consumption and emissions in the
The Environmental Protection Agency has very little data on the
emissions for heavy diesel trucks and does not differentiate among freeways,
ramps, arterials, etc. Furthermore, the speed correction factors for trucks
have only been developed using the older model (MOBILE5), thus they are
questionable in that, as previously discussed, the factors may be overestimated
by the older model. Table 10 provides the estimated changes in speed correction
factors from 55 and 60 mph to 65 mph. Figure 27 illustrates the extrapolated
estimates calculated by Monsere, Newgard, Dill, Rufolo, Wemple, Bertini and
Miliken, C. (2004) using Environmental Protection Agency’s (2001) data.
The change in speed
limit would impact only the “running emissions” that are produced when the
engine is warm and the vehicle is in motion. But these are only a part of the
total emissions produced. Therefore, the amount of increase in overall
emissions due to increased highway speed is probably overestimated by these speed
correction factors.
In addition to speed, the roadway geometry
also has a large impact on the emission rates. Kean (2003) found that the
carbon monoxide (CO) emissions for light duty vehicles increased more with
speed while going uphill and varied little with speed while going downhill.
Furthermore, vehicle acceleration and deceleration were found to have a
significant impact on the emission rates.
In 1997, E.H.
Pechan and Associates estimated the impact of the increased speed limit in 1995
on emissions using the MOBILE5 model. They found that the emission of volatile
organic compounds on roadways with higher speed limits increased by 1 to 4%,
while the NOx and CO increased by much higher percentages (1-35% and
1-38%, respectively). According to den Tonkelaar (1994), for automobiles the
increase in CO and NOx emissions with speed is greater than those of
hydrocarbons, especially for CO, which was observed to increase rapidly beyond
90 kph.
|
|
Total
Hydrocarbons (THC) |
Carbon Monoxide
(CO) |
Oxides of
Nitrogen (NOx) |
|
55 to 70 mph |
16% |
24% |
16% |
|
60 to 70 mph |
10% |
15% |
10% |
|
65 to 70 mph |
5% |
7% |
5% |
|
|
Total
Hydrocarbons (THC) |
Carbon Monoxide
(CO) |
Oxides of
Nitrogen (NOx) |
|
55 to 65 mph |
-2% |
24% |
45% |
|
60 to 65 mph |
0% |
14% |
23% |
2.8 Effects of Speed and Speed Differentials
on Roadway Wear
There is an issue
of roadway wear as a function of highway speed limits. Chatti (1996) studied
the impact of speed on pavement strains. The effect of vehicle speed on
pavement strains was significant. Increasing vehicle speed from 2 mph to 40 mph
caused a decrease of approximately 15 to 30% in transverse strains and 30 to
40% in longitudinal strains. However, one main issue with this study, relative
to the current effort, was that the speed data did not exceed 40 mph.
Luskin (2001)
studied the impact of truck operations on the highway infrastructure, and
concluded that, for a truck moving over smooth pavement, the load transmitted
to the pavement would be static. An increase in the operating speed of the
truck would not affect the intensity of stress on the pavement, but it would
reduce the duration for which the vehicle would be on pavement, thus reducing
the amount of pavement damage. Akram, Scullion, and Smith (1993) studied the
effect of operating speed on pavements using a multidepth deflectometer.
Evaluation of vertical compressive strain data showed that sub grade strains at
the bottom of the asphalt layer decreased substantially with an increase in
vehicle speed. However, this study considered only speeds up to 55 mph.
While there have
been very few scientific studies conducted to investigate the relationship
between higher speeds (speeds above 55 mph) and road wear and maintenance, there
seems to be a common consensus among researchers that the amount of wear and
tear caused on the roadway is directly proportional to the time during which
the roadway is exposed to the vehicle’s tires. Therefore, as the speed limit
increases, the amount of time that the tires will remain in contact with the
unit area of the road decreases. Thus, wear caused by the tires on that
particular unit area of road will decrease. Overall, as the traveling speed of
the vehicles increases, the time for which the vehicles will be traveling on
the road decreases, thus decreasing the roadway wear.
Although there have
been no direct studies of the issue, there could be another important
relationship between speed differentials and roadway wear. Generally, as the
speed differential between automobiles and trucks, or among trucks, increases,
the amount of maneuvering increases. These maneuvers include decelerating and
accelerating and moving laterally across lanes. These activities could have a
very large effect on roadway wear.
This research effort used a number
of approaches to assess the effects of speed differentials for heavy trucks and
lighter vehicles. The approach included observations of both truck and
automobile driver behaviors on highways with different speed limit
configurations. In addition, opinion data were collected from over-the-road
truck drivers, fleet safety and maintenance personnel, and engineers from the
truck, engine, and tire manufacturers. Computer simulation of highway traffic
was used to investigate the effect of speed differentials on the amount of
interaction among vehicles. Finally, safety and operational data from
participating fleets were used to address the effect of speed differentials on
the trucking industry.
3.1 Measurement of Traffic Speeds on Highways
with Different Limits
The research literature discussed in
the previous section generally evaluates highway safety issues using the
available accident and fatality databases (i.e.,
Different sites were investigated on
rural interstate highways in
All of the sites that were chosen were
rural interstate highways that were flat and relatively straight for at least
two miles prior to the site. The objective of the study was to address the
highway geometry that is representative of the majority of rural interstate
highway miles in the
The data were collected between 11
am and 4 pm on weekdays to reduce the effect of commuter and weekend traffic.
During the data collection period, the weather was clear and visibility was
good. The speeds of both trucks and light vehicles were measured with a Prolaser
II, Doppler lidar, manufactured by Kustom Signals, Inc.
When collecting traffic speed data,
the relative levels of enforcement can obviously affect the results. Although
it is difficult to characterize the enforcement levels at the various sites,
there were no speeding citations observed at any site during any of the data
collection periods.
3.2 Computer
Simulation Evaluation of Speed Differentials on Vehicle Interactions
As indicated in the
Literature Review, an important factor in traffic flow and highway safety is
speed variance, which is a measure of the distribution of vehicle speeds on a
roadway. Speed variance is often
represented by the difference between the 85th and 50th
percentile vehicle speed. The conclusion
of a number of studies has been that higher speed variance increases the risk
of two-vehicle accidents. This is simply the result of an increase in the
number of interactions among vehicles (passing or being passed). Speed
differentials, whether due to posted speed limits or company policies, increase
the speed variance on highways. A computer simulation was developed to
quantitatively investigate the relative number of vehicle interactions that
result from traveling either faster or slower than the average traffic speed.
The observed speed measurements from the sites in
3.3 Assessment
of Speed Limiters Use on Heavy Trucks
In addition to the limitations
placed on heavy trucks by posted state-regulated posted speed limits, the speed
of many trucks is also limited by electrical/mechanical devices that are used
to restrict the maximum speed of the truck. These devices, which were
originally mechanical (i.e., speed governors), now control the speed through
the electronic control module (ECM) on truck engines. The primary reason for fleets using speed
limiters is to improve fuel consumption. With the diesel fuel prices currently over
$3.00 per gallon, fuel efficiency is an important issue for trucking companies,
as well as for the shippers and consumers who eventually pay the additional
costs of goods shipped by truck.
To determine the impact that company
speed limitation policies have on traffic flow, surveys were conducted of
over-the-road drivers, trucking company representatives, and truck sales
organizations. The surveys of drivers were conducted at 11 truck stops in
Representatives of commercial
carriers were surveyed by telephone, visits to their headquarters, and through
personal communications at professional/trade organization meetings (i.e., ATA
Technology and Maintenance Council, SAE Truck and Bus Meeting, Great American
Truck Show, etc.). Information as to company policies on speed limiting devices
was also collected as part of a more general survey that is discussed in a
later section of this report.
3.4 Survey
of Truck Drivers’ Opinions
Truck drivers are an important
stakeholder in the context of speed differentials that result from both
regulatory speed limits and company policies.
Truck drivers were surveyed at the truck stops discussed above to obtain
their opinions about truck speed in general and speed differentials, in
particular. Out of the total of 205 drivers surveyed, 115 were “company
drivers” who drive trucks owned by commercial fleets. Sixty eight (68) were
owner-operators and the remaining 22 drivers did not indicate their status. Of
the owner-operators, 20 were leasing their trucks to fleets and 48 operated
under their own authority.
The drivers were surveyed as they
filled their vehicles with fuel or in the restaurant or drivers’ lounge at
truck stops. The survey is provided in Appendix F. The majority of the drivers completed the
entire survey; however, due to limited time available for some drivers, an
abbreviated list of questions was used to obtain the most critical information
for the study. These questions are indicated with asterisks on the survey shown
in Appendix F. Prior to completing the
survey, the drivers signed the Informed Consent (Appendix I) as per the
requirements of the
3.5 Survey
of Carrier Fleet Safety and Maintenance Personnel
As previously discussed, carrier fleets
often restrict the speed of their trucks, which results in truck-automobile
speed differentials, independent of the posted speed limits on rural
interstates. The speed policies adopted by companies are primarily the result
of two overriding factors: safety and economics. To address these considerations, safety and
maintenance personnel from commercial fleets were surveyed. The opinions of
these individuals were obtained by telephone survey, web survey, and personal
interviews at company facilities or professional/trade association
meetings. The survey instruments used
for the safety and maintenance personnel are provided in Appendix F and G. The survey questions addressed both the
effects of truck speed and speed differentials on both safety and operating
costs.
As with the truck driver surveys,
both the responses and the rationale behind the responses from the safety and
maintenance personnel were obtained to determine the basis for the policies
used by the companies. The results and
conclusions drawn from those interviews are presented in the Results section of
this report.
3.6 Survey
of Equipment Manufacturers of Trucks, Engines and Tires
One of the primary reasons for speed
limitations adopted by both fleets and owner-operators is related to the
effects of speed on operational costs. As indicated in the Literature Review,
there is very little information in the research literature relating to the
effect of speed on operational costs. To the extent that the information was
available in the public domain, it was generally provided in materials that are
distributed by the manufacturers of the various components (engines, tires,
etc.). To address the issue of truck speed on operational costs, the
manufacturers of the equipment were surveyed. These surveys were primarily
conducted by telephone and by personal communication at professional/trade
association meetings (i.e., American Trucking Association’s Technology and
Maintenance Council meetings, Society of Automotive Engineer’s Bus and Truck
meetings, etc).
3.7 Comparison
of Fleet Experience in States with Different Speed Limits
When companies
adopt truck speed limit policies that are lower than the traffic speed, it
effectively results in a speed differential for that fleet. As the posted
automobile speed increases (i.e., 65, 70 or 75), the result is that the
effective speed differential increases for the fleet. To analyze the impact of
the "effective" speed differentials between trucks and light
vehicles, participating companies were requested to provide their accident data
from selected states for the past four years (2001 through 2004). Twenty-two states were selected based on
their posted speed limits. Eleven states had differential speed limits and 11
states had uniform speed limits. The maximum speed limits in the selected
states varied from 65 to 75 mph.
The accident type
(lane change, passing, rear-ended, etc.), weather conditions during the accident,
and the highway type on which the accident occurred were included in the data
set. The monthly vehicle miles traveled by the companies' trucks in each of the
states were also requested. Although the number of vehicle miles traveled on
rural interstates was not available, this value was estimated as a proportion
of the total state miles. The speed limits for the trucks in the participating
companies were 62 and 65 mph. Therefore, the "effective" speed differentials were the difference between the
|
Speed Differential States |
Uniform
Speed Limits |
|||
|
|
75 |
65 |
|
75 |
|
|
75 |
65 |
|
75 |
|
|
75 |
65 |
|
75 |
|
|
70 |
65 |
|
75 |
|
|
70 |
60 |
|
75 |
|
|
70 |
55 |
|
75 |
|
|
70 |
55 |
|
70 |
|
|
65 |
60 |
|
65 |
|
|
65 |
55 |
|
65 |
|
|
65 |
55 |
|
65 |
|
|
65 |
55 |
|
65 |
limited truck speeds and the posted
speed limit for trucks and automobiles in a particular state. For example,
although there is no regulated speed differential in Arizona (75 mph for both
trucks and light vehicles), the "effective" speed differential for
the 62 mph fleet is 13 mph. Whereas, the "effective" speed
differential for the fleet in the state of Kentucky was 3 mph (uniform 65 mph
limit).
The objective of
this phase of the study was to compare the fleet accident experience across the
states that have different speed limits that result in different
"effective" speed differentials. The analyses of the fleets'
experience with respect to different types of accidents are presented in the
Results section.
3.8 Financial Cost-Benefit Analysis of Operating Speeds
As discussed earlier, one important
reason that commercial trucking firms have lower operating speeds is to reduce
the operating and maintenance costs. However, the reduction in operating costs
by reducing speed is also accompanied by a reduction in company revenue, in
that the truck assets potentially travel fewer miles per year. To evaluate the
relative costs and benefits associated with lower operating speeds, operating
and maintenance data were obtained from the participating companies. Estimates
of the relative net revenues associated with different speed limit policies are
presented in the Results section.
4.1 Traffic
Speed Measurements under Different Speed Limits Configurations
This section of the report addresses
the distribution of vehicle speeds on highways that have different speed limit
configurations. The results illustrate how the posted speed limits affect the
distribution of traffic speeds. In addition, the data are divided into heavy
trucks and light vehicles (referred to in this report as automobiles). Four
configurations were selected to
represent the range of both absolute speeds and speed differential
configurations.
|
State |
Automobiles |
Trucks |
Differential |
|
|
75 |
75 |
0 |
|
|
70 |
70 |
0 |
|
|
70 |
65 |
5 |
|
|
65 |
55 |
10 |
For each configuration, the
distribution of vehicle speeds is presented. In addition, the separate
distributions for trucks and automobiles are presented. To illustrate the
relative number of vehicles at various speeds, the combined truck and
automobile distributions are documented as proportions. However, due to the
fact that the volume of trucks and automobiles differs significantly within a
site and from site to site, frequencies are used to compare the truck and automobile
distributions. The statistics that are used to represent the traffic flow are
the mean and standard deviation of the speeds. To represent the dispersion of
speeds observed at the sites, the 85th percentile and median speeds
are provided. There are two primary methods of calculating speed variance
reported in the literature: (a) standard deviation of the individual vehicle
speed and (b) the difference between the 85th percentile speed and
the median speed (50th percentile). The present study defines speed
variance as the difference between the 85th percentile and the
median speed. The last statistic reported for each configuration is the
compliance rate (proportion of vehicles traveling at or below the posted speed
limit).
4.1.1
Speed data were
collected at two sites on Interstate 1-40 near
The general shape
of the distribution shown in Figure 22 is similar to those found in the
research literature for many different sites around the
Figure 23
illustrates the speed distributions for automobiles and trucks separately. The
speed of 362 automobile and 169 trucks were measured. The mean of the
automobile speed distribution was 73.51 mph and the standard deviation was 4.32
mph. The 85th percentile was 78 mph, the median speed was 74 mph and
the speed variance was 4 mph. The compliance rate for automobiles was 21.8%.
The mean speed for trucks was 66.70 mph and the standard deviation was 3.69
mph. The 85th percentile speed was 70 mph, the median speed was 66
mph, and the speed variance was 4 mph. The compliance rate for trucks was
32.5%. The data illustrate that, although the posted speed differential was
just 5 mph, the real speed differential between the automobiles and trucks was
6.8 mph.
4.1.2
Speed data were
collected on Interstate I-57, near
Figure 25
illustrates the speed distributions of automobiles and trucks, respectively,
for the
The mean of the
automobile speed distribution was 73.24 mph, which was 8.24 mph above the
posted speed limit. The standard deviation was 5.67 mph, the 85th percentile
speed was 79 mph, and the median was 73 mph. The speed variance for automobiles
was 6 mph. The compliance rate for the automobiles was only 7.2%.
The mean speed for
trucks at the
the authorities as
excessively speeding. The truck drivers expressed a perception that the speed
limit enforcement for trucks was significantly stricter than for automobiles at
the
4.1.3
Speed data in
Figure 27
illustrates the separate speed distributions for 611 automobiles and 257 trucks
from the
The mean speed of
trucks was 68.61 mph, which is 1.39 mph below the posted speed limit. The
standard deviation was 4.55. The 85th percentile was 70 mph, median
speed was 66 mph, and the speed dispersion was 4 mph. The compliance rate for
the trucks was 69.6%. Although the difference in the posted speed limit for
trucks in
4.1.4
Speed data were collected
on the Cherokee turnpike (
The mean speed for
all vehicles was 74.24 mph and the standard deviation was 4.93 mph. The 85th
percentile speed was 79 mph, the median was 74 mph and the speed variance was 5
mph.
Figure 29
illustrates the speed distributions for automobiles and trucks at the
4.1.5 Summary of Speed Data from Different Speed Configurations
Both the statistics
and the shapes of the vehicle speed distributions are important in evaluating
the effects of regulatory speed differentials on driver behavior and highway
safety. In particular, the separate distributions for automobiles and trucks
provide insight that is not provided by the combined data. A summary table that
presents the statistics for each of the speed configurations is provided in
Table 11.
The objectives of posted speed
limits are to both reduce the negative effect of vehicles going at excessive
speeds and to improve the flow of traffic. This is the reason that minimum
speed limits (e.g., 45 mph) are imposed on highways. A significant amount of
the research literature attributes the cause of accidents to "speeding."
However, most accident reporting systems define speeding as "exceeding the
posted limits or driving too fast for conditions." The result of this
definition is that many, if not most, two-vehicle crashes that are
characterized as being caused by speeding occur when the vehicle is actually
traveling slower than the posted speed limits. The result is that the number of
accidents attributed to exceeding the posted speed limit is often
overestimated. From the data provided in this section, the amount of
overestimate is possibly even more severe for heavy truck accidents. For the
purposes of this report, speeding is defined only as the amount that the
vehicle is exceeding the posted speed limit.
4.1.5.1 Speed Differentials and Compliance
One issue that is
important from a regulatory perspective is the compliance rate for the
different configurations. Table 12 illustrates the amount that the average
speed exceeds the speed limits and the compliance rates for both automobiles
and trucks.
|
State |
|
|
Traffic |
Automobile |
Truck |
|
|
|
Average (mph) |
71.2 |
73.2 |
64.2 |
|
|
|
Standard Deviation |
6.54 |
5.67 |
4.00 |
|
|
|
Sample Size |
1140 |
878 |
262 |
|
Autos: |
65 mph |
Proportion of Trucks |
0.23 |
|
|
|
Trucks: |
55 mph |
Compliance (%) |
|
7.17 |
0.0 |
|
(ADT = 19900) |
85th% (mph) |
78 |
79 |
68 |
|
|
|
|
50th% (mph) |
71 |
73 |
64 |
|
|
|
Speed Variance |
7 |
6 |
4 |
|
|
|
|
|
|
|
|
|
|
Average (mph) |
71.4 |
73.5 |
66.7 |
|
|
|
Standard Deviation |
5.19 |
4.32 |
3.69 |
|
|
|
Sample Size |
531 |
362 |
169 |
|
Autos: |
70 mph |
Proportion of Trucks |
0.32 |
|
|
|
Trucks: |
65 mph |
Compliance (%) |
|
21.82 |
32.54 |
|
(ADT = 22000) |
85th% (mph) |
77 |
78 |
70 |
|
|
|
50th% (mph) |
72 |
74 |
66 |
|
|
|
|
Speed Variance |
5 |
4 |
4 |
|
|
|
|
|
|
|
|
|
|
Average (mph) |
71.5 |
72.6 |
68.6 |
|
|
|
Standard Deviation |
5.16 |
4.95 |
4.55 |
|
|
|
Sample Size |
858 |
611 |
247 |
|
Autos: |
70 mph |
Proportion of Trucks |
0.29 |
|
|
|
Trucks: |
70 mph |
Compliance (%) |
|
31.42 |
69.64 |
|
(ADT = 34831) |
85th% (mph) |
77 |
77 |
73 |
|
|
|
50th% (mph) |
72 |
73 |
69 |
|
|
|
|
Speed Variance |
5 |
4 |
4 |
|
|
|
|
|
|
|
|
|
|
Average (mph) |
74.2 |
74.8 |
72.3 |
|
|
|
Standard Deviation |
4.93 |
4.61 |
5.63 |
|
|
|
Sample Size |
154 |
121 |
33 |
|
Autos: |
75 mph |
Proportion of Trucks |
0.21 |
|
|
|
Trucks: |
75 mph |
Compliance (%) |
|
52.89 |
72.72 |
|
(ADT = 3500) |
85th% (mph) |
79 |
80 |
77 |
|
|
|
50th% (mph) |
74 |
75 |
72 |
|
|
|
|
Speed Variance |
5 |
5 |
5 |
|
Automobile speed limit |
Truck speed limit |
Average automobile speed above posted limit (mph) |
Compliance (%) |
Average trucks speed above posted limit (mph) |
Compliance (%) |
|
65 |
55 |
+8.2 |
7.3% |
+9.2 |
0.0% |
|
70 |
65 |
+3.5 |
21.8% |
+1.7 |
32.5% |
|
70 |
70 |
+2.6 |
31.4% |
-1.4 |
69.6% |
|
75 |
75 |
-0.8 |
52.9% |
-2.7 |
72.7% |
Figure 30
illustrates the compliance rates for automobiles and trucks in different speed
limit configurations. These compliance data are consistent with the results
reported in the literature, in that when the posted speed is significantly
below the design speed for a highway, the compliance rate can be very low. In
these cases, the motorists ignore the posted limits and adopt a speed criterion
based on the traffic speed. When the posted speed is closer to the design
speed, motorists tend to comply more closely.
It can also be observed from the data
in Table 12 that the compliance rate of trucks is often higher than that of
automobiles. At the
4.1.5.2
Posted Speed Limits
and Mean Speeds and Differentials.
Table 13 illustrates that for the observed rural highways that had
similar design speeds, the mean speeds for automobiles were very similar, even
if the posted speeds were quite different. Although the speed limit in
Figure 31. Mean
Speed Limits for Different Speed Configurations
|
Automobile Speed Limit |
Truck Speed Limit |
Mean Traffic
Speed (mph) |
Mean Automobile
Speed (mph) |
Mean Truck Speed (mph) |
Posted Speed
Differential (mph) |
Observed Speed
Differential (mph) |
|
65 |
55 |
71.2 |
73.2 |
64.2 |
10 |
9.0 |
|
70 |
65 |
71.4 |
73.5 |
66.7 |
5 |
6.8 |
|
70 |
70 |
71.5 |
72.6 |
68.6 |
0 |
4.0 |
|
75 |
75 |
74.2 |
74.8 |
72.3 |
0 |
2.5 |
Again, it should be
noted that the
The data also
illustrate that there was an effective speed differential between automobiles
and trucks, even if there was no posted speed differential. For example, the
observed differential at the
4.1.5.3 Posted
Speed Limits and Speed Variance
As presented in the
Literature Review, many studies have observed that the interaction among
vehicles is an important factor in determining the potential risk for
two-vehicle accidents on highways. The number of interactions is represented by
the standard deviation and the speed variance. In this research, both measures
are presented. Each measure has advantages and disadvantages. The speed data
were measured during this study as integers (69, 70, 71, etc.). The result is
that the speed variance statistic is also an integer value and is, therefore, a
relatively insensitive measure of traffic speed dispersion. The standard
deviations shown in Table 14 provide a more sensitive measure of the variation
in traffic speeds.
|
Automobile Speed Limit |
Truck Speed Limit |
Standard
Deviation for Traffic Speed (mph) |
Standard
Deviation for Automobile Speed (mph) |
Standard
Deviation for Truck Speed (mph) |
|
65 |
55 |
6.54 |
5.67 |
4.00 |
|
70 |
65 |
5.19 |
4.32 |
3.69 |
|
70 |
70 |
5.16 |
4.95 |
4.55 |
|
75 |
75 |
4.93 |
4.61 |
5.63 |
These data
generally follow the trend cited in the literature that indicates that as the
traffic speed increases, the standard deviation is reduced. For example, the
variance is highest at the
The variation among
truck speeds appears to depart from the relationship where the amount of
variation decreases as the speed increases. Figure 32 illustrates the standard
deviation for all vehicles, automobiles and trucks, respectively. The standard
deviation for speed is lower in states having slower speed limits for trucks (
4.1.5.4 Speed Differentials and Clustered Congestion
As discussed
earlier in the Literature Review section, imposition of differential speed
limits could lead to traffic congestion. This argument was evident when the
sequential traffic data observed at
With differential speed limits,
automobiles tend to travel at speeds significantly higher than those trucks.
This results in faster-moving automobiles traveling in the left lane, while the
slower moving trucks get “stuck” in the right lane. An effect of truck speed
limiters with different maximum speeds is that the trucks with higher limits or
no limiters could move faster, but get “stuck” in the right lane behind the
slower moving trucks, thus
Figure 33. Sequential Traffic Arrival at
Different Speed Configurations
Figure
34. Illustration of Localized Congestion
leading to
a bottleneck situation. When a truck with a slightly higher limit (e.g., 2 mph)
attempts to pass the slower truck, passing can take a significant amount of
time. For example, if a truck with a 65
mph limit passes a truck with a 62 mph limit, in a 75 mph speed zone, traffic
tends to experience “clustered” congestion (Figure 35).
Figure 35.
Illustration of “Clustered” Congestion
4.2 Impact
of Speed Differentials on the Number of Vehicle Interactions
In this section,
the number of interactions an individual vehicle will have with other vehicles
operating at different speeds in the traffic flow is modeled. The goal was to
investigate the number of times a vehicle passes and is passed as a function of
their individual speed relative to the traffic speed. It was assumed that the
“reference vehicle” is operating at a uniform speed limit on a rural interstate
highway. All the vehicles were assumed to be traveling at steady speeds in
free-flowing traffic, and there was no speed fluctuation due to traffic
congestion when one vehicle passed another vehicle.
For the purposes of
the model, the distribution of automobile and truck speeds that were observed
at the Rolla,
The computer
simulation calculated the total number of passing incidents involving the
reference vehicle at different operating speeds. Each time a passing incident
occurred, it was noted whether the passing or passed vehicle was a truck or an
automobile. The total number of “passing” and “being passed” incidents were
combined to determine the total number of interactions that would be
encountered during the complete 1000 miles trip. This procedure was repeated
for operating speeds from 60 to 80 mph. The results are illustrated in Figure 36.
Figure 37 shows the results obtained after repeating the same procedure with
the data collected from
Figures 36 and 37 illustrate
that the number of interactions is minimized when the “reference vehicle” is
traveling at the average traffic speed. Figures 38 and 39 illustrate the relative
frequency of “passing” and “being passed” incidents by vehicle type (automobile
versus truck). It can be seen that the relative frequency of a truck passing automobiles
is very low, which is counter to publics’ perception that trucks frequently
pass automobiles. It should be noted that the percentage of trucks was higher
at the
Figure 40 indicates
that, as the speed of the individual vehicle deviates from the mean traffic speed,
the number of interactions increases and the potential for being involved in a two-vehicle
accident increases. The interactions with other vehicles were minimized at the
average speed of traffic, which was 1.8 mph above the posted speed limit for
Figure 41
illustrates the number of interactions for a posted speed differential of 65
mph for automobiles and 55 mph for trucks. The number of interactions for a
vehicle moving at 60 mph is 149% higher than going at traffic speed; whereas,
the number of interactions for a vehicle traveling at 80 mph, which is 15 mph
above the posted speed limit, is only 70% higher when compared to the frequency
of interactions at the average traffic speed. A truck traveling at the speed
limit (55 mph) would have over four (4) times the number of interactions (407 %
more) compared to a truck going at traffic speed.
4.3 Use
of Speed Limiter Use on Heavy Trucks
In this section,
the results of a survey that was administered to obtain information from truck
drivers on speed limiter settings are presented. The distribution of speed limiter settings
based on the truck driver category and fleet characteristics are discussed in
detail.
Table 15. Summary of Results Obtained from Truck
Driver’s Survey
on Speed Limiter Use
|
Driver category |
Drivers surveyed |
Number of drivers
with speed limiters |
Percentage of drivers
having speed limiters |
|
Company Drivers |
136 |
123 |
90.4 |
|
Lease Drivers |
16 |
11 |
68.8 |
|
Owner-Operators (owning just the tractor) |
38 |
15 |
39.5 |
|
Owner-Operators (owning both tractor and
trailer) |
24 |
6 |
25.0 |
|
Did not Identify Themselves |
22 |
21 |
95.5 |
|
Total |
236 |
176 |
74.6 |
4.3.1 Driver
Category and Speed Limiter Settings
The classification
of the 236 drivers surveyed is shown in Table 15. Speed limiters were on 74.6% of the
trucks. Of the 176 with limiters, the
breakdown by driver category and the proportion of each category that had
limiters are shown in Table 15. In addition to the driver survey, thirty nine
(39) trucking companies were surveyed. Of the 39, 34 used speed limiters on
their trucks. The four that did not have limiters were companies that only
hired contract drivers.
4.3.2 Distribution
of Speed Limiter Setting
Figure 42
illustrates the distribution of the speed limiter settings on trucks from a
combination of the surveys from drivers and company personnel. The “No Limiter”
category indicates that the drivers who responded did not have a speed limiter
or governor. Most of these drivers were owner-operators. To see the difference
between the speed limiter distribution for company drivers and owner-operators,
the data were divided into two categories as shown in Figures 43 and 44.
Figures 43 and 44
illustrate that most of the company drivers have a speed limiter set at or
lower than 70 mph, whereas the majority of the owner-operators do not use a
speed limiter or have their speed limiter settings in the high seventies. This
illustrates a major difference in the approach of companies and
owner-operators. The companies believe that they can maximize their profits by
lowering speed to save fuel and maintenance costs. The owner-operators feel
that they can maximize their profits by traveling at a higher speed, and,
therefore, cover more distance in less time. In addition, the owner-operators
do not want to be tied down by the speed limiter on open rural roads in states
like
The owner-operators
can be further divided into two sub-categories: (1) lease drivers (Figure 45)
and (2) independent drivers who operate under their own authority (Figure 48).
The figures illustrate that a higher percentage of independent drivers do not
have speed limiters, while more of the lease drivers have speed limiters set on
their truck. Most of the owner-operators did not have speed limiters. The ones
who had limiters indicated that they used them to reduce the potential of
getting a “very expensive” speeding tickets.
4.3.3 Driver
Experience and Speed Limiter Setting
The scatter plot of
the relationship between driver experience and speed limiter setting is shown
in Figure 47. There does not appear to be a systematic relationship between the
factors. No statistically significant relationship was observed.
4.3.4 Fleet
Size and Speed Limiter Setting
There was a strong
relationship observed between the size of a carrier fleet and the speed limiter
settings. Figure 48 indicates that many of the larger fleets tend to use lower
speeds.
4.4 Opinions of Truck Drivers
The results
obtained from the truck drivers’ surveys and the reasoning offered by the
drivers is summarized below. It should be noted that the surveys are the
opinions of the truck drivers, which may or may not be valid.
4.4.1 Characteristics
of Vehicles and Routes
As
previously discussed, a disproportionate number of drivers who stop at the
truck stops are long haul drivers. Among the truck drivers surveyed, the trip
lengths (home base to home base) of 55.19% of the drivers were more than 7
days, 40.09% of the drivers had trip length between 2 to 7 days, and only 4.72%
of the drivers were out for a single day trip. When vehicles were classified on the basis of the type of cargo,
following
proportions were observed: 55.62% dry vans, 26.04% reefers, 10.65% flat bed,
3.55% tankers and 4.14% miscellaneous tankers (doubles, triples etc). Of the
total, 88.30% of the carriers were truck load, while only 11.70% were less than
truck load (LTL).
Although there are many engine
manufacturers worldwide, only three were observed to be widely used in our
sample: Detroit Diesel (45.11%), Caterpillar (30.98%) and Cummins (17.39%).
Other engine manufacturers, which included Volvo, Mercedes, Mack etc.,
contributed only 6.52% of the engines used. Among the drivers surveyed, 60.81%
of the drivers had 10 speed gears on their trucks, while 27.03% of the drivers
had 13 speed gears on their trucks, and 12.16% of the drivers had others..
4.4.2 Effects
of Vehicle Interactions
The first set of
questions related to the interactions among vehicles and the driver’s
perceptions of the relative risk of different activities. The interaction
between a truck and another vehicle is a critical event during highway driving
for both the truck driver and the other motorist. From a truck driver’s point
of view, there are three critical stages when a truck is passing an automobile:
(1) beginning of pass, (2) traveling parallel and (3) pulling back into the
lane. The relative importance of these three stages depends upon the traffic
conditions, road conditions, driver’s maneuvering technique, and driver’s
perception. The responses are divided into two scenarios: trucks passing
automobiles and then automobiles passing trucks. The truck drivers were relatively evenly
split on their perceptions of which causes more risk: a truck passing an
automobile or an automobile passing a truck (47% and 53%, respectively).
For the maneuver
where a truck passes an automobile, 13% of the truck drivers stated that the
beginning of the maneuver was the most dangerous, 50% felt that driving
parallel was most dangerous, and 37% considered re-entering the right lane the
most dangerous.
Many of the truck
drivers who considered the initial part of passing most dangerous addressed the
perception issue that many automobile drivers dislike driving behind trucks. As
a result, when the truck begins to pass, the automobile drivers often speed up
so that they would not be passed. They subsequently slow down again until the
truck attempts to pass. Furthermore, when the truck tries to shift from the
left lane to the right lane in order to pass a
slower-moving vehicle, other automobiles coming from behind in the left
lane often speed up to restrict the truck’s ability to enter the left lane. The
truck drivers contended that these actions are often the cause of collisions or
near-misses.
Half of the truck
drivers responded that traveling parallel to another vehicle is the most
dangerous period of a passing maneuver. The truck drivers perceive that some
automobile drivers are frightened by the size of a passing truck (Figure 49).
To increase separation, automobile drivers sometimes veer toward the shoulder
of the road and can loose control of their vehicles. Another observation from
the truck drivers was that the automobile drivers sometimes fixate on the
wheels of the passing truck, and they tend to get “sucked into” the truck. This
concentration on the wheels might be related to the fear of the tire tread
separation. During inclement weather conditions (e.g., heavy rain or wind),
both control and visibility of the automobile driver are compromised when being
passed by a truck. Although these events occur both when the truck is passing
the automobile and when the automobile is passing the truck, the perception is
that the effects are exaggerated when the truck passes the automobile.
Figure 49. Impact
of Truck Passing Automobile
For many of the
truck drivers that contended that the time when the truck is pulling back into
the right lane is the most dangerous part of the maneuver, the issue of the
resistance of some automobile drivers to follow a truck was mentioned. It was
contended that this concern sometimes results in the automobile driver speeding
up when being passed by a truck, making it more difficult for the truck to
re-enter the right lane. Visibility of automobiles in a potential “blind spot”
was also cited as a cause of many accidents and near-misses when re-entering
the right lane.
For the maneuver in
which an automobile is passing a truck, 5% of the truck drivers stated that the
beginning of the maneuver was the most dangerous, 53% felt that driving
parallel was most dangerous, and 42% considered re-entering the right lane the
most dangerous. One of the issues stated
by the truck drivers pertaining to the initiation of the passing action was the
misjudgment of the truck speed by the automobile drivers. When the truck is judged to be moving slower
than it actually is, by trying to get around the truck quickly, the motorist
sometimes end up at a very high speed and loses control of their vehicle. When the truck is judged by the motorist to
be faster than it actually is, the result can be that the automobile rear-ends
the truck. This impression of the truck drivers is supported by the accident
data. Researchers have attributed the misjudgment of the truck speed and the
rapid closure rate to the large image projected by the rear of the truck
trailer.
The truck drivers,
who responded that traveling parallel is the most dangerous part of the
maneuver, also cited the same issues that occur when the automobile is passing
the truck (fear of tire separation, veering away from the truck, etc.). The
truck drivers stated that the motorists concern about the wind and reduced
visibility effects that are associated with inclement weather also sometimes
causes motorists to pass trucks at excessive speeds, which increases the risk
of the motorist loosing control.
The 42% percent of
truck drivers for which the period when automobiles are pulling back into the
right lane is the most hazardous part of the maneuver frequently referred to
being “cut-off” by the automobiles (Figure 52). This response of the motorist
is also related to the fact that they often pass with higher than cruising
speed, pull in front of the truck and then apply the brake to reduce their
speed. Another scenario that relates to pulling back into the right lane occurs
when an automobile passes a truck and then immediately needs to reduce speed in
order to enter the exit ramp. Because
the truck is unable to decelerate as fast as an automobile, these activities
sometimes result in rear-end collisions. Similarly, the automobiles that are
behind the truck do not anticipate the truck applying brakes and might hit the
truck from the rear.
One of the outcomes
of lower posted speed limits or speed limiting company policies for trucks is an
increase in the number of vehicle interactions where automobiles pass trucks.
The truck drivers stated the opinion that uniform speed limits significantly
reduce the frequency and risk associated with vehicle interactions.
Eighty-seven percent (87%) of the truck drivers responded that speed
differentials, whether due to regulated speed limits or company policies,
increase the risk of accidents. Ten percent (10%) of the truck drivers stated
the opinion that there is no effect of speed differential limits on
Figure 50.
Illustration of an Automobile “Cutting-off” a Truck
accidents. They
contended that there are advantages and disadvantages that usually cancel out,
with the result being that overall safety would not be affected. The remaining
3% of the truck drivers responded that they felt that having trucks move slower
than automobiles improves safety due to operating and handling differences in
the vehicles (braking distance, maneuverability, etc.).
With respect to the
types of accidents, 43% of the truck drivers stated that speed differentials
increase the probability of side collisions. Fifty-three percent indicated that
side collisions would not be affected and 4% indicated that they would decrease.
There was a general
consensus among truck drivers (76%) that the traffic speed enforcement in
states having differential speed limits is much stricter than in states that
have uniform speed.
4.4.3 Effects
of Speed Differentials at On-Ramps and Off-Ramps
Another safety
issue addressed by the truck drivers is related to the vehicle interaction at
on-ramps. According to the truck drivers, restricted truck speed has a number
of implications at on-ramps. First, slower trucks tend to get “trapped” in the
right hand lane at on-ramps (Figure 51). The inability to move to the left lane
to avoid merging traffic is frustrating to both truck drivers and the merging
motorists. The interaction with merging traffic involves inherent risks that do
not occur when driving in the flow of traffic. When motorists are merging onto
the highway, they often assume that trucks are moving faster than they are. The
result is that the motorist often reduces speed to merge behind the truck. Because the truck is going slower than other
traffic, this causes congestion.
Figure 51. Truck
Interacting with Merging Traffic
The truck drivers also indicated
that another problem related to restricted truck speeds (speed limiters) is the
inability of trucks to reach traffic speed when merging into traffic at
on-ramps (Figure 52). This causes issues for both the truck and the flow of
traffic and potentially increases the risk of accidents on-ramps.
Figure 52. Truck
Merging into Traffic
4.4.4 Effects
of Speed and Speed Differentials on Driver Fatigue
One of the topics of disagreement
between many truck drivers and company management personnel is the impact of
speed on fatigue. It is interesting to note that there does not appear to be
any published literature on the effect of driving speed on fatigue for either
automobiles or trucks. This is the case even though the effect of driving time
has recently received an extensive amount of attention in the context of “hours
of service” regulations. To investigate the truck drivers’ opinions on the
relative effects of “driving time” and “vehicle speed,” they were asked which
situation results in less fatigue: driving 60 mph for 7 hours or driving 70 mph
for 6 hours. In both cases, 420 interstate highway miles would be covered.
Eighty-seven percent (87%) of the truck drivers indicated that driving faster
for a shorter period would result in less fatigue and drowsiness. This response
was potentially confounded by the fact that drivers are usually paid on a per
mile basis. Therefore, driving faster leads to more income per hour.
Many of the drivers
stated that the handling characteristics of the trucks have improved
significantly over the past decade and that driving at higher speeds is not as
tiring as it was previously. However, many of drivers indicated that driving
above 75 mph increases stress and fatigue. Some drivers (13%) felt that driving
70 mph is too fast, takes more energy and increases fatigue. Most of the
drivers stated that, irrespective of their individual speed, driving with the
average traffic speed minimizes the fatigue. They contended that driving either
above or below the traffic speed causes them more fatigue. Some drivers
indicated that although driving significantly below traffic speed reduces the
number of maneuvers (lane changes), it can increase the boredom and can make
them “drowsy”, thus increasing the risk of running off of the road.
Some company
managers contended that drivers who drive at higher speeds (e.g., 70 mph) take
the same amount of time to cover a given distance as is taken by drivers
driving at slower speeds because the drivers of faster vehicles stop more often
for breaks and the breaks are longer. In
response to this question, seventy-one percent (71%) of the drivers stated that
their driving time between each stop is independent of the speed they travel.
This response appears to be related to the fatigue issue previously discussed,
which is based on time, not distance traveled.. Twenty-nine percent (29%) felt
that they take more frequent breaks when they travel at higher speeds.
Approximately half of these drivers indicated that by traveling at higher
speeds, more distance is traveled in less time, and so they can “afford” to
stop more frequently and still make their deliveries on time. This assumes that
the routing schedule uses an artificially low vehicle speed.
4.4.5 Effects
of Speed Limits on Driver Retention
Truck driver
retention is one of the more serious problems currently being faced by the
trucking industry. Operating speed of the company vehicles could be one of the
factors that affect driver retention. Of the surveyed truck drivers, 68% said
that the company’s speed limit policies affect driver retention. They stated
that if companies set the speed limits of their trucks lower, it would
indirectly affect the driver’s paycheck. Because the drivers are often paid per
mile, lower vehicle speed would translate into fewer miles traveled and less
income for the drivers. Lower speed limits also translate into lower pay per
hour and less personal time per mile traveled. The literature has shown that,
for many drivers, personal time can have a larger effect than monetary factors.
However, 32% stated that, as long as they keep getting healthy paychecks, the
company’s speed limit policy does not affect their decision to remain with the company.
4.4.6
Effects of Speed
and Speed Differentials on Operating Costs
The literature
indicates that fuel costs are considered to be the single most significant
factor in the overall operating costs for trucks that are associated with
vehicle speed. To better understand this opinion, the truck drivers were
surveyed about the impact of vehicle speed on the fuel efficiency. Fifty-five
percent (55%) stated that an increase in speed from 60 mph to 70 mph would
decrease the fuel efficiency. Twelve percent (12%) indicated that truck engines
can be tuned and the axle ratio can be set up in a way as to provide best fuel
efficiency higher speeds. There were 11%
who believed that, fuel efficiency would not be affected up to 65 or 70 mph,
however, beyond that fuel efficiency would start decreasing. Twelve percent
(12) contended that modern truck engines are manufactured to provide best fuel
efficiency at speed in the range of 65 mph to 70 mph and that fuel efficiency would
improve as speed is increased from 55 mph to 65 or 70 mph. However they
indicated the opinion that, beyond 70 mph, the fuel efficiency would decrease
for the current engine configurations. Ten percent of the drivers stated that
fuel efficiency would improve with is operating speeds increased from 55 to 75
mph.
As indicated in the
literature review, some trade reports indicate that higher operating speed
increases some of the maintenance costs. The truck drivers were surveyed to
obtain their opinion of the relationship between speed and maintenance costs.
For reference purposes, the drivers were asked to compare the maintenance costs
for 60 mph versus 70 mph. Most of the drivers (64%) stated that, assuming that
the maintenance is done at regular intervals (by mileage), the maintenance
costs are independent of the truck’s speed. Some of the drivers (28%) felt that
higher speeds would cause more wear on the engine and thus increase the
maintenance costs. Only 8% of the drivers thought that operating at 70 mph would
have lower maintenance costs compared to operating at 60 mph.
The effect of speed
on tire wear was another factor that was included in the truck driver survey.
Again, for reference purposes, the drivers were asked to compare the wear
associated with driving 60 mph versus 70 mph. Fifty-one percent (51%) of the
drivers responded that the tire wear would remain the same, while 45% indicated
that the higher speed would increase tire wear. Only 4% suggested that higher
operating speeds would decrease tire wear. The group of drivers, who felt that
tire wear would remain the same, irrespective of the speed, believed that if
the correct tires are chosen for the speed and the correct tire pressure is
maintained, there would not be additional wear at higher speeds. The other
group of drivers, who believed that increasing speed increase tire wear
referred to increased tire heat at higher speeds. The smallest group of
drivers, who thought increasing speed would decrease tire wear, believed that,
for a given distance, reducing the exposure time for the tires would be
beneficial.
4.4.7 Comparison
of Owner-Operator and Company Driver Opinions
As previously
discussed, there is a difference between both the use of speed limiters and the
speed limit setting used by owner-operators and commercial fleets. Speed
limiters are used very little by owner-operators and, when used, they are often
set at higher speeds. The owner-operators have control of the settings, whereas
company drivers do not. The drivers were asked the question, “If you were paid
the same every month, irrespective of the miles traveled, what safe speed would
you drive on rural interstate highways?” The most frequent choice was 70 mph
(see Figure 53). This is probably lower than the general public assumes that
truck drivers would choose.
Figure 53.
Preferred Speed of Travel by Truck Drivers
Figure 54
illustrates that fewer owner-operators indicated that they would prefer the
higher speed of 75 mph than did company drivers. It is interesting that the
individuals that have had the opportunity of driving faster tended to feel that
the lower speed of 70 mph is preferable. One potential reason could be that the
owner-operators have operated at the higher speeds and found that they are not
as safe and efficient. However, the company drivers’ opinions are based on less
experience driving at the high speeds and not being responsible for the
operating costs. It should be emphasized that this particular question did not
address traffic speed or speed differentials between trucks and light vehicles.
Figure 54. Speed
Preferred by Company Drivers and Owner-Operators
There was also a
difference between the company drivers and owner-operators as to the effect of
speed on fuel consumption. Sixty-two percent (62%) of the owner-operators
indicated that operating at higher speeds would reduce fuel efficiency. Only 50% of the company drivers responded
that higher speeds reduced fuel efficiency. It is interesting that a relatively
high proportion of both groups contended that higher speeds do not
significantly reduce fuel efficiency and often prefaced their statement with
the assumption that the truck engine and transmission are intended for the
higher speeds. In particular, it is
interesting that 38% of the owner-operators who paid for their own operating
costs (fuel, tires, insurance, etc.) indicated that traveling at faster speeds
was both safe and efficient. When the
company drivers were asked why thy think companies limit their trucks to lower
speeds, the majority responded that it is due to insurance costs rather than
fuel, tire or maintenance costs.
When asked what
maximum speed limit for automobiles and trucks should be used on flat
interstate highways, 93% indicated that they would prefer a uniform speed
limit, independent of the absolute limit. The highest percentage of drivers
(62%) indicated that the appropriate truck speed should be 70 mph (refer to Figure
57). Of the remaining drivers, 19% indicated 65 mph and 18% indicated a
preference for 75 mph, respectively.
Again, the fact that 82% of the drivers actually prefered to have limits
that are 70 mph or lower is probably not consistent with the driving public’s
assumption of truck drivers’ preferences.
Figure 55. Speed
Limits Preferred by Truck Drivers for Trucks
4.5 Opinions
of Carrier Fleet Safety and Maintenance Management
Nearly all of the commercial
fleets that were surveyed have speed limiters on their company vehicles. The
only exceptions were fleets that only used contract drivers. Most of the fleets
that had both company drivers and contract drivers require speed limiters only
for the company drivers. Most of the companies operated between 62 and 70 mph.
Flatbed, tanker and refrigerated trucks tend to operate at higher speed limits
(70 to 75 mph).
Most of the safety
managers indicated a firm opinion that higher speeds result in a higher
frequency and, particularly, severity of accidents. They also indicated an
opinion that higher speeds increase stress and driver fatigue, with the result
that drivers take more frequent and longer breaks. The contention expressed by
many fleet managers was that, over an extended trip, the total travel time would
be the same for drivers having speed limiters set at 65 mph and 75 mph. When
drivers were questioned about this opinion, they explained that the
management’s opinion might be accurate if the delivery schedules do not
accommodate the higher truck speeds. If the schedule is based on an average
speed that was established from historical data with a lower speed, there is no
benefit for the driver to arrive before the delivery time. The drivers contend
that the additional break frequency and duration is due to excess schedule
time, rather than due to additional stress or fatigue related to the higher
speeds.
Some companies indicated
that they use the speed limiter setting as an incentive for improved safety
and/or fuel efficiency. Drivers are allowed to travel at slightly higher speeds
based on their safety and fuel consumption records. These companies have found that allowing an
increased speed of one to five miles per hour can be an effective reward for
many the drivers.
In some companies, new
drivers are restricted to a lower speed limit than the experienced drivers.
After a period of time, and sometimes based on their safety record, their
operating speed is raised to the company’s nominal limit. The purpose of this
process is to reduce probability of accidents for less experienced drivers.
However, the literature indicates that there might actually be a higher risk of
accidents at speeds that are slower than the traffic speed due to the increased
number of vehicle interactions.
The majority of the
safety managers indicated that they believed that differential speed limits on
highways cause more accidents and all of these managers stated that automobiles
and trucks should operate at uniform speed. The most frequent speed that was
indicated for a uniform limit was 65 mph, although some indicated that 70 mph
would be acceptable. None of the safety managers suggested speed limits higher
than 70 mph.
The
consensus from the maintenance managers surveyed indicated that an increase in
the operating speed of one mph decreases fuel efficiency by 0.08 mpg to 0.1
mpg. This value is much lower than the 0.14 mpg decrease published by The
Maintenance Committee (TMC). One company reported that their fuel efficiency
had actually gone down by only 0.1 mpg after increasing their operating speed
by 3 mph. No conclusions were drawn from these preliminary results.
Regarding the tire
wear, the consensus of the maintenance managers was that tire wear increases beyond
a 65 mph operating speed; although there were no data available to support the
view. One manager indicated that the company had observed no difference in tire
cost between the trucks that operate in states that have a 55 mph speed limit
and those that operate in other states where the company limit of 65 mph
determines the maximum speed. Regarding preventive maintenance costs, the
maintenance managers indicated that, if preventive maintenance is done at
regular intervals on the basis of mileage, higher operating speeds would not
cause more engine wear. None of the companies modified their maintenance
schedules (i.e., oil changes, etc.) based on vehicle speed.
4.6
Opinions of
Original Equipment Manufacturers
In addition to
reviewing published literature and surveying commercial fleet managers,
engineers from the companies that manufacture the trucks, engines, and tires
were surveyed. These surveys consisted of discussions at professional and trade
meetings such as the Technology and Maintenance Council meetings held by the
American Trucking Association, Society of Automotive Engineers Bus and Truck
Meeting, etc. In addition, a number of personal communications by telephone
were used to solicit the opinions of the original equipment manufacturing
company personnel.
4.6.1
Opinions of Engine
Manufacturers
The primary issue
being addressed with this group related to the effect of truck speed on rural
interstates on the engine wear and life of the engine. The effect of changing driving
speed from 60 mph to 70 mph was addressed. One engine manufacturer indicated
that by increasing travel speed from 60 mph to 70 mph, the engine life would be
reduced by 20%. This estimate was based on the opinion that the increased fuel
consumption is directly related to engine life and that the 1987 Maintenance
Council estimate of fuel consumption as a function of speed was still valid.
Two other major engine manufactures both indicated that a change from 60
mph to 70 mph would not have a significant effect on the engine life, as long
as maintenance was performed at the prescribed intervals. None of the
manufactures, including the one that contended that higher speed reduces engine
life, recommends more frequent maintenance (i.e., oil changes, engine rebuild,
etc.) for trucks traveling at higher speeds. This is consistent with the fleet
data that indicated that the maintenance intervals were not affected by the
maximum speed allowed by the different fleets.
Another source of
information that supported that the travel speed does not significantly affect
engine life is that the fleets and owner-operators that purchase used trucks do
not use the speed that the truck was driven in their purchasing decisions. The
only issue considered was that the regular maintenance was performed at the
appropriate times based on the miles traveled.
A critical issue
addressed by all of the engine manufacturers was that the engine configuration
(i.e., horsepower) and the transmission be based on the truck cruise speed. The
size of truck engines being purchased have been increasing. This is supported
by the data collected in the drivers’ surveys that indicated that newer trucks
generally had much larger engines than the older trucks.
4.6.2
Opinions of Tire
Manufacturers
The opinions of the
tire manufactures varied with respect to the effect of truck speed on tire wear
and tire life. One of the manufacturers indicated that there is a significant
increase in tire wear as speed increases. One basis for the opinion was the
increase in tire temperature with increased speed. However, other tire
manufacturers contended that, as long as the correct tire speed rating is used,
the tire material can accommodate the higher speed. In addition, although the
tires are somewhat hotter at higher speeds, they are hot for a shorter period
because the time required to drive a given distance is shorter. These
manufacturers stated that the effect of truck speeds, below 75 mph, is “in the
noise” compared to other factors that affect tire wear and tire life. At higher
speeds (i.e., 75 mph), tire irregularities become more of a problem than tire wear.
At these speeds, the inertia (tire growth) can also become a problem.
One manufacturer cited
that recent, unpublished data indicated that the increase in rolling resistance
of newer commercial truck tires is between 2% to 3% for an increase in speed from
60 mph to 70 mph. This is significantly
below the estimates in the range of 15% provided in other tire and engine
manufacturer documents.
The one area where
there was consensus among all groups, manufacturers, fleet management, and
drivers, is the criticality of maintaining correct tire pressure for the weight
and speed of the truck. There is a large amount of emphasis provided by these
groups, as well as federal agencies, to increase the awareness of the
importance of tire pressure.
4.7 Comparison
of Fleet Experience in States with Different Speed Limits
The accident data were
obtained from the participating companies for the previous three years
(2001-2004). The maximum truck speeds were limited to 62 and 65 mph. By
comparing the experience of the fleet in states that have different automobile
speed limits, the “virtual” speed differential was investigated.
4.7.1 Selection of Accident Data
The accident data
for the companies were obtained for the period of January 2001 through
September 2004. The data were sorted based on the type of road on which the
accident occurred. Although rural interstates are the focus of this report, the
databases were not categorized in this manner. Therefore, the category of
four-lane, divided highway was selected for the analysis.
The data were
sorted to isolate the conditions where the maximum speed could be a determining
factor. For example, sleet, snow, and fog conditions were not included. The
data focused on both clear conditions and rain.
Although rain does impair visibility, it is a condition in which drivers
often maintain their maximum speed.
The data were also sorted
based on the type of accidents. Only those accidents for which the speed of the
vehicle could have been a cause of the accident were chosen. Accidents
associated with other conditions (e.g., mechanical failure, hitting animals,
etc.) were eliminated. Although the
absolute speed of the vehicle has an affect on the risk and severity of these
accidents, they are not directly associated with the issue of speed
differential. Accidents that could not have occurred on the rural interstates
(i.e., pedestrian, overhead obstacles, hit parked, etc.) were also
eliminated. The primary types of
accidents that were included for the purposes of this analysis included: hit by
other, lane change left, lane change right, miscellaneous avoidable, passing,
rear-end (truck hitting automobile), sideswipe-merge and turnover. Although it
would have been beneficial to be able to differentiate accident types such as
automobile rear-ending or automobile sideswiping the truck, all such accidents
were categorized as “hit by other.”
4.7.2 Analyzing Accident
Data by State Speed Limits
The states were
grouped according to their posted speed limits. Some states had uniform limits
(65, 70 and 75 mph) and other states had posted speed differentials (65/55,
65/60, 70/55, 70/60, 70/65, mph). The
“virtual” speed differential for the fleet would be the difference between the
company imposed limit of 62 mph or 65 mph and the posted speed for automobiles.
Therefore, the “virtual” speed differentials for the fleet varied from 0 mph to
13 mph, depending upon the state.
The states were
grouped on the basis of their maximum posted automobile speed limit. The first
group consisted of states having a maximum
automobile speed limit of 65 mph (IL, IN, KY, WI, PA, OR, OH and IA), the
second group consisted of states having a maximum automobile speed limit of 70
mph (CA, AR, MI, WA and MO), and the third group consisted of states having a maximum
automobile speed limit of 75 mph (MT, NM, NV, OK, TX, WY and AZ).
Unfortunately, the
data from the participating companies were not separated by miles traveled on
interstates. Therefore, valid accident rates (per million miles traveled) could
not be calculated. To correct for the fact that different miles were traveled
on rural interstates in various states, the data were normalized by using the
proportion of occurrence for each accident type instead of comparing the
absolute number of accidents. Table 16 illustrates the proportion of each type
of accident within each group.
|
Accident Type |
Group I (65 mph) |
Group II (70 mph) |
Group III (75 mph) |
|
Hit by Other |
48.86 |
54.55 |
52.81 |
|
Lane Change Left |
3.04 |
3.03 |
3.90 |
|
Lane Change Right |
7.15 |
5.39 |
6.93 |
|
Misc. Avoidable |
8.68 |
9.09 |
5.63 |
|
Passing |
0.15 |
0.34 |
0.00 |
|
Rear-end A to B |
10.65 |
8.42 |
9.52 |
|
Sideswipe - Merge |
17.66 |
17.17 |
18.61 |
|
Turnover |
3.81 |
2.02 |
2.60 |
|
Total Accidents |
100 |
100 |
100 |
The only difference
that was statistically significant (p< 0.05) was the “hit by other”
category. The proportion of total
accidents in the “hit by other” category was significantly higher in the 70 mph
states than in the 65 mph states. This
is potentially due to the increased number of interactions in which the other
vehicle must maneuver around the truck.
However, if this were the total explanation, it would be expected that
the proportion for the 75 mph group would have been greater than the 70 mph
group, which it was not.
4.8 Financial
Cost-Benefit Analysis of Operating Speeds
The operating costs
were estimated from a combination of the values from the literature, surveys of
the drivers, and surveys of company maintenance personnel. Although the
specific values would vary somewhat for different organizations, the basic
concept of the cost-benefit analysis would be consistent for different fleets. From
the maintenance data obtained from the participating companies, the fuel consumption
was estimated to be 6.23 mpg at the speed of 62 mph. The estimate of the amount
of fuel efficiency reduction due to increased vehicle speed was based on the literature
review, interviews with fleet operations personnel, and preliminary data from a
participating company. The participating company was evaluating the fuel
consumption on a test fleet on which the speed limiters were set 2 mph above
the rest of the fleet. The decrease in
fuel efficiency estimates ranged from 0.1 mpg per mph (from the Technology and
Maintenance Council), to 0.08 mpg/mph (from surveys of maintenance managers), down
to 0.03 mpg/mph (from a participating company’s preliminary results). The value
for the reduced fuel efficiency for the first analysis was selected to be the
high estimate of 0.1 mpg/mph. With respect to the impact of vehicle speed on
tire wear, the estimates ranged from “no increase” (from some tire
manufacturers and fleets that have vehicles in different speed zones) to a 1%
decrease in tire life for each mph increase in speed (from the Technology and
Maintenance Council and other tire manufacturers). The value of 0.5% for each
mph increase was assumed for this analysis. Based on discussions with
maintenance managers of trucking companies, it was assumed that increased speed
would not have any significant impact on other maintenance costs on a per mile
basis. The price of fuel was assumed to be $2.00 per gallon.
From a survey of
the participating companies and other commercial fleets, the direct variable
costs associated with vehicle speed were estimated. The context for the
analysis is long-haul operations on rural interstates. It was determined that
the direct costs, independent of the drivers’ pay, was 29.3% of total revenue.
The breakdown by category was as follows:
Cost Category Percentage Revenue
Fuel 15.4 %
Tires 1.6 %
Maintenance Costs 4.3 %
Profit 8.0 %
The number of miles
traveled per truck, per year was estimated to be 130,000 for the purposes of
this analysis. This value is somewhat higher than some companies and is lower
than the average annual miles traveled by many owner-operators, based on the
surveys during this study. The breakdown of costs for the base speed of 65 and
70 mph are shown in Table 17.
For this scenario
comparison, the increase in revenue per truck ($20,846) is less than the
increase in incremental operating costs due to the higher speed. This would
result in a net reduction in profit of $2,371 per truck. However, the driver’s
pay increased by $3,200 due to the increase in total miles. To the extent that
the additional wages improve driver retention, the reduction in the costs
required to replace drivers might offset the decrease in profit per truck. The
cost of replacing a driver is approximately $5,000 to $8,000.
With more modern
fleets that have electronically controlled engines, more effective aerodynamics
and higher horsepower engines, the additional cost per mile in fuel is more
likely to be .05 mph per mph. In this case, the annual reduction in profit per
truck would be $328. If the lower estimate
of .03 mpg/mph is used, based on the preliminary
Table 17. Per-Truck
Cost Analysis for 0.1 mpg/mph Fuel Efficiency Loss
|
Speed (mph) |
65 |
70 |
|
|
|
|
|
|
|
Fuel cost ($/gallon) |
2.00 |
|
|
|
Fuel (mpg) |
6.23 |
5.73 |
0.10
mpg/mph |
|
Tires (% of total revenue) |
1.60 |
1.64000 |
0.50%
/ mph |
|
Maintenance (% of total revenue) |
4.30 |
4.30 |
|
|
Driver ($/mile) |
$
0.32 |
$
0.32 |
|
|
|
|
|
|
|
Annual Miles |
130000 |
140000 |
|
|
Total Revenue |
$270,997 |
$291,843 |
|
|
Gallons consumed |
20867 |
24433 |
|
|
Fuel cost |
$41,734 |
$48,866 |
|
|
Tire cost |
$4,336 |
$4,786 |
|
|
Maintenance cost |
$11,653 |
$12,549 |
|
|
Drivers Pay |
$41,600 |
$44,800 |
|
|
Other cost (70.7%) |
$191,595 |
$206,333 |
|
|
Operating Revenue |
$249,317 |
$272,534 |
|
|
|
|
|
|
|
Profit (8%) |
$21,680 |
$19,309 |
Reduction of $2,371 |
Table18. Per-Truck
Cost Analysis for 0.05 mph/mph Fuel Efficiency Loss
|
Speed (mph) |
65 |
70 |
|
|
|
|
|
|
|
Fuel cost ($/gallon) |
2.00 |
|
|
|
Fuel (mpg) |
6.23 |
6.08 |
0.03 mpg/mph |
|
Tires (% of total revenue) |
1.60 |
1.64 |
0.50% / mpg |
|
Maintenance (% of total revenue) |
4.30 |
4.30 |
|
|
Driver ($/mile) |
$ 0.32 |
$ 0.32 |
|
|
|
|
|
|
|
Annual Miles |
130,000 |
140,000 |
|
|
Total Revenue |
$270,997 |
$291,843 |
|
|
Gallons consumed |
20,867 |
23,026 |
|
|
Fuel cost |
$41,734 |
$46,053 |
|
|
Tire cost |
$4,336 |
$4,786 |
|
|
Maintenance cost |
$11,653 |
$12,549 |
|
|
Drivers Pay |
$41,600 |
$44,800 |
|
|
Other cost (70.7%) |
$191,595 |
$206,333 |
|
|
Operating Revenue |
$249,317 |
$269,721 |
|
|
|
|
|
|
|
Profit (8%) |
$21,680 |
$22,122 |
Increase of $442 |
empirical fleet
data, there would actually be an annual increase profit gain of $442 per truck
by changing the company speed policy or the posted truck speed limit from 65
mph to 70 mph. The assumptions used in this analysis are obviously not
representative of all trucking operations under all conditions. For example,
the effective cost of fuel (accounting for surcharges) has a large effect on
the costs associated with the reduced fuel efficiency. In addition, the
assumption was that the trucks are long-haul operations that are always on
interstate highways. However, for that portion of a fleet’s operations that are
spent on rural interstates, this type of analysis should apply.
5. Discussion
This study
addressed the safety and financial costs and benefits of higher speed limits
and of speed differentials between large trucks and other vehicles on rural
interstate highways. This section of the report presents conclusions drawn
from: (a) review and analysis of existing literature, (b) collection and
analysis of speed, accident, and maintenance data, and (c) analysis of opinions
of various stakeholders: truck drivers, safety and maintenance managers of
companies, and original equipment manufacturers of trucks, tires, and engines.
5.1 Summary of Research on Truck Speed Effects on
Traffic Flow and Safety
5.1.1 Impact of Speed Limits on Traffic Speed
Increases and
decreases of the posted speed limits have been found to affect traffic speeds
to various degrees by different studies. The concept of “design speed,” often
defined in terms of the 85th percentile traffic speed, is frequently
discussed in the context of setting speed limits. Although this concept has
been shown to be useful for two-lane roadways with complex geometries, it does
not appear to be applicable for four-lane rural interstate highways. The 85th
percentile speed of unrestricted traffic on rural interstates would be much
higher than the limits that are generally considered to be acceptable. One of
the reasons that studies have observed a large amount of variation in traffic
speeds on highways with the same physical characteristics has been the level of
enforcement. If speed limits are not strictly enforced, motorists choose their
own “comfortable” operating speeds.
A factor that has affected
the observed increase in the traffic speed when limits have been raised has
been the time frame over which the data are collected. The change in traffic
speed after a limit change is characterized by two stages, an initial
transition phase and, subsequently, an adaptation phase. During the
“transition” phase, only a few motorists increase their speeds immediately up
to or above the new speed limits. The adaptation phase begins when the
motorists become comfortable with the higher traffic speeds and increase their
speed. If the magnitude of increase in the average speed is calculated soon
after increasing the speed limit (during transition), the increase in the
average speed is lower than if it is measured later, after the adaptation
phase. Another important aspect of the transition phase is that the speed
variance (distribution of vehicle speed in the traffic flow) is higher than it
is after the adaptation phase. This speed variation has important safety
implications which will be discussed later in this section.
An important issue that
previous studies in the research literature have not addressed is the traffic
mix of heavy trucks and light vehicles when investigating the relationship
between speed limits and traffic speed. As illustrated in this report, the
speeds of many (if not most) trucks are limited to below posted speeds by
engine speed limiters. Since large trucks constitute a significant portion (15
to 45%) of rural interstate traffic, an increase in the posted limit of 10 mph
does not produce the same amount of increase in the mean traffic speed. The
level from which the speed limit was raised (from 55 to 65 mph or from 65 to 75
mph) was also been found to affect the amount of increase in the mean speed. An
increase in the speed limit from 55 to 65 mph on rural interstate highways increases
the mean traffic speed by 3 to 6 mph; whereas, an increase in the speed limit
from 65 mph to 75 mph increases mean speed by only 2 to 4 mph. One reason for
this is that most trucks can increase their speed from 55 to 65 mph; however, a
significant portion of the trucks can not increase from 65 to 75 mph.
The speed data
collected during this study illustrated that, although the posted speed limits
for automobiles differed by 10 mph (65 versus 75 mph), the mean speeds differed
by only 1.6 mph (73.2 to 74.8 mph). The posted speed limits for heavy trucks
had a larger effect. The 15 mph
difference in posted limits for trucks (55 versus 70 mph) resulted in mean
truck speeds that differed by 4.4 mph (64.2 and 68.6, respectively). These data
support the research literature that has frequently indicated that motorists
tend to drive at a speed with which they are comfortable, regardless of the
posted limits.
Even when the
posted speed limits are the same for heavy trucks and automobiles (uniform
limits), the average speed of trucks is 3 to 4 mph slower than the average
speed of automobiles. This is primarily due to the fact that most trucks have
speed limiters that restrict their speed. However, the truck drivers contend
that it is also the result of different levels of enforcement for heavy trucks
and automobiles. The compliance rates differed significantly for the four speed
limit configurations studied during this effort. The compliance rate for the
highest, uniform limits (75/75 mph) were 53% and 73% for automobiles and
trucks, respectively. However the compliance rates for the lower differential in
speed limits (65/55 mph) were 7% and 0%, for automobiles and trucks,
respectively. This supports the contention in the literature that, if the
limits are set at what is considered to be arbitrarily low values, motorists
will not adhere to the limit.
5.1.2 Impact of Speed Limits on Rural
The fact that sections of interstate highways with virtually identical
physical characteristics have very different speed limits in different states
illustrates that there are many factors unrelated to the roadway and traffic that
affect the setting of speed limits. For similar rural interstate highways, the
speed limits range from 65 to 75 mph for automobiles and from 55 to 75 for
heavy trucks. A good of a dramatic and immediate change in speed limit occurs
when crossing the Nevada-California state line on Interstate I-15. The speed
limit for heavy trucks decreases by 20 mph (from 75 to 55 mph), although the
roadway does not change at that point.
The large number of safety
studies that were discussed in the Literature Review indicates that this issue
has received a great amount of attention. Unfortunately, many of the studies
involve more advocacy than science. One section of this report addresses the
methodological issues associated with much of the research on the relationship
between speed limits and highway safety. For example, the studies that analyzed
the number of fatalities during the transition periods immediately after speed
limits were increased often found very large increases in the number of
fatalities. However, other studies that measured fatality rates or accident
rates over a longer time frame often concluded that there was little or no
negative impact of the speed limit increases. Similarly, many sources in the
popular press refer to the statistics that indicate that more than one-third of
the highway accidents are associated with “speeding.” However, speeding is
defined as “traveling faster than the posted limits” or “traveling too fast for
conditions.” Because there is no differentiation of these two categories in
much of the literature, the effect of the posted speed limits on the number of accidents
and fatalities is probably highly exaggerated in the popular literature.
5.1.3 Causes
and Impact of Speed Variance
Although there is a
large amount of controversy over the magnitude of the effect that increases in posted
speed limits have on highway safety, there is a relatively strong consensus
among both researchers and practitioners that a higher variance of vehicle speeds
in the traffic flow increases the risk of accidents. This relates to the
intuitive argument that the more interactions there are among vehicles, the
higher the probability of a collision event occurring. Even when the traffic
density is high, traveling on an interstate highway without passing or being
passed would involve fewer opportunities for two-vehicle collisions than if the
variation in vehicle speeds is high.
Various factors that
affect traffic speed variance are enforcement, the design speed of the highway,
and the percentage of trucks among traffic. High enforcement results in the
reduction of the number of motorists traveling at excessively high speeds,
which results in lower speed variance among vehicles. If the speed limit is set
far below the effective design speed of the highway, some motorists will adhere
to the limits, but most will choose a higher speed at which they feel
comfortable. This will increase the speed variation among vehicles. From the
traffic speed measurements taken during this study, it was observed that the
rural interstate with a posted speed limit of 65 mph had much more “speeding”
(i.e., low compliance) than was observed for the interstates with higher limits.
Another characteristic of speed limits that
increases the speed variance is differential speed limits. If the posted limit
for automobiles is higher than for heavy trucks, there will naturally be more
variation in vehicle speeds. Company policies that restrict the maximum speed
of their fleet with limiters on the engines also increase the amount of speed variance
on interstate highways. As the proportion of trucks on a highway increases, the
amount of speed variance increases.
Changes in posted
speed limits also affect the speed variance. During the transition period, some
drivers adapt slowly to the higher limits while others immediately travel at or
above the new limit. This temporary behavioral difference of these two groups increases
the amount of speed variance. This
phenomenon has been cited as being a potential confounding factor when
investigating the impact of increased speed limits on the number of accidents
or fatalities. If the safety data for the transition period are used as the
basis of comparison, the conclusion could be that there is a large negative
impact of increased speed when, in fact, the increase in accidents could be
due, at least in part, to the increased
speed variance.
When the effect of increased
speed limits on speed variance of automobiles and trucks were studied individually,
different trends were observed for the two vehicle categories. Speed variance
among automobiles decreased with increased speed limits. For trucks, increasing
speed limits up to 65 mph resulted in reduced speed variance. However, increases
in speed limits beyond 65 mph increased speed variance among trucks. Higher
speed limits tend to divide truck traffic in two parts: one consisting mainly
of owner-operators, who can travel at higher speeds, and the other consisting
mainly of company drivers who can not travel at higher speeds due to the use of
speed limiters. Results of the traffic speed measurements collected during this
study support these conclusions. It was observed that the speed variance among
automobiles on highways with 65, 70, and 75 mph speed limits decreased (5.67,
4.95, and 4.61 mph, respectively); whereas, the speed variance among trucks on the
same 65, 70, and 75 mph speed limit highways increased (3.69, 4.55,
and 5.63 mph, respectively).
With respect to
speed variation, most of the studies that analyzed the effect of vehicle speed
on the risk for an individual vehicle concluded that the probability of being
involved in a crash follows a U-shape curve as a function. The risk increased
for both vehicles going faster and slower than the traffic speed with the
minimum value being at or slightly above the mean speed of traffic. A computer
simulation used in this study indicated that, for the interstate with posted differential
speed limits of 65/55 mph, the number of interactions for a truck traveling at
the speed limit (55 mph) would be more than four times the number of
interactions for a truck traveling at mean traffic speed.
One of the common
misconceptions that motorists have is that they are often passed by trucks. However,
results of the simulation study indicated that the frequency of automobiles
being passed by trucks is very low. Using the traffic speed data from the
uniform 70 mph sites, an automobile traveling at the mean traffic speed (71.5
mph) would be passed by only 30 trucks during a 1000 mile trip on a rural
interstate.
5.1.4. Impact of
Speed on Crash Severity
Most of the studies
in the research literature have concluded that the severity of an accident
increased with increased speed. Although the improvements in passive safety
systems, such as seatbelts, airbags, and vehicle crash worthiness, have reduced
the impact of speed on severity, basic physics indicates that a crash at higher
vehicle velocities results in higher impact forces. This is particularly the
case for heavily loaded trucks. The
difference in braking distance between automobiles and heavy trucks is also
affected by the speed of the vehicles. Although recent advances and projected
future improvements in brake technology for trucks is reducing the brake
distance differential, this is one of the most valid reasons for restricting
truck speeds to lower than automobile speeds. One misconception that is often
cited in the popular literature relates to the relationship between truck
weight and braking distance. Due to the increased normal forces on the roadway
surface, the braking distance for a fully loaded truck is not higher than for
an empty truck.
The relationship
between speed and crash severity is one of the reasons that research studies
that use the number or rate of fatalities, rather than accidents indicate a much
higher impact of higher speeds on highway safety. Even when the number of
accidents does not increase, or even when the number decreases, the number of
fatalities can increase because the accidents, when they do occur, are more
severe.
5.1.5 Impact of Differential Speed Limits on
Highway Safety
The fact that public policy makers
have come to different conclusions about the efficacy of speed differentials is
illustrated by the fact that states have adopted speed limits that range from a
15 mph differential to uniform limits for both automobiles and heavy
trucks. Although there have been a
number of studies that have investigated the safety implications of posted
speed differentials between automobiles and heavy trucks, the results have been
inconclusive. The studies have either compared data from states that have
different configurations (Differential Speed Limits, DSL, or Uniform Speed
Limits, USL) or data for states that changed from one configuration to the
other. A representative conclusion is from the Federal Highway Administration’s
Technical Report (FHWA-HRT-04-126, 2004) states that: “Overall, the study was
not able to isolate or measure the effect of USL/DSL changes. The effect of the
DSL, if any, is not enough to be detected in the aggregate speed data that were
analyzed.”
One very important factor that has
not been addressed by the research studies that have investigated posted speed
differentials between automobiles and heavy trucks was the impact of speed
limiters that are installed on most commercial trucks. To the extent that this
resulted in an effective differential, even for states that had uniform speed
limits, the studies were inherently flawed. This is one of the reasons that the
various studies have found differing results.
Proponents of lower
truck speed limits cite the fact that trucks require longer braking distances
for any given speed and lower truck speeds help equalize the stopping distance.
Truck drivers contend that their higher seat position allows a longer site
distance (multiple vehicles forward), reducing the effect of the differences in
braking distance. Opponents of lower
truck speed limits have suggested that the differential speeds increase the
speed variance and, therefore, has a negative impact on highway safety. It is
likely that both of these arguments are correct. This would indicate that differential
speed limits have two effects: (1) the positive effect that results from improved
vehicle dynamics (braking and maneuvering) for trucks at lower speeds; and (2) the
negative effect of increasing speed variation and the number of interactions
among vehicles. These two effects of
differential speed limits act in opposite directions and ultimately result in
no observable effect on highway safety data.
When the truck drivers
were asked for their opinions of speed differentials, most stated that differential speed limits increase interactions among
vehicles and increase the probability of rear-end, side-swipe, and on-ramp
accidents. Two scenarios that dominated the drivers’ concerns were associated
with on-ramps. The first safety issue related to trucks being “trapped” in the
right lane and the increased risk of continually encountering merging traffic.
The second issue involved trucks not being able to reach traffic speed when
merging into traffic flow. They also indicated a concern that lower truck
speeds result in congestion and clustering of traffic and bottleneck situations
on highways. The majority of the truck drivers indicated that a uniform speed
limit of 70 mph for both automobiles and trucks would be both the safest and
the most efficient configuration for rural interstate highways. It was
interesting to note that the drivers that generally have the ability to travel
faster than 70 mph (owner-operators) also agreed that a 70 mph limit would be most
appropriate.
Most of the company safety managers who were surveyed also expressed the
opinion that differential speed limits increase the probability of accidents on
rural interstate highways. However, many of the safety managers felt that a
uniform limit of 65 mph would be the best alternative. Some managers indicated
that new, less experienced drivers might benefit more from lower truck speeds,
with more experienced drivers being able to handle the higher speeds. Other
managers indicated that this policy would put less experienced drivers at
additional risk due to the increase in the number of vehicle interactions that
they would experience. The effect of
company policies that restrict maximum speeds does not appear to affect the
insurance premiums paid. From discussions with insurance carriers, it was
determined that only the company’s experience ratings were considered and that
the company’s speed policies were not included in the rate-setting process.
5.1.6 Effect of
Speed on Driver Fatigue
Fatigue is a
contributing factor in as many as 30-40% of all heavy truck crashes. Although
research has been conducted to study the factors causing truck driver fatigue,
there is no empirical data indicating that increased speed increases fatigue.
However, there are studies that have found that operating time has significant
impact on truck driver fatigue. The relationship among of driving time, fatigue,
and accident risk has been extensively documented in the context of the recent
changes in truck driver “hours-of-service” regulations. One of the methods of
reducing driving time and fatigue without reducing transport efficiency or
driver pay, would be to travel at a higher speed. From an hours-of-service
perspective, an important issue is whether it would be safer to drive for 10
hours at 70 mph than it would be to drive for 11 hours at 64 mph.
When the truck
drivers were surveyed about their opinions on fatigue, most of them stated that
driving faster for a shorter duration of time would result in less fatigue and
drowsiness. In addition, the consensus of drivers was that driving at the
average traffic speed reduces fatigue.
Most of the company
safety managers indicated the opinion that traveling at higher speeds results
in more fatigue. A comment frequently expressed by managers was that, even when
drivers are allowed to use higher speeds, they do not get to their destinations
sooner because they stop more frequently and take longer breaks. However, most
of the truck drivers stated that their driving time between each stop is
independent of the speed they travel and that their stops are based on time
rather than distance. The drivers did indicate that, if the scheduling of the
delivery time is not adjusted for the higher speed, then there is no benefit in
getting to the destination early. In this case, they would distribute their
time rather than waiting at the destination. However, they indicated a
preference for getting to the destination sooner if the delivery schedule was
adjusted for the higher speeds.
5.2 Effect of Speed on Operational Costs
5.2.1 Effect of Speed on Fuel Efficiency
One of the primary reasons for commercial trucking
firms limiting the maximum speed of their trucks is the reduction in fuel
consumption which is the highest operational cost per truck. The rule of thumb
provided by the trucking trade organization, The Maintenance Council (now the
Maintenance and Technology Council), and
some of the engine manufacturers is that each increase in one mph of speed above
55 mph will decrease the fuel efficiency by 0.1 mpg. However, this estimate is
based on studies that were conducted nearly 20 years ago. The engines,
electronic controls, aerodynamics, etc. are very different for trucks being
purchased today. The survey of maintenance and operations managers indicated
that a more accurate estimate for current fleets is probably 0.08 mph for each
mile per hour increase in speed. Some recent, unpublished data, indicate that,
for rural interstates, the cost of increased speed is 0.03 to 0.05 mpg per mile
per hour increase.
In addition to the absolute vehicle speed, speed
variance in the traffic flow also has an effect on fuel efficiency when both
trucks and automobiles decelerate and accelerate to maneuver around slower traffic.
As illustrated by the computer simulation in this study, speed differentials
significantly increase the number of interactions among vehicles. The negative
impact of traffic speed variation on fuel efficiency has not been addressed in
the research literature or as a policy issue.
When speed policies are considered, it is important to
consider that the driver effect is estimated to be double the effect of vehicle
speed. It might be possible that by improving retention, the costs associated
with higher speeds might, to some extent, be offset by the ability of more
experienced drivers to conserve fuel.
The survey of the truck
drivers indicated that they agreed that speeds beyond 65 mph decrease fuel
efficiency. The drivers tended to focus on the impact of the appropriate truck configuration
(engine, transmission, etc.) if higher speeds are used. It is interesting to
note that the owner-operator drivers, who have direct knowledge of their
individual operating costs, acknowledge the additional fuel cost associated
with higher speeds; however, as a group, they preferred higher speeds due to
the increased revenue, more flexible scheduling, and the benefits of increased
personal time.
5.2.2 Effect of Speed on Roadway Wear
Although the
literature search was extensive, no study that specifically addressed the
impact of heavy truck speed on the required maintenance of limited-access
highways was found. The basic laboratory research indicated that an increase in
the operating speed of the truck would not increase roadway surface stress. The
consensus of the researchers surveyed indicated that, to the extent that there
was an effect, it would be that higher speeds reduce pavement wear based on the
fact that the forces are exerted on individual segments of the roadway for a
shorter period. Another widely held
consensus was that, as speed variability increases, the increased level of vehicle
maneuvering, braking, acceleration, and deceleration would increase the amount
of wear on the roadway.
5.2.3 Effect of Speed on Tire Costs
There was no
objective research data found in the public literature that related to the
effect of speed on tire wear at the speeds appropriate for rural interstates.
In the survey, some of the tire manufacturers indicated that a truck speed change
from 65 to 75 mph reduced the tire life. This estimate was as high as a 1%
reduction in tire life for each additional 1 mph. The primary reason for the reduction was reported
to be the increased tire temperatures at higher speeds. The higher temperatures
affect the number of times the casings can be retreaded. Other manufacturers
stated that, as long as the correct tire rating was used and the pressure was
appropriate for the load and speed, the amount of additional tire wear
associated with the higher speed would be negligible. With respect to tire temperature, these
manufacturers indicated that, although the tires were hotter, the materials
were adequate to accommodate those temperatures and the exposure time during
which they were hot was actually lower on a per-mile basis. However, there is
no objectively verifiable data available to check the validity of either of
these opinions, although one manufacturer had preliminary data that indicated
that tire speed was relatively unimportant relative to the other factors (i.e.,
correct pressure).
The majority of the maintenance
managers surveyed indicated that tire wear increases beyond a 65 mph operating
speed. One of the participating companies indicated that they had observed no
significant difference in tire cost between the trucks that operate in states
that have a 55 mph speed limit and those that operate in other states where the
company limit of 65 mph determines the maximum speed. Most of the truck drivers
surveyed expressed that correct tire selection and tire pressure have a much more
significant impact on tire wear than the operating speed.
5.2.4 Effect of Speed on Engine Life and Routine Maintenance
Costs
With respect to the
effect of higher truck speeds on engine life, the opinions of the manufacturers
were again split. The estimates of the additional engine wear ranged from no
effect to a 20 % reduction in engine life for a truck with a 70 mph operating
speed compared to a 60 mph speed. As with estimates for other operating costs,
the configuration of the truck (engine, transmission, etc.) is important. If
the vehicle is not configured for higher speeds (i.e., low horsepower, wrong
gear ratios, insufficient cooling system, etc.) engine wear can increase
significantly at higher speeds.
None of the engine manufactures,
including the one that contended that traveling at higher speeds reduces engine
life, recommended more frequent maintenance intervals on a mileage basis for
trucks traveling at higher speeds. This is consistent with the company and
driver survey data that indicated that the maintenance intervals were not
affected by the maximum speed allowed by different fleets. Another point to
note is that fleets and owner-operators that purchase used trucks do not use
the speed at which the truck traveled in their purchasing decisions; rather
they are only concerned with the fact that maintenance was performed at the
appropriate intervals based on the number of miles traveled.
5.3 Financial Cost-Benefit Analysis of
Operating Speeds
The financial
cost-benefit analysis illustrated how the results are very sensitive to
estimates of the operational costs associated with increased truck speed.
Unfortunately, although there is are many opinions, there is very little
verifiable data that can be used to make these estimates. Therefore, the
combination of the literature, survey results, and participating company data
were used to derive estimates for the analysis. The analysis used estimates of
the increased revenue that could result from higher speeds on rural interstates
and estimates of the costs associated with those higher speeds. The results
ranged from an annual decrease in net profit per truck of $2,371, for the higher
estimates of speed-related operational costs to a net profit increase of $442
for the lower estimates. Even the costs derived using the higher estimates
could be offset, to some extent, if the higher speeds and increased pay would
improve driver retention. In addition, the number of trucks necessary for the
same annual mileage would be reduced, lowering the truck inventory costs for
commercial fleets.
5.4 Conclusions
The focus of the study was on absolute
and differential speed limits for heavy trucks on rural interstate highways. Although
there is an abundance of opinion on many
of the issues, there is very little empirical, verifiable, and
scientifically valid data available from either public or private sources. The
current effort assessed the research and applications literature, measured
traffic flow under different speed limit configurations, and surveyed the
stakeholders that were affected by the policies. The object of the stakeholder
surveys was to obtain their opinions and, more importantly, the basis for those
opinions. It is evident that there is a need for additional research in many of
the areas relevant to the maximum speed for heavy trucks. The data from the Large Truck Crash Causation Study should provide better detailed
information that could assist in evaluating the safety implications of speed
differentials between automobiles and heavy trucks. To satisfactorily address
the issue, additional current and valid information is required about the
operational costs of higher truck speeds that apply to both trucking operations
and the general public.
The decisions pertaining to the
state regulated absolute and/or differential speed limits for trucks will
continue to be a political, as well as a technical issue. Similarly, the policy
decisions of commercial trucking organizations related to maximum truck speeds involve
many factors beyond those addressed in this study. The objective of this effort
was to provide information that both regulatory agencies and trucking
operations could use when making decisions related to maximum truck speeds, in
general, and speed differentials between automobiles and heavy trucks, in
particular.
6.
References
Abelson,
P.
(2002). Isn’t it Time to Bring NHTSA’s
Standards for Brakes in Line with Today’s Technology? Land Line Magazine.
Abraham, J.M. and Abdulhai, B. (2001), Analysis of Highway Speed Limits,
Department of Civil Engineering, University of Toronto.
Addis, G. (1999). To speed.…Or not to Speed. Expedite.com Magazine.
Advocates for Highway and Auto Safety. (1995). States at Risk: Repealing the National
Maximum Speed Limit Means More Deaths, Injuries, and Costs to Society.
National Transportation Library.
Advocates for Highway and Auto Safety. (2001).
Truck Driver Fatigue.
http://www.saferoads.org/issues/fs-truckdriverfatigue.htm
Agent, K.R., Pigman, J.G. and Webber, J.M.
(1998). Evaluation of Speed Limits in
Agent, K.R., Pigman, J.G. and Weber, J.M.
(1998). Evaluation of Speed Limits in
Akram, T., Scullion, T. and Smith, R.E. (1993) Using
the Multidepth Deflectometer to Study Tire Pressure, Tire Type, and Load
Effects on Pavements, Research Report 1184-2 Vol. 2. Project No. 1184.
Texas Transportation Institute,
American Association of
American Association of
Ang, B.W., Fwa, T.F. and Poh, C.K. (1991).
Statistical Study on Automobile Fuel Consumption. Energy, 16 (8).
Ang-Olson, J. and Schroeer, W. (2003). Energy
Efficiency Strategies for Freight Trucking: Potential Impact on Fuel Use and
Greenhouse Gas Emissions, Journal of the
Transportation Research Board No. 1815.
Ashenfelter, O. and Greenstone, M. (2002) Using Mandated Speed Limits to Measure the
Value of a Statistical Life. Journal of Political Economy, 112
(1), 2.
Atkinson, K. (1996). States’ Attitude Towards Speed Limits: Summary Chart. Reasonable
Drivers Unanimous. (http: //sunsite.unc.edu/rdu).
Balkin, S. and Ord, J.K. (2001). Assessing the Impact of Speed Limit
Increases on Fatal Interstate Crashes. Journal of Transportation and
Statistics. April 2001, 1-26.
Bamfield, J.D. (1989). Accidents Before and After the 65 mph Speed Limit in
Banasiak, D. (1997). Speed Limits Up, But Deaths
are Down. Roads and Bridges. 35 (6),
10-11.
Bartle, S.T., Baldwin, S.T.,
Baum, H.M., Esterlitz, J.R. and Zador, P., and
Penny, M. (1991). Different Speed Limits for Cars and Trucks: Do They Affect Vehicle
Speeds. Transportation Research Records
1318, 3-7.
Baum, H.M., Lund,
A.K. and Wells, J.K. (1989). The Mortality Consequences of Raising the Speed Limit to 65
mph on Rural Interstates. American
Journal of Public Health, 79, 1392-1395.
Baum, H.M., Lund,
A.K. and Wells, J.K. (1990). Motor Vehicle Crash Fatalities in the Second Year of 65 MPH
Speed Limits. Journal of Safety Research,
21, 1-8.
Baum, H.M., Wells,
J.K. and Lund, A.K. (1991). The Fatality Consequences of the 65 mph Speed Limits, 1989. Journal of Safety Research, 22,171-177.
Baxter, J.J. (1999) The Mythology of Setting Low
Speed Limits. Consumer’s Research. April, 26-27.
Bedard, P. (1996). Why We Need More Speed and
Less Frank J. Wilson. Car and Driver. 42 (1), 24-43.
Binkowski, S.E., Maleck, T.L.,
Blower, D. and Campbell, K.L. (2002). The Large Truck Crash Causation Study.
UMTRI-2002-31. Transportation Research Institute,
Borsje, J.F. (1995). The Effects of the 1988 Differentiation of Speed Limits in the
Brackett, R.Q. and Ball, K. (1990). The Safety Impact of the 65 mph Speed Limit
in
Bridgestone/Firestone Commercial Truck Tires.
(2004). Guide to Large Truck Fuel Economy
for a New Millennium.
Broderick, A. (1975). Fuel Consumption of Tractor-Trailer Trucks as Affected by Speed Limit
and Payload Weight. U.S Department of Transportation (DOT).
Brown, D.B., Maghsoodloo, S. and
Cameron, M. (2003). Potential Benefits and Costs of Speed Changes on Rural Roads.
Australian Transport Safety Bureau CR 216.
Cameron, M., Newstead, S., and Vulcan, P.
(1994). Analysis of Reductions in
Carroll, R.J (2004), Impact of
Local/Short Haul Operations on Driver Fatigue: Focus Group Summary and Analysis, TechBrief
FHWA-MCRT-99-002,
Carroll, R.J (2004), Light Vehicle- Heavy Vehicle Interactions: A Preliminary Assessment
Using Critical Incident Analysis, Federal Motor Carrier Safety
Administration, http://www.fmcsa.dot.gov/safetyprogs/research/briefs/lv-hv_inteeractions_tech_brief.htm
Casey, S.M. and
Cerelli, E.C. (1981). Safety Consequences of Raising the National Speed Limit from 55 mph to
60 mph (
Chang, G.L., Carter, E.C. and Chen, C.H. (1993).
Intervention Analysis for the Impacts of the 65 mph Speed Limits on Rural
Cirillo, J.A. (1968) Interstate System Accident
Research Study II, Interim Report II, Public
Roads, 35 (3), 71-75.
Cirillo, J.A. (2003) Testimony of Julie Anna
Cirillo, Senate Highways and Transportation Committee, Senate Bill
94, Owner Operator Independent Drivers Association.
Coesel, N. and Rietveld, P. (1998). Time to Tame Our Speed? Costs, Benefits and
Acceptance of Lower Speed Limits. Proceedings, 9th International
Conference, Road Safety in Europe,
Coffman, Z., Stuster, J. and
Warren, D. (1998). Synthesis of Safety
Research Related to Speed and Speed Management, FHWA-RD-98-154.
Cooper, K.R. (2003). Truck
Aerodynamics Reborn- Lessons from the Past, SAE 2003-01-3376, 2003 SAE
International Truck and Bus Meeting and Exhibition,
Council, F.M., Harkey, D.L., Nabors, D.T.,
Khattak, A.J., and Mohamedshah, Y.M. (2003). Examination of Fault, Unsafe
Driving Acts and Total Harm in Car-Truck Collisions. Transportation Research Record, 1830, 63-71.
Craft, R. (2002). Rear-End Large Truck Crashes, Federal Motor Carrier Safety Association,
http://www.fmcsa.dot.gov/safetyprogs/research/briefs/rear.pdf
Cummins (2003). Secrets of Better Fuel Economy-
The Physics of MPG, Cummins MPG Guide.
Deierlein, B. (2000). Managing fuel consumption,
Fleet Equipment
den Tonkelaar,
W.A.M. (1994) . Effects of Motorway Speed Limits on Fuel Consumption and
Emissions. The Science of the total
Environment, 146/147, 201-207.
Ding, Y. (2000). Quantifying the Impact of Traffic-Related and Driver-Related Factors on
Vehicle Fuel Consumption and Emissions. MS Thesis, Civil and Environmental
Engineering, Virginia Polytechnic Institute and
Donald, D. and Cairney, P. (1997). Higher
Dornsife, C. (2001). Fatal Accidents Double on Montana’s Interstates, National Motorist
Association, Waunakee
E.H. Pechan and Associates, Inc. (1997). "The Effects of Raising Speed Limits on Motor
Vehicle Emissions”, prepared for U.S. EPA, EPA Contract No. 68-03-0035,
March, 1997.
Elvik, R. and Vaa, T. (2004), Speed and Road Safety: Synthesis of Evidence
from Evaluation Studies, Institute of Transport Economics.
European Transport Safety Council (2001). The
Role of Driver Fatigue in
Farkhan, M.(1999). Tire Costs Eating Your
Profit? Waste Age, 91.
Farmer, C.M., Richard A. and
Federal Motor Carrier Safety Administration
(2003) Hours of Service of Drivers,
Driver Rest and Sleep for Safe Operation,
Federal Motor Carrier Safety Administration
(2003) Results
from the 2003 R&T Stakeholder Forums. http://www.fmcsa.dot.gov/safetyprogs/research/briefs/Forum_Report_digital_version_all.htm
Fell, D. (1987). A New View of Driver Fatigue. Draft version. Rosebery, NSW: Traffic
Authority of NSW.
Fieldwick, R. and Brown, R.J. (1987). The Effect
of Speed Limits on Road Casualties. Traffic
Engineering Control, 28,
635-640.
Fildes, B.N. and Lee, S.J. (1993). The Speed Review: Road Environment,
Behavior, Speed Limits, Enforcement and Crashes. Report CR 127 (FORS) CR 3
/ 93 (RSB) (Federal Office of Road Safety and the Road Safety Bureau, Roads and
Traffic Authority of
Finch, D.J., Kompter, P., Lockwood, C.R. and
Maycook, G. (1994). Speed, Speed Limits,
and Crashes. Project Report 58 (
Fitzgerald, R.W. (1989). Rural Truck Speed Differentials. The 1986/87 National Study.
Federal Office of Road Safety Report.
Fowles, R. and Loeb, P.D. (1989) Speeding,
Coordination and the 55 Mph Limit: Comment. American
Economic Review, 79, 916-921.
Freedman, M. and Esterlitz, J.R. (1990). Effect
of the 65 mph Speed Limit on Speeds in Three States. Transportation Research Record, 1281, 52-61.
Freedman, M. and Williiams, A.F. (1992). Speeds
Associated with 55-MPH and 65-MPH Speed Limits in Northeastern States. ITE Journal, 62(2), 17-21.
Ganote, D.P. (1997). Uniform Speed Limit for Cars and Trucks Urged. Proponent Testimony
for the
Ganote, D.P. (1999). Fatality Rates on German Autobahns, National Motorist Association,
Garber, N.J. and Ehrhart, A.A. (2000). The Effect of Speed, Flow, and Geometric
Characteristics on Crash Rates for Different Types of
Garber, N.J., and Gadiraju, R. (1988). Speed
Variance and Its Influence on Accidents.
Garber, N.J. and Gadiraju, R. (1989). Factors
Affecting Speed Variance and its Influence on Accidents. Transportation Research Record, 1213, 64-71.
Garber, N.J. and Gadiraju, R. (1990). Safety
Impact of Truck Restrictions. Transportation
Executive Update, 4(2), 24.
Garber, N.J. and Gadiraju, R. (1991). Impact of Differential Speed Limits on
Highway Speeds and Accidents. Final Report for AAA Foundation for Traffic
Safety. January 1991.
Garber , N.J. and Gadiraju, R. (1992). Impact of
Differential Speed Limits on the Speed of Traffic and the Rate of Accidents. Transportation Research Record, 1375, 44-52.
Garber, N.J. and Graham, J.D. (1989). The
Effects of the new 65 Mile-Per-Hour Speed Limit on Rural Highway Fatalities: A
State-by-State Analysis. Accident
Analysis and Prevention, 22(2),
137-149.
Garber, N.J. and Joshua, S. (1989). Characteristics of Large-Truck Crashes in
Garber, N.J., Miller, J.S., Yuan, B. and Sun X.
(2003). The Safety Impacts of
Differential Speed Limits on Rural Interstate Highways, Transportation
Research Board.
Godwin, S.R. (1992). Effect of the 65 m.p.h.
Speed Limit on Highway Safety in the
Godwin, R. and Kulash, D.J. (1988). The 55 mph
Speed Limit on US Roads: Issues Involved, Transport
Reviews, 8, 219-335.
Goodyear (2004). Radial Truck Tire and Retread
Service Manual. Goodyear.
Governors Highway Safety Association (GHSA). Survey of the States: Speeding.
http://www.statehighwaysafety.org/html/publications/pdf/surveystates2005/surveystates_speeding.pdf
Gruberg, R. (1999). Speeding-Related Multi-Vehicle Fatal Crashes Involving Large Trucks.
Office of Motor Carrier Research Document Analysis Division.
Hall, J.W. and
Hall, D.E. and Moreland, J.C. (2001).
Fundamentals of Rolling Resistance. Rubber
Chemistry and Technology, 74 (3), 525.
Hamelin, P. (1987). Lorry Drivers' Time Habits in Work and their
Involvement in Traffic Accident. Ergonomics,
30 (9), 1323-1333.
Harkey , D.L. and Mera, R. (1994). Safety Impacts of Different Speed Limits on
Cars and Trucks. Final Report, FHWA-RD-93-161 (
Harkey, D., Robertson, H.D. and Davis, S.E.
(1989). Assesment of Current Speed Zoning Criteria. Transportation Research Record, 1281, 40-51.
Hart's European Fuels News (2001). European
Commission Wants Speed Limiters on All Lorries and Buses,
Hauer, E. (1971). Accidents, Overtaking and Speed Control. Accident
Analysis and Prevention, 3, 1–13.
Hauer, E., and F.J. Ahlin. (1982). Speed
Enforcement and Speed Choice. Accident Analysis and Prevention, 14 (4),
267–278.
Harwood, D.W., Potts, I.B. and Torbic,
D.J. (2003). Synthesis 3 Highway/Heavy
Vehicle Interaction. Commercial Truck Bus Safety Synthesis Program,
National Research Council,
Houston, D.J. (1999). Implication of the 65-MPH
Speed Limit for Traffic Safety. Evaluation
Review, 23(3), 304-315.
Haworth, N. (1998), Fatigue and Fatigue Research: The Australian
Experience. Paper Presented to 7th Biennial
Australasian Traffic Education Conference, Speed, Alcohol, Fatigue, Effects,
Insurance Institute of Highway Safety. (1991).
Is Speed Variation, Not Speed Itself, the Real Problem? Accident
Reconstruction Journal, 3 (6), 16.
Jernigan, J.D.,
Jiao, K, Li, Z.Y., Chen, M. and Wang, C.T.
(2004). Effect of Different Vehicle
Speeds on Mental Fatigue of Healthy Drivers, SAE 2004-01-0234.
Johansson , P. (1996). Speed limitation and
Motorway Casualties: A Time-Series Count Data Regression Approach. Accident Analysis and Prevention, 28,
73-87.
Joksch, H.C. (1975). An Empirical Relationship
between Fatal Accident Involvement per Accident Involvement and Speed. Accident Analysis and Prevention, 7,
129-132.
Jones,
Judycki, D. (1994). Technical Advisory: Motor Vehicle Accident
Costs. Technical Advisory 7570.2.
Kamerud, D.E. (1988). Benefits and Costs of the
55 mph Speed Limit: New Estimates and Their Implications. Journal of Policy Analysis and Management. 7 (2), 341-352.
Kamerud, D.B. (1988). Evaluating the New 65 mph Speed Limit. In John D. Graham (ed.),
Preventing Automobile Injury: New Findings From Evaluation Research.
Kean, A.J., Harley, R.A. and
Khan, N. and Sinha, K.C. (2000). An Analysis of Speed Limit Policies for
Kloeden, C.N., Ponte, G. and
Knipling, R.R., Waller, P., Peck, R.C., Pfefer,
R., Neuman, T.R., Slack, K.L. and Hardy, K.K. (2004). A Guide for Reducing Collisions Involving Heavy Trucks, NCHRP
Report 500, Transportation Research Board.
Kostyniuk, L.P., Streff, F.M. and Zakrajsek, J.
(2002). Identifying Unsafe Driver Actions that Lead to Fatal Car-Truck
Crashes, AAA Foundation for Traffic Safety.
Krammes, R.A., Fitzpatrick, K., Blaschke, J.D.
and Fambro, D.B. (1996). Speed,
Under-standing Design, Operating, and Posted Speed. Report 1465-1 (Texas
Transportation Institute).
Krell, K.E. and Heinz, R.K. (1977). Costs and Benefits of General Speed Limits:
Report for the thirty-seventh Round Table on Transportation Economics held in
Land Line Magazine. (2005). 2005 State Index,
Legislative Watch. http://www.landlinemag.com/Legislative_Watch/2005/leg_watch.htm
Langlotz,
B. (1999). Did Raising Freeway Speed
Limits Affect Traffic Safety? National Motorist Association Foundation.
March 1999.
Lave,
Lave,
Lave,
Ledolter, J., and Chan, K.S. (1994). Safety Impact of the Increased 65 mph Speed
Limit on
Levy , D.T. and Asch , P. (1989). Speeding,
Coordination, and the 55 mph Limit: Comment. American Economic Review, 79, 913-915.
Lissaman, P.B.S. (1975). Development of
Devices to Reduce the Aerodynamic Resistance of Trucks, 750702, Society of
Automotive Engineers.
Liu, X.G. (1998). Travel Speed and Speed Differential and Their Effects on Traffic Safety,
Transport Association of
Luskin, D.M. and Walton, C.M. (2001) Effects of Truck Size and Weights on Highway Infrastructure and
Operations: A Synthesis Report, FHWA/TX-0-2122-1, Texas Department of Transportation.
Mace, D. J. and Heckard, R. (1991). Effect of the 65 mph Speed Limit on Travel
Speeds and Related Crashes. Final Report HS 807 764, March (
Mannering. Fred L and Kilareski, Walter P. (1998). Principles
of Highway Engineering and Traffic Analysis, Wiley,
McCarthy, P.S. (1988). Highway Safety and the 65 mph Maximum Speed Limit: An empirical Study.
AAA Foundation for Traffic Safety,
McCarthy, P.S. (1994). An Empirical Analysis of
the Direct and Indirect Effects of Relaxed Interstate Speed Limits on Highway
Safety. Journal of Urban Economics,
36, 353-364.
McCarthy, P.S. (1994). Relaxed Speed Limits and
Highway Safety: New Evidence from
McCarthy, P.S. (2001). Effect of Speed Limits on
Speed Distributions and Highway Safety: A Survey of Recent Literature. Transport Reviews, 21(1), 31-50.
McKnight, A.J., Klein, T.M. and Tippetts, A.S.
(1989). The Effect of the 65 mph Limit on
Speeds and Accidents.
McKinght, A.J. and Klein, T.M. (1990).
Relationship of 65 Mph Limit to Speeds and Fatal Accidents. Transportation Research Record, 1281,
71-77.
McPhee, A.T. (1996). Are Higher Speed Limits
More Deadly? Current Science, 81(14),
14-16.
Meier, K.J., and Morgan, D.R. (1981). Speed
Kills: A Longitudinal Analysis of Traffic Fatalities and the 55 mph Speed
Limit. Policy Studies Review, 1,
157-167.
Monsere, C.M., Newgard, C., Dill, J. Rufolo, A.,
Wemple, E., Bertini, R.L. and Miliken, C. (2004). Impacts of Issues Related to Proposed Changes in Oregon’s Interstate
Speed Limits, Portland State University, Center for Transportation Studies,
Research Report, September 2004.
Munden, J.M. (1967). The Relation between a Driver's Speed and His Accident Rate.
Transport and Road research laboratory, LR-88,
Muster, T. (2000). Fuel Saving Potentials and Costs Considerations for US Class 8 Heavy
Duty Trucks Through Resistance Reductions and Improved Propulsion Technologies
until 2020, Energy Laboratory, MIT
Cambridge, Massachusetts, # MIT EL
00-001.
Najjar, Y.M., Russel, E.R., Stokes, R.W. and
Abu-Lebdeh, G. (2002). New Speed Limits On
Nakao, D.K. (1989). After Sixty-five? Analysis of
the Speed on Rural Interstate Freeways. California Department of
Transportation, Division of Traffic Operations.
National Highway Traffic Safety Administration.
(1989). The Effects of the 65 mph Speed
Limit During 1987: A Report to
National Highway Traffic Safety Administration.
(1992). The Effects of the 65 mph Speed
Limit Through 1990: A Report to
National Highway Traffic Safety Administration.
(1998). Report to Congress: The Effect of
Increased Speed Limits in the Post-NMSL Era, US Department of
Transportation, Washington, DC.
National Highway Traffic Safety Administration.
(2004) Traffic Safety Facts: 2003 Data.
DOT-HS-809-767. http://www-nrd.nhtsa.dot.gov/pdf/nrd-30/NCSA/TSF2003/809767.pdf
National Research Council. (1998). Managing Speed: A Review of Current Practice
for Setting and Enforcing Speed Limits. Transportation Research Board, Special Report 254.
Neeley, G.W. and
Nilsson, G. (1977). Trials with Differentiated Speed Limits during the Years 1968-1972.
B527.
Nilsson, G. (1990). Reduction in the Speed Limit from 110 km/h to 90 km/ h during Summer
1989, Swedish Road Safety Office, Linkoping, VTI Rapport 358A.
North Carolina Department of Motor Vehicle.
(undated). Effect of Speed on Braking Distance. http://www.dmv.dot.state.nc.us/driverlicense/CDL/handbook/speed.html
Northern Echo. (2000). Keep Fuel in Your Tank, Echonews Edition, Darlington (
Office of Road safety (2004), Fatigue Facts,
Oregon Trucking association. (2003). Stopping
Distance of Car versus Truck, http://www.ortrucking.org/htmlpages/stopping.htm
Oron-Gilad, T., Ronen, A., Cassuto, Y. and
Pant, P.D., Adhami, J.A. and Niehaus, J.C.
(1991). Effects of the 65-mph Speed Limit on Traffic Crashes in
Parker, M.R. (1992). Effects of Raising and Lowering Speed Limits, US Department of
Transportation, Technical Report FHWA-RD-92-084, FHWA, Washington, DC.
Patterson, T.L., Frith, W.J., Povey, L.J. and
Keall, M.D. (2002). The Effect of Increasing Rural Interstate Speed Limits in
the
Persaud, B., Parker, M., Knowles, V. and Wilde, G. (1997) Safety,
Speed and Speed Management: A Canadian Review; File No. ASF 3261-280, Road
Safety and Motor Vehicle Regulation, Transport
Pfeffer, R.C.,
Stenzel, W.W. and Lee, B.D. (1991). Safety Impact of the 65-Mph Speed Limit: A
Time-Series Analysis. Transportation
Research Record, 1318, 22-33.
Radwan, A.E. and Sinha, K. (1978). Effect of the
National Speed Limit on the Severity of Heavy-Truck Accidents. Traffic Quarterly. 32 (2), 319-328.
Rajbhandari, R. and
Daniel, J. (2002). Impacts of The 65-mph
Speed Limit on Truck Safety, 82nd Meeting of Transportation Research Board.
Raju, S., Souleyrette, R. and Maze, T.H. (1998).
Impact of the 65 Mph Speed Limit on
Renski, H., Khattak, A.J. and Council, F.M.
(1999). Effect of Speed Limit Increases on Crash Injury Severity: Analysis of
Single-Vehicle Crashes on
Research Triangle Institute. (1970). Speed
and Accident, Volume II, Report No. FH-11-6965,
Retting, R.A. and Greene, M.A. (1996). Traffic Speeds Following Repeal of the
National Maximum Speed Limit: Preliminary Results, Insurance Institute for
Highway Safety.
Roads and Traffic Authority. (2004). Driver fatigue, Roads and Traffic Authority.
Rock, S.M. (1995). Impact of the 65 mph Speed
Limit on Accidents, Deaths, and Injuries in
Ryan, G.A., Wright, J.N., Hinrichs, R.W. and
Safety Management System Task Force on Speed
Limits. (2002). Update Report on Speed
Limits in
Sagberg, F. (1999). Road Accidents Caused by
Drivers Falling Asleep. Accident Analysis and Prevention, 31(6),
639-649.
Salusjarvi, M. (1988). The Speed Limit Experiments on Public Roads in
Saricks, C.L. and Tompkins, M.M. (1999). State-Level Accident Rates of Surface
Freight Transportation: A Reexamination. Technical Report ANL/ESD/TM-150.
Schmidt, F. and
Sidhu, C.S. (1990). Preliminary Assessment of
the Increased Speed Limit on Rural Interstate Highways in
Sliogeris, J. (1992). 110 Kilometer per Hour Speed Limit - Evaluation of Road Safety Effects. Report No. GR 92-8, August (
Smiley, A. (1999). Driver Speed Estimation: What Road Designers Should Know. Conference,
Transportation Research Board 78th Annual Meeting, Workshop on Role of
Geometric Design and Human Factors in Setting Speed.
Smith, R.N. (1990). Accidents Before and After the 65 MPH Speed Limits in
Snyder, D. (1989). Speeding, Coordination, and the 55-mph Limit:
Comment. The American Economic Review. 79, 922–925.
Solomon, D. (1964). Accidents on Main Rural Highways Related to Speed, Driver, and Vehicle.
Society of Automotive Engineers (SAE). Joint TMC/SAE Fuel Consumption Test
Procedure - Type II. Surface Vehicle
Recommended Practice SAE J1321, 1986.
Spencer T. (2003). Testimony of Todd Spencer, Senate
Highways and Transportation Committee, Senate Bill 94, Owner
Operator Independent Drivers Association.
Srinivasan, R. (2002). Characteristics of Traffic Flow and Safety in 55 and 65 mph Speed
Limits: Literature Review and Suggestions for Future Research, NJ
Department of Transportation,
Stuster, J. (1999). The Unsafe Driving Acts of Motorists in the
Vicinity of Large Trucks,
The Commercial Drivers License. (2000). Controlling Speed.
http://drivingrules.net/cdl/cdlsecb/b6speed.htm
The Maintenance Council. (1996). 55 vs. 65+, An Equipment Operating Costs
Comparison, The Maintenance Council, American Trucking Association.
The Maintenance Council. (1998). The Fleet Manager’s Guide to Fuel Economy,
The Maintenance Council, American Trucking Association.
Thiriez, K., Radja, G. and Toth, G. (2002) Large
Truck Crash Causation Study Interim Report. DOT-HS-809-527.
http://www-nrd.nhtsa.dot.gov/pdf/nrd-30/NCSA/Rpts/2002/809-527.pdf
Transport
http://www.tc.gc.ca/pol/en/report/TruckActivity/Chapter9.htm
Treat, J.R.,
Truck writers of
Upchurch, J. (1989).
US Department of Energy. (2003). The Transportation Energy Data Book, 23rd Edition, http://cta.ornl.gov/data/index.shtml
Wagenaar, A.C.,
Streff, F.M. and Schultz, R.H. (1990). Effects of the 65 MPH Speed Limit on Injury
Morbidity and Mortality. Accident
Analysis and Prevention, 22, 571-585.
West, L.B. and Dunn, J.W. (1971) Accidents,
Speed Deviation and Speed Limits. Traffic Engineering, 41 (10), July
1971.
Williamson, A.M., Feyer, A.M., Friswell, R. and
Sadural, S. (2001). Driver Fatigue: A
Survey of Professional Long Distance Heavy Vehicle Drivers in
Wilmot, C.G. and Khanal, M. (1999). Effect of
Speed Limits on Speed and Safety: A Review. Transport
Reviews, 19 (4), 315-329.
Wislocki, J. (2003). Truck Related Highway
Fatalities Fall Again. Transport Topic,
(3548), 6.
Yuan, B. and Garber, N.J. (2002). Effects of Differential Speed Limits on
Vehicle Speed and Crash Characteristics Using Hypothesis Tests. Center for
Transportation Studies,
Zlatoper, T.J. (1991). Determinants of Motor
Vehicle Deaths in the
A. Speed Limits Before 55 mph
NMSL in 1974: (Atkinson ,1996)
|
State |
Speed
Limit |
|
|
70 |
|
|
70 |
|
|
75 |
|
|
75 |
|
|
70 |
|
|
70 |
|
|
60 |
|
|
60 |
|
|
70 |
|
|
70 |
|
|
70 |
|
|
70 |
|
|
70 |
|
|
70 |
|
|
75 |
|
|
75 |
|
|
70 |
|
|
70 |
|
|
70 |
|
|
70 |
|
|
65 |
|
|
70 |
|
|
65 |
|
|
70 |
|
|
70 |
|
|
no speed limit |
|
|
75 |
|
|
no speed limit |
|
|
70 |
|
|
60 |
|
|
70 |
|
|
65 |
|
|
70 |
|
|
70 |
|
|
70 |
|
|
70 |
|
|
75 |
|
|
65 |
|
|
60 |
|
|
70 |
|
|
75 |
|
|
75 |
|
|
70 |
|
|
70 |
|
|
65 |
|
|
70 |
|
|
70 |
|
|
70 |
|
|
70 |
|
|
75 |
B. 1987 Speed Limit Increase: (Baum, 1989; Advocates of Highway Safety,
1995)
|
State |
Implementation Date for States
which Increased Speed Limits in 1987 |
Implementation Date for States
which Increased Speed Limits after 1987 |
|
|
|
|
|
|
N/A |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
N/A |
|
|
Dist of |
N/A |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
C. 1995 Speed Limit Increase. (NHTSA, 1998)
|
State |
Implemented |
Speed
Limit Change |
|
|
|
To 70 mph
on Interstates |
|
|
|
|
|
|
|
To 75 mph
on Rural Interstates |
|
|
|
To 70 mph
on Rural four-lane divided highways |
|
|
|
To 70 mph
on Rural Freeways |
|
|
|
To 75 mph
on Highway |
|
|
|
|
|
|
|
To 65 mph
on Interstate |
|
Dist of |
|
|
|
|
|
To 70 mph
for some Interstate segments |
|
|
|
To 70 mph
on Interstate and look-alikes |
|
|
|
|
|
|
|
To 75 mph
on Interstates |
|
|
|
65 on
Urban Interstate |
|
|
|
|
|
|
|
To 65 mph
on selected four-lane divided |
|
|
|
To 70 mph
on Interstates |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
To 60 or
65 mph on selected Urban Interstates |
|
|
|
To 65 mph
on 13 Major Interstates and Highways |
|
|
|
To 70 mph
on Interstates |
|
|
|
|
|
|
|
To 70 mph
on Interstates |
|
|
|
To 70 mph
on Interstates |
|
|
|
Unlimited
during day; to 65 mph at night |
|
|
|
To 75 mph
on Interstates |
|
|
|
To 75 mph
on Interstates |
|
|
|
|
|
|
|
|
|
|
|
To 75 mph
on Interstates |
|
|
|
|
|
|
Aug-96 |
To 70 mph
on Interstates |
|
|
|
|
|
|
|
To 65 mph
on Interstate |
|
|
Dec-95 |
To 70 mph
on Interstates and four-lanes |
|
|
|
|
|
|
Dec-95 |
On
Turnpikes roads to 75 mph; Selected roads to 65 mph |
|
|
|
To 65 mph
on some Interstates |
|
|
|
|
|
|
|
To 75 mph
on Interstates |
|
|
|
To 65 mph
on some Urban Interstates |
|
|
|
70 mph for
Cars (65 mph at night) and 60 mph for Trucks (55 mph at night) |
|
|
|
To 75 mph
on Interstates |
|
|
|
|
|
|
|
|
|
|
|
To 70 mph
on Interstates |
|
|
|
|
|
|
|
|
|
|
|
To 75 mph
on Rural Interstates |
D. Rural Interstate Speed Limits. (Insurance Institute for Highway Safety)
|
State |
State
Abbreviation |
Speed Limit |
|
|
|
70 |
|
|
AK |
65 |
|
|
AZ |
75 |
|
|
AR |
70
[trucks: 65] |
|
|
CA |
70
[trucks: 55] |
|
|
CO |
75 |
|
|
CT |
65 |
|
|
DE |
65 |
|
Dist of |
|
N/A |
|
|
FL |
70 |
|
|
GA |
70 |
|
|
HI |
60 |
|
|
ID |
75
[trucks: 65] |
|
|
IL |
65
[trucks: 55] |
|
|
IN |
65
[trucks: 60] |
|
|
IA |
65 |
|
|
KS |
70 |
|
|
KY |
65 |
|
|
LA |
70 |
|
|
ME |
65 |
|
|
MD |
65 |
|
|
MA |
65 |
|
|
MI |
70
[trucks: 55] |
|
|
MN |
70 |
|
|
MS |
70 |
|
|
MO |
70 |
|
|
MT |
75
[trucks: 65] |
|
|
NE |
75 |
|
|
NV |
75 |
|
|
NH |
65 |
|
|
NJ |
65 |
|
|
NM |
75 |
|
|
NY |
65 |
|
|
NC |
70 |
|
|
ND |
75 |
|
|
OH |
65
[trucks: 55; 65 on turnpike] |
|
|
OK |
70 (75
on Turnpike) |
|
|
OR |
65
[trucks: 55] |
|
|
PA |
65 |
|
|
RI |
65 |
|
|
SC |
70 |
|
|
SD |
75 |
|
|
TN |
70 |
|
|
TX |
day:
75 night: 65 [trucks: 65] |
|
|
UT |
75 |
|
|
VT |
65 |
|
|
VA |
65 |
|
|
WA |
70
[trucks: 60] |
|
|
WV |
70 |
|
|
WI |
65 |
|
|
WY |
75 |
E. Summary of
Speed Data at Individual Sites
|
|
|
Traffic |
Auto. |
Truck |
|
State |
|
|
|
|
|
Effingham (IL) I-55
South (1) 65/55
mph |
Average (mph) |
71.3 |
73.8 |
64.2 |
|
Standard Deviation |
6.66 |
5.43 |
4.27 |
|
|
Sample Size |
353 |
260 |
93 |
|
|
Compliance (%) |
|
5.77 |
0 |
|
|
85th % (mph) |
78 |
79 |
68 |
|
|
50th % (mph) |
71 |
73 |
64 |
|
|
Speed Variance |
7 |
6 |
4 |
|
|
|
|
|
|
|
|
Effingham (IL) I-55
North 65/55 mph |
Average (mph) |
71.6 |
72.8 |
63.8 |
|
Standard Deviation |
6.26 |
5.69 |
3.44 |
|
|
Sample Size |
370 |
318 |
52 |
|
|
Compliance (%) |
|
5.66 |
0 |
|
|
85th % (mph) |
77 |
77 |
67 |
|
|
50th % (mph) |
71 |
72 |
64 |
|
|
Speed Variance |
6 |
5 |
3 |
|
|
|
|
|
|
|
|
Effingham (IL) I-55
South (2) 65/55 mph |
Average (mph) |
70.7 |
73.2 |
64.4 |
|
Standard Deviation |
6.68 |
5.84 |
4.02 |
|
|
Sample Size |
417 |
300 |
117 |
|
|
Compliance (%) |
|
10 |
0 |
|
|
85th % (mph) |
77 |
79 |
68 |
|
|
50th % (mph) |
71 |
73 |
64 |
|
|
Speed Variance |
6 |
6 |
4 |
|
|
|
|
|
|
|
|
Rolla (MO) I-40
East 70/70 mph |
Average (mph) |
71.8 |
73.2 |
68.7 |
|
Standard Deviation |
5.37 |
5.26 |
4.25 |
|
|
Sample Size |
284 |
196 |
88 |
|
|
Compliance (%) |
|
28.06 |
71.59 |
|
|
85th % (mph) |
77 |
78 |
73 |
|
|
50th % (mph) |
72 |
73 |
68 |
|
|
Speed Variance |
5 |
5 |
5 |
|
|
|
|
|
|
|
|
Rolla (MO) I-40
West 70/70 mph |
Average (mph) |
71.8 |
73.3 |
68.4 |
|
Standard Deviation |
5.33 |
4.88 |
4.73 |
|
|
Sample Size |
270 |
187 |
83 |
|
|
Compliance (%) |
|
24.6 |
72.29 |
|
|
85th % (mph) |
77 |
77 |
73 |
|
|
50th % (mph) |
72 |
73 |
69 |
|
|
Speed Variance |
5 |
4 |
4 |
|
|
|
|
|
|
|
|
|
|
Traffic |
Auto. |
Truck |
|
Joplin (MO) 70/70
mph |
Average (mph) |
70.8 |
71.5 |
68.7 |
|
Standard Deviation |
4.75 |
4.54 |
4.75 |
|
|
Sample Size |
304 |
228 |
76 |
|
|
Compliance (%) |
|
39.91 |
64.47 |
|
|
85th % (mph) |
76 |
76 |
73 |
|
|
50th % (mph) |
71 |
71 |
69 |
|
|
Speed Variance |
5 |
5 |
4 |
|
|
|
|
|
|
|
|
Ozark (AR) I-40
South 70/65 mph |
Average (mph) |
71.5 |
73.5 |
66.7 |
|
Standard Deviation |
5.35 |
4.50 |
4.05 |
|
|
Sample Size |
361 |
255 |
106 |
|
|
Compliance (%) |
|
21.96 |
31.13 |
|
|
85th % (mph) |
77 |
78 |
70 |
|
|
50th % (mph) |
72 |
74 |
67 |
|
|
Speed Variance |
5 |
4 |
3 |
|
|
|
|
|
|
|
|
Ozark (AR) I-40
North 70/65 mph |
Average (mph) |
71.0 |
73.5 |
66.7 |
|
Standard Deviation |
4.83 |
3.85 |
3.01 |
|
|
Sample Size |
170 |
107 |
63 |
|
|
Compliance (%) |
|
21.5 |
34.92 |
|
|
85th % (mph) |
76 |
77 |
70 |
|
|
50th % (mph) |
71 |
74 |
66 |
|
|
Speed Variance |
5 |
3 |
4 |
|
|
|
|
|
|
|
|
|
Average (mph) |
74.2 |
74.8 |
72.3 |
|
Standard Deviation |
4.93 |
4.61 |
5.63 |
|
|
N |
154 |
121 |
33 |
|
|
Compliance (%) |
|
52.89 |
72.72 |
|
|
85th % (mph) |
79 |
80 |
77 |
|
|
50th % (mph) |
74 |
75 |
72 |
|
|
Speed Variance |
5 |
5 |
5 |
F. Truck Driver’s Survey:
Company driver
Owner-operator (owns tractor only)
Owner-operator
(leasing truck) Owner-operator
(owns tractor & trailer)
Driving for Owner-operators
NoSingle day 2-7 days more than 7
days
Previous load/unloading point (State) _____________,
next load/unload point _______________.
Dry vans Flat
beds Doubles,
Triples
Reefers Tankers
Others ___________
Less
than Truck Load (LTL) When you are beginning to
pass
When you are traveling parallel to the other car
When you are pulling back
into the right lane
When the car is beginning to
pass
When the car is traveling parallel to your truck
When the car is pulling back
into the right lane
When a
car passes a truck When a truck passes a car
Increase accidents Decrease accidents No effect
Increases Decreases Does not change
Side
collision
