Rollover of Heavy Commercial Vehicles

UMTRI-99- 19

August, 1999

C. B. Winkler
R.D. Ervin
The University of Michigan
Transportation Research Institute
2901 Baxter Road, Ann Arbor, Michigan 48 109

for
Volvo Truck Corporation, AB
Goteborg, Sweden
and the
Great Lakes Center for Truck and Transit Research
Ann Arbor, Michigan
 Technical Report Documentation Page
1. Report No. 2. Government Accession No. 3. Reciplent's Catalog No.

4. Title and Subtitle 5. Report Date

August, 1999
Rollover of Heavy Commercial Vehicles 6. Performing Organization Code

8. Performing Organlzatlon Report Nio.

7. Author@)
Winkler, C. B.; Ervin, R.D.
-

9. Performing Organlzation Name and Address 1 10. work unit NO. (TRAIS)
The University of Michigan
Transportation Research Institute 11. Contract or Grant NO.

2901 Baxter Road, Ann Arbor, MI 48109-2150

-
13. Type of Report and Perlod Coverejd
12. Sponsoring Agency Name and Address
Final Report
Volvo Truck Corporation, 1
B
Great lakes Center for Truck and Transit Research

15. Supplementary Notes

I
I
14. Sponsoring Agency Code

16. Abstract
The state-of-the-art understanding of rollover of the commercial vehicle is reviewed. Accident statistics are
presented which highlight the severity and lethal nature of rollover crashes. Physical and statistical evidlence
for the linkage between vehicle roll stability and the actual occurrence of rollover accidents is presented. The
fundamentals of static roll stability are described in detail and then enhanced with discussion of dynamilc
considerations of the rollover process. The text concludes with a discussion of the evolving use of intelligent
electronic systems and active vehicle control for reducing the occurrence of rollover. Appendices include a
bibliography of the literature on heavy-vehicle rollover.

17. Key Words 18. Distrlbutlon Statement

truck, articulated vehicle, stability Unrestricted

1
19. Security Classif. (of thls report) 20. Security Classlf. (of this page) 21. No. of Pages 22. Prl~ce

None None 60
1. ROLLOVERACCIDENTS AND VEHICLEROLLSTABILITY ...............................................1
Rollover and accident severity ...........................................................................
1
Roll stability and the occurrence of rollover accidents ........... ...... ................2
2 . THEMECHANICS OF STATICROLLSTABILTY ........................................................ 7
Introduction and the simplified roll-plane model ................................................ 7
Rollover of the rigid vehicle .............................................................................. 9
The vehicle with compliant tires ........................................................................ 9
The vehicle with roll-compliant suspension ...................................................... 10
The influence of lash in suspensions and in the fifth-wheel coupler ..................11
The influence of multiple suspensions ............................................................ 12
Other mechanisms influence static roll stability ...............................................14
Measuring rollover threshold with the tilt-table experiment ............ . .........,.....16
3. DYNAMIC CONSIDERATIONS IN ROLLOVER OF HEAVYVEHICLES ............................... 19
Introductions ....................................................................................................
19
The influence of sloshing liquids and other moving loads ...........................,.... 19
Rollover in dynamic maneuvers ........................................................................ 26
4 . ROLLOVERAND THE INTELLIGENT
HIGHWAYNEHICLE............................
SYSTEM 31

APPENDIX A . NOTESON ACCIDENT-DATA SOURCES ...................................................

39
APPENDIX B . ANALYSES
OF THE INFLUENCE OF ROLL STABILITY ON ROLLOVER
. .
ACCIDENTS .................... ...............................................................................
41
Percent of accidents involving rollover as a function of roll stability ................41
Rollover accident rate as a function of roll stability ..........................................
43
APPENDIX C. A PARTIAL
BIBLIOGRAPHY OF THE LITERATURE ON HEAVY-VEHICLE;
ROLLOVER .............................................................................................................
45
 1. ROLLOVERACCIDENTSAND VEHICLEROLLSTABILITY
Rollover accidents are of special concern for commercial vehicle safety. Rollover
accidents are especially violent and cause greater damage and injury than other accidents.
Moreover, the relatively low roll stability of the commercial truck promotes rol1ove:r and
contributes to the number of truck accidents. These ideas are generally confirmed by the
accident record.'
Rollover and accident severity
Commercial truck rollover is strongly associated with severe injury and f~taliti~esin
highway accidents. According to data from the General Estimates System (GES)~,
rollover occurred in about 3.6 percent of U.S. truck accidents in 1995. However, from the
Truck and Bus Crash Fact Book (T&BFB)2 for the same year (the most resent available),
rollover occurred in '7.9 percent of tow-away accidents involving trucks, 9.9 percent of
injury accidents, and 12.3 percent of fatal accidents.[l13
In the preceding statistics, fatalities and injuries refer to anyone involved in the
accident including occupants of other vehicles or pedestrians. The association of rollover
with injuries to the truck driver is even stronger. Again from the 1995 T&BFB, dea~thor
incapacitating injury is about ten times more likely to occur to the truck driver in rollover
accidents than in nonrollover accidents. Further, about 50 percent of truck-driver deaths
and 47 percent of incapacitating injuries occurred in accidents in which the truck rolled
over. (Statistics for other years are similar to those shown in figure 1.[2-41)

Rollover involvement in U.S. truck accidents

By se\/erity-1995
II U.S. truck driver injury by rollover-1!395

All
(GES)
Tow-
away
L-
Injury
iT&BFB) - Fatal

1I Rollover No rollover
Figure 1. Rollover is strongly associated with accident severity and with serious injury to truck drivers
I

' That is, they are confirmed by the U.S. accident record. The points made in this section derive from
statistical analyses of U.S. accident-data files. We would expect similar analyses of European da,ta
sources to yield similar results.
See the descriptive notes on data sources in appendix A.
Numbers in brackets refer to bibliographic references at the end of this text.
 Roll stability and the occurrence of rollover accidents
The low level of basic roll stability of commercial
trucks sets them apartfrom light vehicles and appears to
be a significant contributing cause of truck rollover
accidents. The basic measure of roll stability is the static
rollover threshold, expressed as lateral acceleration in
gravitational units (g). The rollover thresholds of
passenger cars are virtually always greater than 1 g.PI
For light trucks, vans, and SUVs, this property lies in the
range of 0.8 to 1.2 g.WI, but the rollover threshold of a
loaded heavy truck often lies well below 0.5 g.
When loaded to legal gross weight, the typical U.S.
five-axle tractor-van semitrailer combination has a
rollover threshold perhaps as high as 0.5 g with a high-
density, low center of gravity (cg) load, but as low as
0.25 g with the worst-case load-one which completely
fills the volume of the trailer while also reaching legal
gross weight.[7-91The typical U.S. five-axle petroleum
semitanker has a rollover threshold of about 0.35 g.1101
Rollover thresholds of common cryogenic tankers for
the transport of liquefied gases are as low as 0.26 g.[111
El-Gindy and Woodrooffe found a variety of logging
trucks operating in Canada to have thresholds ranging
from 0.23 to 0.31 g.[121 Individual vehicles with rollover
thresholds well below 0.2 g can occur occasionally.[e~g~~
1314

Drivers regularly maneuver vehicles at well over 0.2

g. The AASHTO guidelines for highway curve design
result in lateral accelerations as high as 0.17 g at the
advisory speed.[211 Therefore, even a small degree of
speeding beyond the advisory level will easily cause
actual lateral accelerations to reach 0.25 g in everyday
driving. On the other hand, tire frictional properties limit
lateral acceleration on flat road surfaces to a bit less than
1 g at the very most. These two observations clearly
imply that the rollover threshold of light vehicles lies
above, or just marginally at, the extreme limit of the
vehicles maneuvering ability, but the rollover threshold
of loaded heavy trucks extends well into the
"emergency" maneuvering capability of the vehicle and

~i~~~~2. l-he rollover threshold of ,411 these examples are estimates obtained from simulation or other
trucks extends deep into calculations. However, sufficient tilt-table measurements are
the maneuvering range reported in the literature to confirm the general findings.[e& 14-20]
sometimes into the "normal" maneuvering range.
Nevertheless, it is relatively hard for truck drivers to perceive their proximity to
rollover while driving. First of all, rollover is very much an either-or situation. It is
something like walking up to a cliff with your eyes closed: even as you approach the
edge, your perception is still one of walking on solid ground, until it is too late. Further,
the actual rollover threshold of the commercial truck changes regularly as the load
changes, so the driver may not have the chance to "get used to" the stability of his
vehicle. Finally, for combinations especially, the flexible nature of the tractor frame tends
to isolate the driver from the roll motions of the trailer, which might act as a cue to
rollover.
These observations provide physical rationale for two safety hypotheses:
- Heavy rrucks are subject to a class of rollover accidents to which light vehicles
are not susceptible, namely, rollover accidents caused directly by inad~er~tently
operating the vehicle beyond the rollover threshold.
- Rollover in heavy-truck accidents is strongly related to the basic roll stability of
the vehicle.
The mass accident data support these hypotheses.
7- The accidents described in the first
Untripped rollovers by type of vehicle hypothesis could be identified as sin,gle-
for single-vehicle accidents in the US
vehicle accidents in which the first
signiJicant event is an untripped rollover.
According to the hypothesis, such
accidents should be nearly nonexistent for
passenger cars but more common foir
trucks.
As is often the case, the perfect
accident file for such an analysis does not
exist. However, GES files (which do not
indicate first event) for 1993 through 1996
show that untripped rollovers account for
more than 20 percent of the single-vehicle
I Cars Tractor-semis
I
rollover accidents for tractor semitrailers,
Figure 3. Untripped rollovers are common for tractor- but they make up less than 4 percent of
semitrailer combinations but rare for cars those accidents for passenger cars. Further,
the Trucks In Fatal ~ c c i d e n t files
s ~ (which
contain no comparable data for cars) for 1994 through 1996 show that untripped,Jirst-
event rollovers account for 26.8 percent of the single-vehicle rollover accidents in that
file.
Accident data also confirm the second hypothesis. Figure 4 shows the results of an
analysis of data from the Bureau of Motor Carrier Safety (BMCS)5 from 1987-199 16. The

See the descriptive notes on data sources in appendix A.

Ervin first presented analyses of this type in the1980s.[7-111 The analysis herein is limited to five-axle
 The effect of vehicle stability on rollover

0.3 0.4 0.5 0.6 0.7

Static rollover threshold, gravitational units
Figure 4. The chance of rollover is strongly influenced by the roll stability of the vehicle

figure reveals a strong, well-behaved relationship between roll stability and the chance of
rollover in a single-vehicle accident. The relationship is nonlinear and of the form that
one would expect. That is, as the vehicle becomes more and more stable, the chance of
rollover asymptotically approaches zero. Conversely, as stability decreases, the
sensitivity of the probability of rollover to stability increases rapidly and the function
becomes quite steep.
The data of figure 4 can be combined with other data sources to allow at least a rough
estimation of the influence of physical roll stability on rollover accident rates (i.e.,
rollovers per kilometer of travel). Data presented by Winkler[221 indicate that, in the
1988-1990 time period, the fleet of all tractor-semitrailer combinations in the U.S.
averaged 8697 rollovers per year while traveling an average of 53,434 million kilometers
a year (yielding an average of one rollover per 6.15 million kilometers). Data indicating

tractor-van semitrailers (the most common U.S. heavy vehicle) to facilitate estimation of rollover
threshold. Threshold estimates are based on median cg heights and typical vehicle properties. Jacknife
accidents are excluded to prevent this peculiar type of accident from distorting the high-stability end of
the curve. A detailed explanation of the analysis is given in appendix B.
 The influence of roll stability on the rollover accident rate of tractor-semitrailers

0.3 0.4 0.5 0.6 0.7

Rollover threshold, gravitational units

Figure 5. An estimate of the effect of roll stability on the rollover accident experience of tractor-
semitrailers

the distribution of tractor-semitrailer mileage by gross weight is presented by Campbell

1231. Combining these data with figure 4 yields the estimates of rollover accident rate
presented in figure 5. (See appendix B for details of the analysis.) As in figure 4, a
second-order polynomial function is fit to the data and is extrapolated to low levels of
stability.
Figures 4 and 5 should be viewed only as estimates. As discussed in appendix El, each
is subject to broad assumptions about the applicability of the data. However, regardless of
their absolute accuracy, the qualitative findings present in these figures are believed to be
valid and significant. Namely, as roll stability declines to low levels, the probability of
rollover in an accident increases rapidly until the vehicle becomes very likely to rollover
in nearly any accident. Moreover, for the low-stability vehicles for which rollover i : such
~
a great concern, relatively small improvements in physical stability can yield rather large
improvements in rollover accident rate.
Introduction and the simplified roll-plane model
All rollover events in the real world are dynamic events to some extent; none arc truly
quasi-static. However, the accident-data analyses presented in the previous chapter show
that there is a very strong relationship between the basic, static roll stability of the heavy
vehicle and the actual occurrence of rollover in accidents. Accordingly, this chapter will
discuss the mechanics of quasi-static rollover in order to explain how this fundamental
performance property derives from the mechanical behavior of the various components of
the vehiclem7
Figure 6 presents a simplified model of a heavy
vehicle in a steady turn in which the vehicle, its
tires, and suspenions have been "lumped" into a
single roll plane. The nomenclature of the figure is
as follows:
ay is lateral acceleration,
Fi are the vertical tire loads, i=l, 2,
h is the height of the cg,
T is the track width,
W is the weight of the vehicle,
Ay is the lateral motion of the cg relative: to the
track,
Q, is the roll angle of the vehicle.
The equilibrium equation for roll moment
about a point on the ground at the center of the
Figure 6. A simplified freebody diagram of track is:
a heavy vehicle in a steady turn W*h*ay= (Fp - Fq)*T12 - W*Ay . (1)
Qualitatively, there are two destabilizing
(overturning) moments acting on the vehicle:
- a moment due to the lateral D'Alambert force acting through the cg, W*h0ay, as
a results of the external imposition of lateral acceleration
- a moment due to the weight of the vehicle acting at position that is laterally
offset from the center of the track, W*Ay.
The first of these results from the external imposition of lateral acceleration while ithe
latter results from the internal compliant reaction of the vehicle.
These two destabilizing moments are opposed by one stabilizing (restoring) moment
which is due to the side-to-side transfer of vertical load on the tires, (F2 - Fl)*T/2 . This
moment is also due to the internal, compliant responses of the vehicle. The maximum

7 The analytical approach used herein (and especially the graphical form introduced in figure 8) was
developed by Mallikarjunarao. It first appeared in [7] and has been presented in various levels of detail
many times since. [e.g., 8,9,24]
possible value of this moment is W*T/2 which occurs when all load is transferred to one
side of the vehicle, i.e., when F2 = W and FA= 0.
One way of interpreting equation 1 and the observation of two destabilizing moments,
is that a vehicle's rollover threshold
derives from (1) a reference rigid-body
Rigid vehicle
(TI2h) stability, which would result if Ay were
zero, and (2) the degradation from that
reference resulting from the lateral
motion of the cg allowed by compliances
within the vehicle.
Figure 7 presents a "case study"
which illustrates how various properties
Compliant tires
of the vehicle contribute to the rollover
threshold according to this view. The

-/ example is of a rather low-stability

vehicle. Its heavy load and relatively high
payload establish a rigid-body stability of
0.45 g. The roll motion allowed by the
compliance of the tires and the
suspension springs drop the stability level
Compliant springs
to about 0.36 g. Free play, or lash, in the

/
suspension springs and fifth-wheel
coupler allow more roll motion, further
reducing stability. Less-than-optimum
distribution of load on the suspensions
lowers stability still more. Structural
Fifth-wheel and
spring lash, multiple compliances in suspensions and the cargo
suspensions body and the off-center positioning of the

.
7
Lateral suspension and
body compliances
payload allow additional lateral
translation of the cg.
The qualitative message of figure 7 is
that roll stability is established by the
summated effects of many compliance
w mechanisms. While the eflect of any one
compliance may be small, virtually all
compliances degrade stability. All the
compliances combined can reduce roll
stability to as little as 60percent of the
rigid-vehicle stability.
Off-center cargo
7 The following paragraphs will review
just how each of the mechanisms of
0.25 figure 7 actually influences stability by
Figure 7. An example case showing various major
progressively examining their individual
influences which determine roll influences On the behavior of equation 1.
 Roll moment
(Small angles are assumed to allow
4 load-transfer moment linearity.)
Rollover of the rigid vehicle
Begin by considering a
vehicle completely rigid vehicle. Figure 8 is a
(right side of eq 1)
graphic representation of equatilon 1
RO1l,A"gle for such a vehicle. Equation 1 has

been arranged with the externallly

applied moment on the left side and
Figure 8. Graphic presentation of the roll-equilibrium the internal vehicle-reaction moments
equation for a rigid vehicle on the right side. The graph of figure
8 is arranged the same way. The left
side of the equation is presented on the left side of the graph in a plot of roll moment (on
the ordinate)-versus lateral acceleration (on the abscissa to the left). The right side of the
equation is presented on the right side of the graph in a plot of roll moment versus roll
angle (on the abscissa to the right).
Because this vehicle is rigid, any finite roll of the vehicle results immediately in1
complete transfer of all vertical load onto the tires on one side of the vehicle. The
unloaded tires would immediately lift from the ground. This is reflected in the plot of
load-transfer moment shown as a horizontal line at the maximum value of WoT12.
However, the offset moment grows proportionately (and negatively) with roll angle as the
cg translates laterally. This behavior is shown in the downward sloping plot of the offset
moment. The sum of the load-transfer moment and the (negative) offset moment
constitutes the total vehicle reaction as expressed by the right side of the equation. 'The
graph shows that this combined function achieves its maximum value at zero roll angle.
The negative slope of this plot at all finite roll angles indicates that the vehicle becomes
unstable immediately as its tires lift from the ground. By projecting the maximum value
of this right-side total onto the plot of the left side of the equation, it can be seen thiat the
maximum lateral acceleration (in gravitational units) that can be sustained by this rigid
vehicle in an equilibrium condition is the ratio of the half track (T12) to the cg height (h).
This well-know, rigid-vehicle stability factor, T12h, is the most fundamental vehicle
property which influences basic roll stability.

Roll moment
The vehicle with compliant tires
Now consider a vehicle with
compliant tires represented by linear
vertical springs. This vehicle would
roll about a point located on the
ground plane at the center of the track.
,I In the process, the tire spring on one
side compresses, increasing its :load,
while the tire spring on the other side
extends, decreasing its load.
Figure 9. Graphic presentation of the roll equation for a Simultaneously, the cg translates
vehicle with compliant tires and suspension
 4 Roll moment A Roll moment
- - - -wjder
- - . track
---- -- - ---

Roll Angle

Lateral accel. I Lateral accel.

I lower +c

Figure 10. Improving roll stability with a lower cg and a wider track

laterally a distance equal to roll angle times the height of the center of gravity (cg height).
Load-transfer moment is shown to develop progressively with roll angle. Full load
transfer and the resulting maximum load-transfer moment are achieved at the "tire-lift-off
angle," $L. At $L, the offset moment has grown to a negative value of W*~*$L. The
maximum value of the total-reaction moment is achieved just as tires lift at a roll angle of
This maximum is less than that which was achieved by the rigid vehicle because of
the non-zero offset moment. At roll angles greater than qL, offset moment continues to
increase but load-transfer moment is saturated. The resulting downward slope of the total
vehicle reaction again indicates an unstable system.
The graphic form of figure 9 can be used to illustrate how the rollover threshold is
influenced by both cg height and track width. Consider figure 10. The plot on the left
shows that lowering the cg improves stability by reducing both destabilizing moments.
Moment due to the lateral D'Alambert force through the cg is reduced directly by
lowering its line of action. The lateral shift of the cg is also reduced, thereby reducing the
associated moment. On the overhand, widening the track of the vehicle improves stability
by increasing the stabilizing moment available from side-to-side load transfer.
The vehicle with roll-compliant suspension
Ay due to susp, roll
The effect of roll compliance of the
suspension is very similar to the effect of tire
compliance except that the additional suspension
roll motion takes place about a roll center which
is typically well above the ground. From figure
11 it is apparent that the height of this suspension
roll center has two influences. (1) For a given roll
angle condition, the lateral displacement of the
cg is less if the suspension roll center is higher.
(2) For a given cg height, roll moment acting on
the suspension due to the D'Alambert force is
less if the suspension roll center is higher. This,
in turn, reduces body roll angle and the resulting
tire roll center lateral displacement of the cg.
Figure 11. Tire and suspension roll motions
occur about different centers
 Roll moment A Roll moment

\ 1 ,v tiffer tires or
LsDension

Roll Angle
- d
Lateral accel.

Figure 12. Improving roll stability with stiffer tires or suspension and by raising the suspension roll center

Figure 12 illustrates how stability is improved by (1) increasing tire or suspension

stiffness, and (2) raising the suspension roll center. Increasing roll stiffness has the irather
straightforward effect of causing the complete side-to-side load transfer to occur at a
lesser roll angle. Consequently, at the point of tire lift, the destabilizing moment from the
lateral displacement of the cg is smaller. This translates to improved roll stability.
The graph on the right shows that raising the roll center has two effects. (1) The
amount of lateral translation of the cg per unit of roll decreases, thus lessening the offset
moment, and (2) the stiffness of the system (load-transfer moment per unit of roll)
increases. Both of these effects are the result of reducing moment about the suspension
roll center by shortening the distance between it and the line of action of the lateral force
through the cg.
The influence of lash in suspensions and in the fifth-wheel coupler
Many leaf-spring suspensions allow vertical free play, or lash, between the spring and
body. The spring must pass
through this lash as it
transitions from c~mpress~ion to
tension. Similarly, fifth-wheel
couplers typically allow for
some vertical free play between
the kingpin and the coupler
plate. As shown in figure 13,
either or both of these can
appear as free play in the rolling
motion of the vehicle,
especially for vehicles with
high centers of gravity.
Figure 14 shows the
influence of free play on the roll
w
stability of the vehicle. As the
region of the lash is
Figure 13. If the cg is high, the sprung mass attempts to "roll off'
the suspension and fifth-wheel prior to rollover of the transitioned, the free-rolling
vehicle motion of the vehicle simply
 suspension or results in an increase in lateral
displacement of the cg and the
associated offset moment, but with no
compensating increase in the load-
the vehicle is transfer moment. Roll stability declines
n accordingly. Note that, as the system
passes through lash, it is locally
Lateral accel.
unstable in roll, as indicated by the
negative slope of the total moment.
Figure 14. Suspension and fifth-wheel lash increase the After the of the lash, there is, in
roll angle required for complete load transfer theory, a small zone in which the
vehicle is again stable. In practice,
however, momentum will typically carry the vehicle through this stable zone to rollover.
The influence of multiple suspensions
To this point, the discussion has assumed that all the tire and suspensions of the
vehicle may be "lumped" and assumed to operate as a single suspension. Now consider
figure 15 in which the load-transfer moments of the three suspensions of a tractor-
semitrailer combination are shown individually. The relative performance of the three
suspensions in the figure is typical of real vehicles. That is:
- The trailer suspension exhibits the highest roll stiffness followed by the tractor
drive-axle suspension and then the steer-axle suspension. The stiffness of the
latter is quite low compared to the other two.
- The loads carried by (and therefore, the maximum load-transfer moments of)
the drive-axle and trailer-axle suspensions are similar and are each considerably
more than the load carried on the steer axle.
Starting from the left side of the graph, consider the process as the vehicle gradually
experiences increasing lateral acceleration and corresponding roll motion. At the
beginning, all tires are on the ground, and the slope (effective roll stiffness) of the total
system is determined by the sum of the stiffness of all suspensions less the negative
influence of the offset moment.
ROII moment Because of its greater roll stiffness, the
maximum trailer suspension transfers load side-
total moment moment to-side most quickly and is first to
drive-axle arrive at the point of tire lift ($3).
I 'moment When this occurs the stiffness of the
moment trailer suspension is "lost" and the
stiffness of the total system declines.
moment However, the total stiffness is still
positive, and the system is still stable
even with the trailer tires off the
offset ground. The process continues and the
moment
next tires to lift from the ground are
Figure 15. The load-transfer moments of the steer-axle, the drive-axle tires (02), At this point,
drive-axle, and trailer suspensions shown
separately the drive-axle stiffness is also lost. The
remaining positive stiffness of the steer-axle suspension is less than the negative
influence of the offset moment. As a result, the total-system stiffness is negative, and the
vehicle is unstable. The peak in the total-system moment occurs at the roll angle $2 and
represents the rollover threshold of the vehicle.
Generalizing on the presentation of figure 15, three classifications of suspensions can
be identified: (1) suspensions whose tires lift off before (i.e., at a smaller roll angle) the
peak of the total-system moment, (2) suspensions whose tires lift off at (and therefore
define) the peak of the total-system moment, and (3) suspensions whose tires remain on
the ground when the total-system moment peaks. Figure 16 illustrates the different
influences brought about by changing the roll stiffnesses of these three types of
suspensions. Starting from the left of the figure, we see that stiffening suspensions of the
first type have no influence on rollover threshold. In essence, this type of suspensioi~
"delivers" all its available stabilizing moment to the system prior to instability.
Delivering that moment even quicker can have no influence. It does not change the roll
angle of the point of instability and it does not change the amount of stabilizing moment
available from this suspension. However, from the second graph, stiffening the second
type of suspension does result in an improvement in rollover threshold. Stiffening this
suspension causes the point of instability to occur at a lesser roll angle. As a result the
offset moment is less and stability improves. Finally, the third graph shows that stiffening
the third type of suspension improves the rollover threshold by increasing the stabilizing
moment which this suspension supplies at the point of instability.
Following the logic of figure 16, it can be shown that the optimum situation for
maximizing roll stability (although, not necessarily for best handling or other concerns)
is for tire lift to occur simultaneously at all suspensions. That is, for the example of figure
16, stiffening the drive-axles is productive only until the angle $2 is reduced to equal $3.
Any additional stiffening of the drive-axle suspension would cause it to become a type-1
suspension for which increased stiffness is not effective. The same principle holds for the
steer axle. Finally, noting that the angle of lift off is a combined function of the effelctive
roll stiffness of the suspension and the load carried by the suspension (i.e., its maximum
load-transfer moment), then to first order, roll stability is optimized when load is
distributed among suspensions in proportion to the distribution of roll stifiess.
Roll moment Roll moment Roll moment increase instability

;*increase in suspension
' stiffness and in moment

no change in the
rnent
at the point of
A , instability
\
Stiffening a suspension that lifts off Stiffening a suspension that lifts off Stiffening a suspension that lifts off
before the peak has no effect on at the peak improves stability by after the peak improves stability by
roll stability. reducing the offset moment. increasing the load-transfer moment.
Figure 16. Increasing the roll stiffness of different suspensions has different influences on roll stabi1it:y
Other mechanisms influence static roll stability
The majority of the lateral displacement of the cg (and thus, the majority of the
destabilizing offset moment) usually results from roll motion due to the vertical and roll
compliances of the tires and suspensions. However, there are many other compliances
within the vehicle, each of which can contribute some additional lateral offset of the cg.
Two such compliances which are known to be significant are illustrated in figure 17.
These are lateral compliance of the suspension and lateral beaming of the vehicle frame.
The figure also illustrates the obvious possibility that the cargo can be placed off center
and thereby contribute to the lateral displacement of the cg.

lateral offset MI

beam com~lianceof the frame

Figure 17. Examples of other mechanisms that can contribute to the destabilizing offset moment

The significance of lateral displacements such as these can be judged by comparing

them to T12. That is, the lateral displacement of the cg is, in effect, a direct reduction of
the half-track. In round numbers, the half-track of an axle with dual tires is about 95 ~ m . ~
Thus, a 1-cm lateral deflection results in loss of stability equal to about 1 percent of the
original rigid-body stability of the vehicle. Lateral suspension deflection may be on the
order of 2 cm. Lateral beaming of the trailer may be 3 cm or more. A variety of other
compliances may each produce displacements on the order of several millimeters, and of
course, the lateral offset of the placement of the cargo can be very s~bstantial.~ While
none of these displacements may seem significant individually, the total influence can
easily account for the loss of a significant portion of the rigid-vehicle stability.
There is a general point of some significance which follows from this discussion.
That is, the roll stability of heavy vehicles typically derives from the summation of T L H
plus a large number of small influences resulting from various compliances. The
corollary observation is that virtually all compliances degrade stability. Thus,

Ninty-five centimeters is the nominal half-track as measured to the center of the dual-tire pair. This is
the appropriate point of reference-not the center of the outer tire nor the outer edge of the tire tread.
As shown in figure 14, the point of roll instability of a tractor-semitrailer combination typically occurs
when the light-side drive-axle tires first lift from the road surface. This occurs at only a few degrees of
roll and with all the tires on the heavily loaded side of the vehicle, including the inside tires of the
dual-tire pairs, still very firmly on the ground. At this point in the rollover process, however, the
vehicle is already unstable and rollover is virtually an established fact. It is only much later in the
process-and much too late to be of any significance-that the inside dual tires will lift off the road
surface.
Fluid cargos are of particular interest in this regard. They will be considered in the next chapter.
Figure 18. The rear end of a torsionally compliant flat-bed trailer rolls over nearly independently of th~e
front end

engineering judgements as to whether individual compliances are "negligible" should not

be made in isolation but should be considered in the context of all such compliances.
Finally, the torsional compliance of the vehicle frame stands out as a uniquely
important element in establishing the roll stability of some vehicles, particularly those
with flat-bed trailers. The photograph of figure 18 speaks clearly to the point. The
photograph was taken on a test track. The rollover event was unintentional and occurred
in nearly a quasi-static fashion.
The vehicle is loaded with four rolls of aluminum sheet. (The intention of the e:xercise
was to test the cargo-restraint system.) Two rolls are at the extreme front of the trailer,
directly over the drive-axle suspension. The other two rolls are at the extreme rear of the
trailer directly over the trailer suspension. The frame of the trailer is so compliant that the
front and rear of the vehicle are nearly independent bodies with respect to roll. The "front
vehicle" benefits f r o ~ nthe low cg height of the tractor mass and, as the picture sho\vs, the
"rear vehicle" has appreciably lower roll stability.
Figure 19 is an example of another vehicle whose stability might be expect to suffer
due to torsional compliance of the frame. In this case, the central location of the load

Figure 19. The actual roll stability of this vehicle is a small fraction of its "rigid-vehicle" stability
owing to roll compliance of the trailer frame
whose cg lies high above the neutral axis of the compliant frame will result in a large
lateral displacement of the cg and correspondingly large destabilizing offset moment.
Measuring rollover threshold with the tilt-table experiment
The tilt-table methodology is a physical simulation of the roll-plane experience of a
vehicle in a steady turn. The method provides a highly resolute means of determining
rollover threshold and examining the mechanism by which this limit is determined.
In this experimental method, the vehicle is placed on a tilt table and is very gradually
tilted in roll. As shown in figure 20, the component of gravitational forces parallel to the
table surface provides a simulation of the centrifugal forces experienced by a vehicle in
turning maneuvers. The progressive application of these forces by slowly tilting the table
serves to simulate the effects of quasi-statically increasing lateral acceleration in steady
turning maneuvers. The tilting process continues until the vehicle reaches the point of roll
instability and "rolls over." (The vehicle is constrained by safety straps to prevent actual
rollover.)
When the table is tilted, the
component of gravitational forces
parallel to the table surface,
W*sin($~),simulates lateral forces,
and the weight of the vehicle itself
is simulated by the component of
gravitational forces that are
perpendicular to the table (i.e.
W*COS(@T), where W is the weight
of the vehicle and @Tis the roll
$ angle of the table relative to the
true gravitational vector). Thus, the
- L forces acting during the tilt-table
test are scaled down by a factor of
c o s ( + ~ )Since
. the important
mechanisms of actual rollover
depend on the ratio of the
Figure 20. The tilt-table experiment centrifugal forces to the vertical,
gravitational forces, it is
appropriate to take the ratio of the simulated lateral acceleration forces to the simulated
weight to represent lateral acceleration when interpreting the results of a tilt-table
experiment. That is:

where:
ays is the simulated lateral acceleration (in gravitational units)
$T is the roll angle of the tilt table
W is the weight of the vehicle.
 The quality of tan(4-r) as an estimate of actual static roll stability depends, in part, on
how closely cos(9-r) approximates unity. In the tilt-table experiment, both the vertical and
lateral loading of the vehicle are reduced by the factor cos(4-r) relative to the loads they
are meant to represent. Because of the reduced vertical loading, the vehicle may rise on
its compliant tires and suspensions relative to its normal ride height, resulting in a slightly
higher cg position and, possibly, a slightly low estimate of the static roll stability limit. At
the same time, static lateral loading is also reduced by the factor cos(4~).This may result
in compliant lateral and roll motions of the vehicle that are relatively small, tending ito
produce a slightly high estimate of the static roll stability limit. The fact that these two
influences tend to cancel each other is clearly advantageous. More importantly, for tlhe
moderate angles of tilt required to test large commercial vehicles, cos(4~)remains
sufficiently near to unity such that accurate representations of all loadings are maintained.
(At a tili. angle simulating 0.35 g lateral acceleration, cos(4.r) is 0.94.)
A second error source in this physical simulation methodology involves the
distribul'ion of lateral forces among the tires of the several axles of the vehicle. Lateral
forces developed at the tire-road interface must, of course, satisfy the requirements of
static equilibrium of lateral force and yaw moments acting on the vehicle. However,
many commercial vehicle units are equipped with multiple nonsteering axles which
results in the system being statically indeterminate. Thus the distribution of lateral
reaction forces among the axles is partially dependent on the lateral compliance
properties of tires and suspensions. The compliance properties that are in play while the
vehicle is sitting on the tilt table are not precisely those that are in play while the vehicle
is in motion on the road. The significance of this error source is dependent on axle
location, and the similarity, or lack thereof, of geometry among the redundant axles and
suspensions. For many commercial vehicles, the close spacing and geometric similarity
of the two axles of each tandem suspension tend to minimize these errors.
A third error source lies in the side slip angle of the tractor and the yaw articulation
geometry of the vehicle. Although tilt-table experiments are conducted with these tvvo
yaw plane angles at zero, the negotiation of real turns at significant speed generally
implies the existence of small, nonzero yaw plane angles. Some reflection on this matter
(and that of the preceding paragraph) reveals that, in practice, static rollover threshold
varies somewhat as a function of turn radius. In this light, the zero-yaw-angle condition is
simply seen as one of many possible test conditions-certainly the one most easily
implemented.
The most fundamental aspects of the mechanics of quasi-static rollover were
presented earlier using very simplified models. Despite the high level of simplifi~ati~on,
the validity of the ideas presented have been confirmed in numerous tilt-table
e~periments.[l7~19,201
Figure 21 presents data gathered in one such experiment.[l71 The test vehicle wais a
five-axle tractor-van semitrailer equipped with air suspensions at the drive axles and the
trailer axles. The figure is a plot of trailer-body roll angle (relative to the table surfalce, of
course) as a function of simulated lateral acceleration (the tangent of the tilt angle). The
annotations in the figure point out features in the data corresponding to many of the
elements of the process which were discussed earlier in this chapter.
Tittable test of a five-axle tractor-van semitrailer combination
Introductions
The accident-data analyses presented in chapter 1 make clear that the static roll
stability is the dominate vehicle quality affecting the chance of a given heavy truck being
involved in a rollover accident. Chapter 2 reviews the mechanics of static stability.
However, virtually all rollover accidents in the real world are dynamic events to some
extent; none are truely quasi-static.
The influence of sloshing liquids and other moving loads"
In the majority of commercial truck operations, the load on the vehicle is fixed and
nominally centered. In certain cases, however, the load may be able to move on the
vehicle, with the potential of affecting the turning and rollover performance. The most
common examples of moving loads are:
- bulk, liquid tankers, with partially filled compartments,
- refrigerated vans hauling suspended meat carcasses, and
-- livestock.
The performance properties of commercial vehicles used in these applications may be
influenced by the free movement of the load in either the longitudinal or lateral
directions. This chapter will present material on the first two types of loads.
Sloshing-liquid loads
In the majority of commercial truck operations, the load on the vehicle is fixed imd
nominally centered. In certain cases, however, the load may be able to move on the
vehicle, with the potential of affecting the turning and rollover performance. The most
important of these is liquid cargo carried in tanks.
In the operation of a bulk-liquid transport vehicle, the moving load that can affect its
cornering and rollover behavior is the presence of unrestrained liquid due to partial filling
of the tank or its compartments. A compartment that is filled to anything less than iits full
capacity allows the liquid to move from side to side, producing the so-called "slosh" load
condition. Slosh is of potential safety concern because (1) the lateral shift of the load
reduces the vehicle's performance in cornering and rollover, and (2) the dynamic mlotions
of the load may occur out of phase with the vehicle's lateral motions in such a way as to
become exaggerated and thus further reduce the rollover threshold.
The motions of liquids in a tank vehicle can be quite complex due to the dependence
of the motions on tank size and geometry, the mass and viscosity of the moving liquid,
and the maneuver being performed.[e.g., 26,27,281 Fundamental analyses of sloshing
liquids in road tankers appeared in the literature from the 1970s.[e.g-,29,301 A number of
more elaborate computer studies arose in the late 1980s and early 1990~.[e~g.~ 3132,331
This discussion is constrained to basic elements that provide insight on the mechanisms
by which fluid motions influence rollover. The mechanisms of slosh are most readily

lo The material of this section was largely created by Ervin and has been presented in [25].
described in simple steady-state cornering, although it is in transient maneuvers that the
most exaggerated fluid displacements take place.
Steady turning
When a slosh-loaded tanker performs a steady-state turn, the liquid responds to lateral
acceleration by displacing laterally, keeping its free surface perpendicular to the
combined forces of gravity and lateral acceleration. Figure 22a illustrates the position of a
partial liquid load in a circular tank which is being subjected to a steady-state cornering
maneuver. The mass center of the liquid moves on an arc, the center of which is at the
center of the circular tank. In effect, the shift of the liquid produces forces on the vehicle
as if the mass of the load was located at the center of the tank.
With more complex tank shapes, even the steady-state behavior becomes somewhat
difficult to analyze. In particular, with unusual tank shapes it becomes more difficult to
describe the motion of the liquid's center of mass as a function of lateral acceleration. As
a contrast to the circular tank, figure 22b illustrates the behavior of liquid in a rectangular
tank. At low lateral accelerations, the liquid movement is primarily lateral, centered at a
point well above the tank center. Hence, its effect is similar to having a very high mass
center. With increasing lateral acceleration, the mass center follows a somewhat elliptical
path.
While the circular tank results in a vehicle with a higher load center, efforts to reduce
the load height by widening and flattening the tank can be expected to increase vehicle
sensitivity to slosh degradation of the rollover threshold. The effect is illustrated by the
plot in figure 23 taken from StrandbergL301 showing rollover threshold versus load
condition in steady-state cornering. For a circular tank, increasing load lowers the
threshold continuously due to the increasing mass of fluid free to move sideways. In this

a. Circular cross section b. Rectangular cross section

Figure 22. Illustration of liquid position in steady-state turning for circular and rectangular tanks
 case, the minimum rollover
I Rollover threshold in a steady turn
threshold occurs at full lolad.
For a vehicle having a
rectangular tank, higher levels
of rollover threshold occur
when the tank is either ernpty
or full, although at
intermediate load conditions
the rollover threshold is
severely depressed due to the
greater degree of lateral
motion possible for the
40 60 80 100 unrestrained liquid. 'Thus, the
Load, percent rectangular tank shape (hl
Figure 23. Rollover threshold in a steady turn as a function of the contrast to the circular) can
percentage of load of unrestrained liquid potentially result in rollover
thresholds with slosh load that
are less than that of the fully loaded vehicle.
Transient turning
In transient maneuvers such as an abrupt evasive steering maneuver (e.g., a rapid lane
change), slosh loads introduce the added dimension of dynamic effects. With a sudden
steering input, the rapid imposition of lateral acceleration may cause the fluid to displace
to one side with an underdamped (overshooting) type of behavior. The difference
between the steady-state and transient maneuvers are primarily a matter of the time
involved in entering the turn. The steady-state type of behavior is observed when the turn
is entered very slowly, whereas the transient behavior applies to a very rapid turning
maneuver. The difference
cI Quasi-steady-state behavior between the two is illustrated
in figure 24, which shows the
way in which the liquid
surface moves in each type of
maneuver. In effect, the liquid
motion occurs much like that
of a simple undamped
Time
pendulum. The response of the
Transient behavior liquid mass to a step input of
acceleration (as in the bottom
illustration of figure 24) would
be seen to displace to an
amplitude which is
approximately twice the level
of the steady-state amplitude.
In a lane-change maneuver in
Time
Figure 24. Motions of the surface of a sloshing load in quasi-steady
which the acceleration goes
and transient turning first in one direction and then
 Normalized spectral density the other, an even more
7 exaggerated response
amplitude can be produced.
6
In general, the degree to
g5 which the dynamic mode is
-0 4
excited depends on the timing
a,
.-
N of the maneuver. The
7i3
E unrestrained liquid will have a
22 natural frequency for its lateral
oscillation which depends on
1
the liquid level and cross-
0 sectional size of the tank. For
0 0.2 0.4 0.6 0.8 1 a half-filled, eight-foot-wide
Frequency, Hz
tanker, this frequency is
Figure 25. Frequency content distributions obtained from steering approximately 0.5 H~ (cycles
input time histories
per second); whereas, a six-
foot-diameter circular tank (typical of an 8,800-gallon tanker) would have a frequency of
approximately 0.6 Hz. As for dynamic systems in general, if the frequency content of
input (lateral acceleration) stays below this natural frequency, the response is largely
quasi-static, but if the input contains substantial power at or above the natural frequency,
the response will be dynamic. Studies of driver steering behavior have shown that in a
demanding steering task, such as an accident-avoidance maneuver, the steering input may
have significant energy near the 0.5 Hz frequency.WI Figure 25 shows the frequency
content of steering motions measured under different driving tasks. The "tracking" task
especially indicates that steering input at or near the 0.5 Hz frequency may be readily
applied by a driver. Thus it is possible to excite these dynamic motions. For example, the
two-second lane change used as a typical evasive maneuver for evaluating rearward
amplification constitutes a lateral acceleration input closely matched to the slosh
frequency.[351 Hence it must be concluded that dynamic slosh motions can be readily
excited on a tanker of normal
Rollover in a transient maneuver, 0.5 Hz size, especially in the course
of evasive maneuvers such as
a lane change.
In transient maneuvers,
the rollover thresholds are
depressed by this dynamic
motion. Figure 26 shows the
estimated rollover threshold
as a function of load for
unrestrained liquids in a
transient maneuver. In the
transient case, even the
I O
20 40 60
Load, percent
80 100
circular tank experiences
reduced rollover thresholds
Figure 26. Rollover threshold in a transient turn as a function of the
percentage of load of unrestrained liquid when partially loaded due to
the fact that the fluid can "overshoot" the steady-state level. Understandably, the elliptical
tanker is even worse.
Though the results shown are derived from analytical studies, experimental tests of
partially loaded tankers generally confirm these observations.l361
Methods for dea1i:ng.with partial liquid loads
In the vocational use of many liquid bulk haulers it is necessary to run at times with
partial loads. This is especially true with local delivery tankers hauling gasoline and
home-heating fuel. The question is what can be done to reduce the sensitivity and hence
the potential risks of using these vehicles, once a substantial fraction of their load has
been delivered? Of course, specifying a vehicle with suspension systems most resistant to
rollover is a first step. However, at least two other aids are available.
Baffles. Baffles are commonly used in tank vehicles, except in special cases where
provisions for cleaning prevent their use (such as bulk-milk haulers). However, the
common baffle arrangement is a transverse baffle intended to impede forelaft movement
of the load. These transverse baffles have virtually no utility in preventing the lateral
slosh influential to roll stability. To improve roll performance, longitudinal baffles are
required. Properly designed, they can substantially reduce the slosh degradation of
cornering and rollover performance.
Figure 27 shows the effect of
different longitudinal baffling
arrangements on an elliptical tank
design. The plot shows the relative
sensitivity to oscillation frequency
at a 50 percent load condition in
transient maneuvers. Note in tlhe
figure that the rollover limit is
always less than that of an
equivalent rigid load in the
frequency range up to 0.5 Hz,
representative of normal driving
maneuvers. The unbaffled tanlc is
most degraded by slosh at the
resonant frequency of 0.5 Hz.
.3 .4 .5 .6 .8 1. Adding one vertical baffle on the
Oscillation frequency, Hz
centerline greatly improves the
Figure 27. Approximate rollover limits as a function of
degradation in the low frequency
harmonic oscillation frequency of an elliptic tank
with 50 percent load and four types of baffles range by pushing the resonant
frequency to approximately 0.8 Hz
(above the frequency of most steering inputs). Adding three vertical baffles further
improves the rollover performance, largely by preventing the significant lateral
movement of the liquid. Horizontal baffling also aids performance, presumably by
interrupting the smooth flow of the sloshing liquid and reducing the overshoot motions.
 Rollover threshold in a step-steer maneuver Compartmentalization. A
8-foot wide semieliptical tank second, and far more
common, method for
improving cornering
performance with tankers
under partial loading
conditions is to subdivide the
tank into separate
compartments. Ideally, the
compartments are completely
emptied on an individual
basis at a drop spot. In this
0 20 40 60 80 100
case, the vehicle is never
Load, percent of total volume subject to a sloshing load.
Figure 28. Rollover threshold as a function of load percentage and The only precaution in this
fractional sloshing volume type of use is that the delivery
route.be planned to empty
from the rear of the vehicle first.
When it is not possible to completely empty each compartment, a reduced slosh
sensitivity exists, but is often not significant as long as only a fraction of the total load is
free to slosh. In these cases, the relevant parameters are the percent load being carried
and the fraction of the load that is free to slosh. Figure 28 shows an estimate of the
rollover threshold in a transient maneuver for an eight-foot-wide semielliptical tanker as a
function of these parameters. The horizontal axis represents the percent (of capacity) to
which the vehicle is loaded; whereas, the individual lines represent the fraction of the
load that is free to slosh. The worst condition is at about 45 percent load with all of the
load free to slosh (the 1.OO line). The key point illustrated in this figure is that, as long as
the sloshing load is never more than 20 percent of the total load, the rollover threshold
will not be less than that of the fully loaded
vehicle.
Hanging-meat loads
The transport of hanging meat in refrigerated
vans is another of the special cases of moving
loads that has been subject to study, both
analytically and experimentally. While these
loads are similar to that of livestock in that there
is a certain amount of lateral movement possible,
they differ in that the loads are suspended from
rails on the ceiling of the van. Thus, the load
hangs much like a pendulum, free to move within
a gap provided for air circulation inside the van,
as modeled in figure 29. In addition to the fact
that this load has an unusually high cg due to its
Figure 29. Analytical model of vehicle suspension above floor level, it also has potential
canying hanging meat
C
.451 for two other effects:
- lateral displacement of the cargo cg in either
quasi-static or dynamic fashion, and
w - a dynamic influence associated with the
4 impact of the load against the trailer wall.
time, sec
The implications of the first point have bleen
well covered, in principle, in the preceding
material. In a quasi-steady turn, the load will
swing sideways until it contacts the wall. The
lateral air gap, then, represents a limit to the
maximuml~eraldisplacement of the cargo cg.
This shift is the most significant influence which
the mobile quality of the hanging load has 0.n the
roll stability of the vehicle.
Dynamically, the pendulum nature of the load
establishes a natural frequency which is typically
a large fraction of 1 Hz. Thus the dynamic
t/
I influence of a potentially oscillating load are
1 1 2 3 similar to those for dynamically sloshing liquids.
time, sec
Regarding impact of the load with the wall of
the trailer, figure 30 shows the time-respons,e
behavior of a typical tractor-semitrailer with a

k
hanging-meat load in a step-steer maneuver as
Tractor calculated from an analytical model with features
as shown in figure 29. The first graph at the top
of the figure shows that the trailer experiences a
Trailer
turn.
history
sharpThis is due0.75
disturbance
at about tointhe
its
seconds
impact
lateralinto
of
acceleration
the
thehanging
step-steer
load

Lefl side with the trailer wall.

The second graph shows that the yaw ra.te of
the tractor is substantially influenced by thils
impact. It can be expected, therefore, that the
driver of the vehicle would be well aware of the
impact of the load with the trailer wall and ,would

\
i,ooj Right side
find it necessary to apply some steering
correction.
However, the third graph, which shows the
o I 2 3 4 roll angles of the tractor and trailer, and the fourth
time, sec
graph, which shows vertical load on the traiiler
Figure 30. Example responses of a tractor-
semitrailer with hanging-meat tires, suggest that the direct influence of this
load in a stepsteer maneuver impact on roll stability is minor. A relatively
slight "ripple" is apparent in both plots around 1
second. However, the frequency content of the impact event is so high that it substantially
exceeds the ability of the vehicle to respond in roll. The primary influence of the hanging
meat on roll stability is the lateral translation of its cg per se.
Rollover in dynamic maneuvers
In practice, quasi-static rollover is nearly impossible to accomplish even on the test
track. The analyses of chapter 2 assume that the lateral acceleration condition is a given
and is sustained (i.e., the condition defining steady state). In practice, a test vehicle can
approach rollover quasi-statically either by very slowly increasing turn radius at a
constant velocity, or by very slowing increasing velocity at a constant radius. (The former
is the more common practice). In either case the, the quasi-static condition can be made
to hold reasonably well until the tires of the drive axles lift. At this point, however, the
vehicle loses traction and typically "scrubs off' speed such that the lateral acceleration
immediately declines and the drive wheels settle back onto the surface. The process may
be repeated any number of times. Strictly speaking, there are at least two exceptions
which can allow quasi-static rollover. (1) The vehicle may be equipped with a locking
differential so that drive thrust can be maintained after lift of tires on the drive axles. (2)
Highly compliant (flat bed) trailers may rollover at the rear without lifting drive-axle tires
(figure 18). Regardless, in real world events, there is virtually always a dynamic
component to the maneuver which, at the least, provides the needed kinetic energy to
raise the cg through its apex height after tires of all axles (or at least all axles other than
the steer axle) have left the ground. However, as can be seen in figure 3 1, for vehicles
with high centers of gravity, the additional elevation of the cg which is required is not
that great.
In the context of rollover, dynamic maneuvers are those in which the frequency
content of the maneuver (and in particular, the lateral acceleration of the vehicle)
approach the natural frequency of the rolling motion of the vehicle. A lightly loading
tractor-semitrailer can be expected to have
natural frequencies in roll in the range of 2
Hz or more-well above the frequency of
steering input which the truck driver can
muster even in emergency maneuvers.
However, a heavily loaded combination with
its payload cg at a moderate height and with
suspensions of average roll stiffness is likely
to exhibit a roll natural frequency near 1 Hz.
A heavily loaded semitrailer with a high cg
and with suspensions of less-than-average
stiffness can have a roll natural frequency as
low as 0.5 Hz. As discussed in the previous
chapter (see figure 25), 0.5 Hz in particular is
well within the range of excitation
frequencies expected in emergency
Figure 31. At the least, rollover requires the maneuvering. Thus, one can expect the
dynamic momentum required to lift potential for harmonic tuning and related
the cg through its apex height.
 Time responses in a two-second lane change, 90 k/hr resonant overshoot to
0.2 promote rollover in
transient maneuvers with
0.1
higher frequency content.
ln
m However, higher
-i
0
,-

ew O
frequency maneuvers also
-w involve yaw dynamics
8m which can complicate roll
-e -0.1 behavior. Figure 32 shows
P
II1 the response behavior of a
-0.2 tractor-semitrailer during a
simulated, 2-second
-0.3 emergency lane change
0 2
Time, seconds
4 6 maneuver.[351 The figure
presents time histories of
Figure 32. In a dynamic maneuver, the acceleration of the semitrailer
lags the tractor and roll lags acceleration
lateral acceleration for the
tractor and for the
semitrailer and roll angle for the combination. When maneuvering at speed, the
semitrailer tends to follow the path of the tractor rather faithfully. Particularly with longer
vehicles, this implies a time lag between the actions of the tractor and the trailer. (Tlnis is
more a result of the tractrix geometry
-
which basically governs the motion of the
trailer rather than a true dynamic
phenomenon.) When the frequency content

expected to lag lateral acceleration. Both of

?ii/,d;

9 significant to note that when the trai1t:r

reaches its maximum roll displacement, the
tractor is well passed its peak lateral
acceleration. Consequently, at this critical
,,,,,- point, the tractor, with its relatively low cg,
is more "available" to resist rollover rhan it
would be in a demanding steady-state turn.
Lateral Rearward Amplification = A /Ayl Thus, in this maneuver, while roll
dynamics are degrading roll stability, the
. e yaw dynamics are compensating to some
extent. The situation (even in this reli2tively
simple maneuver) is complex and the: net
result depends on the tuning of the
Peak lateral accel
of the 2nd trailer frequency content of the particular
Figure 33. In rapid obstacle-avoidance maneuvers, maneuver, the frequency sensitivities of the
rearward amplification may result in vehicle in yaw, and the natural frequency
premature rollover of the rear trailer
 and damping of the vehicle in roll.
Dynamics can play a unique role
in the rollover of multiply articulated
vehicles. As illustrated in figures 33,
vehicles with more than one yaw-
articulation joint (i.e., truck-trailer
combinations, doubles, or triples)
may exhibit an exaggerated response
of the rearward units when
0 .25 0.50 .75 1.00 1.25 1.50 performing maneuvers with
Maneuver frequency, Hz unusually high frequency content.
Figure 34. Rearward amplification of a Western double (see The phenomenon is known as
figure 33) at 88 klhr as a function of maneuver
frequency and test methodL351
rearward amplification and is often
quantified, as shown in the figure,
by the ratio of the peak lateral response of the rearward unit to that of the tractor.[35l"
Figure 35, shows that rearward amplification is a strong function of the frequency
content (and the type) of the maneuver. Because rearward amplification is nearly unity at
low frequencies, these vehicles behave very well in normal driving. However, since
rearward
amplification tends
to peak in the
frequency range
characteristic of
quick, evasive
maneuvers, these
vehicles are also
The B-train has no dolly but has a 5th-wheel coupler at the rear of the 1st trailer
I I I I quite susceptible to
rollover of the rear
trailers during
emergency
maneuvering.
Numerous
approaches to reduce
rearward
amplification of
multitrailer vehicles
have been proposed,
Figure 35. The B-train and C-train, originally introduced in Canada, exhibit less most of which are
rearward amplification than the standard A-train

" The subject of rearward amplification is covered extensively in the literature. Early work in the 1960s
and1970s [37-401 was followed by a large effort, primarily by UMTRI, in the 1980s and 1990s. This
work ranges from linear analyses of closed-form [41], on the stability plane, and in the frequency
domain [42], through extensive simulation and test-track experimental studies [43,44,45], to real-world
field-test studies [46].
based on different arrangements for coupling trailers. (See [44] for a review of many types
of innovative couplings.) The most successful have been the so-called B-train and C1-train
which are compared to the reference A-train in figure 35. Both of these vehicles eliminate
the yaw and roll degrees of freedom associated with the pintle-hitch coupling between the
semitrailer and the full trailer. Eliminating.
"
the yaw articulation indirectly improves
roll stability by reducing rearward
amplification. For example, the A-train in
figure 35 would typically have a rearward
amplification of about 2 (figure 34), but
the rearward amplification of the B-train
and C-train in the figure would typically
be less than 1.5.
However, by coupling the two triailers
in roll, the B- and C-train configurat:ions
dramatically improve dynamic roll
stability directly. Figure 36 presents time
histories of an A-train in a maneuver
similar to that of figure 33. The figuire
shows that the lateral acceleration arid roll
motions of the two trailers are about 90
degrees out of phase. Thus, when the
second trailer reaches its critical condition
of maximum lateral acceleration ancl roll
angle, the first trailer has passed its peak
and returned to near-zero in these two
measures, and actually has substantial roll
momentum in the opposite direction. Thus,
when these two trailers are coupled in roll
as in a B- or C-train, the vehicle can
perform very severe lane changes (i,e.,
with peak lateral accelerations of the
tractor on the order of 0.5 g) without
experiencing rollover as it is extremely
difficult for one trailer to "drag over" its
o 2 4 6 8 out-of-phase partner. [441 (Of course:, the
Time, seconds
mechanical loads on the coupler ancl dolly
Figure 36. Time histories of an A-train double in an frame may be very high in such
evasive lane-change maneuver maneuvers, introducing the risk of
mechanical failure of these parts.)
In recognition of this powerful mechanism for improving dynamic roll stability, Ervin
introduced a performance measure known as the dynamic load transfer ratio (DLTR). [431
Where:
FL~ is the vertical load on the left-side tires of axle i
FR~ is the vertical load on the right-side tires of axle i
m is the first axle of the roll unit
n is the last axle of the roll unit
DLTR is defined for each roll unit of a vehicle, i.e., for each group of units of the vehicle
which rolls independently of other units. (For example, the A-train of figure 35 has two
roll units, the tractor semitrailer and the full trailer, but the B-train and C-train of that
figure are each single roll units.) DLTR can be calculated for each instant during a
dynamic maneuver according to equation 1. The maximum value so determined is the
DLTR of the vehicle for that maneuver. DLTR is zero for a laterally symmetric roll unit
at rest and is unity when all the tires on one side of a roll unit have lifted from the ground.
Within a given class of vehicle configuration, DLTR transfer has been shown to be
largely a direct result of the static rollover threshold (SR) and rearward amplification
(RA). In 1990, Winkler and Bogard showed that, over a broad range of A-train doubles,
rearward amplification and static rollover threshold are very good predictors (r2 = 0.91)
of DLTR. [471 Similarly, McFarlane et al., obtained very strong correlations between
DLTR and static rollover threshold within some 19 (fairly narrow) vehicle
configurations. At the same time, the differences in the correlation models among the
various configurations was very substantial. [481
The broad messages to be taken from these findings are twofold: (1) the most
powerful means for improving dynamic rollover stability of a given vehicle configuration
is to increase static roll stability or, in the case of multiply articulated vehicles, to
decerease rearward amplification; (2) dynamic roll stability varies a great deal between
configurations, and DLTR provides an effective means to compare this property among
all heavy vehicles.
1.o
.-0
1

2
, 0.9
&'
V)
s
2
u
0.8
RI
-0
.o 0.7
5
C
% 0.6
u
aC

.- 0.5
C

w"
0.4
0.4 0.5 0.6 0.7 0.8 0.9 1.0
Actual dynamic load transfer ratio
Figure 37. Estimated DLTR versus actual DLTR for
a broad range of A-train doubles [471
 4. ROLLOVER
AND THE INTELLIGENT
HIGHWAYNEHICLE
SYSTEM
Modern electronics are beginning to be applied to the problem of heavy vehicle
rollover in the form of so-called intelligent systems either on-board the vehicle or within
the highway infrastructure.
Since a disproportionate number of commercial-vehicle rollover crashes occur on exit
ramps (17 percent according to [@I), highway-infrastructure systems have concentrated
on active signing for advisory speeds on exit ramps. The methods explored vary
significantly in complexity. For example, Freedman et al. examined the effectiveness of
speed-advisory signs employing flashing lights activated when any truck was observed
entering the ramp at excessive speed. [501 On the other hand, Strickland et al., descrilbed
prototype installations which selectively display the message "Trucks reduce speed,"
based on automated observations of speed, weight, and height of individual vehicles. [sll
At least three approaches aimed at reducing the occurrence of commercial-vehicle
rollover through on-board systems are being pursued (at least at the research-and-
development level).
Perhaps the most direct method is active roll control which is intended to directly
improve the roll stability of vehicles during critical events. Kusahara et al., describe a
prototype active roll stabilizer installed on the front suspension of a medium duty
commercial truck. [521 Similar devices to be installed on all suspensions of either unit
trucks or tractor-semitrailer combinations have been described and are under
development at Cambridge University. [53,54,551
Another approach employing on-board intelligence is the roll-stability-advisory
(RSA) or rollover-warning systems. A "stability monitoring and alarm system" was
advertised for application on commercial vehicles as early as the late 1980s. [561 More
recently, RoadUser Research of Melbourne, Australia, has developed and installed a
rollover-warning system in limited numbers for use on tank vehicles. The system
produces an audible warning for the driver based on real-time measurement of lateral
acceleration which is compared to ;I
predetermined, worst-case static rollover
threshold for the vehicle. UMTRI has
developed a prototype RSA which
includes a visual display to the driver
comparing the current lateral
acceleration of the vehicle together with
the static rollover threshold of the
vehicle in left- and right-hand turn!;.
[20,571 The rollover thresholds are
calculated in real time based on sig;nals
from on-board sensors. Thresholds for
each new loading condition are
determined after only a few minutes of
Irecall recent maximumsl normal driving.
Figure 38. The driver display of the UMTRI prototype
RSA
 Another approach to the reduction of rollover crashes is active yaw control of the
vehicle intended to prevent lateral acceleration from exceeding the rollover threshold of
the vehicle. The approach uses selective application of individual wheel brakes to apply
appropriate yaw moments and/or to simply slow the vehicle. Palkovics, in association
with El-Gindy and others, has published a number of research articles based on this
approach, and the ideas presented are being introduced in commercial applications. [58-
611 UMTRI has developed and demonstrated a prototype system especially for reducing
rearward amplification in multitrailer vehicles. [20,571 Development of this system
continues with expectations of commercial application.
Sweatman, P.; Tso, Y. 1988. The influence of suspension characteristics on the rollover stability of
articulated vehicles. Australian Road Research Board, Vermont South, Victoria. 19 p. Sweatman, P.
F., Glynn, L., and George, R. M., eds. Truck Designers Sprung? Symposium on Heavy Vehicle
Suspension Characteristics. Proceedings. Vermont South, ARRB, 1988. Pp. 179-197.
Winkler, C. B. 1987. Experimental determination of the rollover threshold of four tractor-
semitrailer combination vehicles. Final report. Michigan University, Ann Arbor, Transportation
Research Institute. 65 p. Sponsor: Sandia Laboratories, Albuquerque, N.M. Report No. UMTRI-87-
3 1.
George, R, 1992. The roll-behaviour of trucks around curves. Australian Road Research Board,
Nunawading, Victoria. 47 p. Report No. ARR 238.
Winkler, C. B.; Zhang, H. 1995. Roll stability analysis of the TARVAN. Final report. Michigan
University, Ann Arbor, Transportation Research Institute. 34 p. Sponsor: Naval Air Warfare Center,
Indianapolis, Ind. Report No. UMTRI-95-24.
Ervin, R.; Winkler, C.; Fancher, P.; Hagan, M.; Krishnaswami, V.; Zhang, H.; Bogard, S. 1998. Two
active systems for enhancing dynamic stability in heavy truck operations. Michigan University, Ann
Arbor, Transportation Research Institute. 216 p. Sponsor: National Highway Traffic Safety
Administration, Washington, D.C. Report No. UMTRI-98-39.
1990. A policy on geometric design of highways and streets - 1990;AASHTO green book. American
Association of State Highway and Transportation Officials, Washington, D.C. 1104 p.
Winkler, C. B.; Bogard, S. E.; Ervin, R. D.; Horsman, A.; Blower, D.; Mink, C.; Karamihas, S.
1993. Evaluation of innovative converter dollies. Volume 11. Appendices A - H. Final report.
Michigan University, Ann Arbor, Transportation Research Institute. 268 p. Sponsor: Federal
Highway Administration, Washington, D. C. Report No. UMTRI-93-47-21 FHWAIMC-941019 (3
volumes)
Campbell, K. L.; Blower, D.; Gattis, R. G.; Wolfe, A. 1988. Analysis of accident rates of heavy-duty
vehicles. Final report. Michigan University, Ann Arbor, Transportation Research Institute. 123 p.
Sponsor: National Highway Traffic Safety Administration, Washington, D.C. Report No. UMTRI-
88-17.
Winkler, C. B.; Fancher, P. S.; MacAdam, C. C. 1983. Parametric analysis of heavy duty truck
dynamic stability. Final report. Michigan University, Ann Arbor, Transportation Research Institute.
170 p. Sponsor: National Highway Traffic Safety Administration, Washington, D.C. Report No.
UMTRI-83-13-11 DOTIHS 806 41 1.
1998. Mechanics of heavy-duty truck systems. Course notes by the staff of the Engineering Research
Division, Michigan University, Ann Arbor, Transportation Research Institute. Published by
Michigan University, Ann Arbor, Engineering Summer Conferences. Yearly since 1985. 1139 p.
UMTRI-81558
Dalzell, J. F. 1967. Exploratory studies of liquid behavior in randomly excited tanks: lateral
excitation. Southwest Research Institute, San Antonio, Tex. 50 p. Sponsor: National Aeronautics and
Space Administration, George C. Marshall Space Flight Center, Huntsville, Ala. Report No. 2.
UMTRI-48860
Komatsu, K. 1987. Non-linear sloshing analysis of liquid in tanks with arbitrary geometries. Japan
National Aerospace Laboratories, Tokyo. 15 p. International Journal of Nonlinear Mechanics, Vol.
22, NO. 3, 1987, pp. 193-207. UMTRI-76467
Krupka, R. M. 1985. Mathematical simulation of the dynamics of a military tank. Northrop
Corporation, Electro-Mechanical Division, Anaheim, Calif. 41 p. Report No. SAE 850416. UMTRI-
71555
Slibar, A.; Troger, H. 1976. Das stationaere Fahwerhalten des Tank-Sattelaujliegerzuges; Dynamic
steady state behaviour of a tractor-semitrailer-system carrying liquid load. Wien Technische
Hochschule, Austria. 21 p. Pacejka, H. B., ed. The Dynamics of Vehicles on Roads and on Railway
Tracks. Proceedings. Amsterdam, Swets and Zeitlinger, 1976. Pp. 152-172. UMTRI-40355 A08
Strandberg, L. 1978. Lateral stability of road tankers. Volume I - main report. Statens Vaeg- och
Trafikinstitut, Linkoeping, Sweden. 83 p. Sponsor: Statens Trafiksaekerhetsverk, Solna, Sweden;
Transportforskningsdelegationen, Stockholm, Sweden. Report No. 138A. UMTRI-40055
Sankar, S.; Rakheja, S.; Ranganathan, R. 1989. Directional response ofpartiallyfilled tank vehicles.
Concordia University, Montreal, Quebec, Canada. 11 p. Report No. SAE 892481. UMTRI-78979
Rakheja, S.; Sankar, S.; Ranganathan, R. 1989. Influence of tank design factors on the rollover
threshold of partialllyfilled tank vehicles. Concordia University, Montreal, Quebec, Canada. 14 p.
Sponsor: Canada, Transport Canada, Transportation Development Centre, Montreal, Quebec. Report
NO. S m 892480. UMTWI-78980
Ranganathan, R. 1993. Rollover threshold of partiallyfllled tank vehicles with arbitrary tank
geometry. Missouri University, Columbia, Department of Mechanical and Aerospace Engineering. 4
p. Institution of Mechanical Engineers. Proceedings. Part D: Journal of Automobile Engineering,
Vol. 207, NO.D3, 1993, pp. 241-244. UMTRI-85810
McLean, J. R.; Hoffmann, E. R. 1973. The effects of restricted preview on driver steering control
and performance. Melbourne University, Department of Mechanical Engineering, Australia. 10 p.
Human Factors, Vol. 15, No. 4, Aug 1973, pp. 421-430. Sponsor: Australian Road Research Board,
Victoria. UMRI-51452
IS0 14791. Road vehicles-Heavy commercial vehicle combinations and articulated buses-Lateral
stability test procedures. International Organization for Standardization, Geneva, Switzerland. 30 p.
Culley, C.; Anderson, R. L.; Wesson, L. E. 1978. Effect of cargo shifring on vehicle handling. Final
report. Dynamic Science, Inc., Phoenix, Ariz. 140 p. Sponsor: Federal Highway Administration,
Bureau of Motor Carrier Safety, Washington, D.C. Report No. 3989-78-221 FHWA-RD-78-76.
UMTRI-41111
Jindra, F. 1965. Handling characteristics of tractor-trailer combinations. General Motors
Corporation, Defense Research Laboratories, Santa Barbara, Calif. 14 p. Report No. SAE 6507:20.
WTRI-01083
Hazemoto, T. 1973. Analysis of lateral stability for doubles. Mitsubishi Motors Corporation, Japan.
18 p. Report No. SAE 730688. UMTRI-27878
Hales, F. D. 1975. 'The rigid body dynamics of road vehicle trains. Loughborough University of
Technology, Leicestershire, England. 6 p. Vehicle System Dynamics, Vol. 4, No. 2-3, July 1975, pp.
104-109. UM'TRI-52530
Strandberg, L.; Nordstroem, 0 . ; Nordmark, S. 1975. Safety problems in commercial vehicle
handling. Statens V'aeg- och Trafikinstitut, Fack, Sweden. 66 p. Commercial Vehicle Braking and
Handling Symposium. Proceedings. Ann Arbor, HSRI, 1975. Pp. 463-528. UMTRI-32725 A161
Fancher, P. S. 1982:.The transient directional response of full trailers. Michigan University, Ann
Arbor, Transportation Research Institute. 22 p. SAE Transactions 1982. Volume 91. Warrendale,
SAE, 1983. Sponsor: National Highway Traffic Safety Administration, Washington, D.C. Report
NO. SAE 821259. UMTliI-47550
Ervin, R. D.; Fancher, P. S.; Gillespie, T. D.; Winkler, C. B.; Wolfe, A. 1978. Ad hoc study of
certain safety-related aspects of double-bottom tankers. Final report. Highway Safety Researchi
Institute, Ann Arbor, Mich. 78 p. Sponsor: Michigan State Office of Highway Safety Planning,
Lansing. Report No. UM-HSRI-78-18-1. UMTRI-40219
Ervin, R. D.; Guy, Y. 1986. The influence of weights and dimensions on the stability and control of
heavy-duty trucks in Canada. Volume I - technical report. Final report. Michigan University, Ann
Arbor, Transportation Research Institute. 277 p. Vehicle Weights and Dimensions Study, Volume 1,
1986. Sponsor: Canroad Transportation Research Corporation, Ottawa, Ontario, Canada. Repoirt No.
UMTRI-86-3511. UMTRI-73754
Winkler, C. B.; Fancher, P. S.; Carsten, 0 . ; Mathew, A.; Dill, P. 1986. Improving the dynamic
performance of multitrailer vehicles: a study of innovative dollies. Volume I -technical report. Final
report. Michigan University, Ann Arbor, Transportation Research Institute. 246 p. Sponsor: Federal
Highway Administration, Washington, D.C. Report No. UMTRI-86-26/U FHWAIRD-861161.
UMTRI-744 11
Winkler, C. B.; Bogard, S. E.; Ervin, R. D.; Horsman, A.; Blower, D.; Mink, C.; Karamihas, S.
1993. Evaluation of innovative converter dollies. Volume I. Technical report. Final report. Michigan
University, Ann Arbor, Transportation Research Institute. 108 p. Sponsor: Federal Highway
Administration, Washington, D. C. Report No. UMTRI-93-47-l/FHWA/MC-941019 (3 volumes)
UMTRI-85474
Winkler, C. B.; Bogard, S. E.; Bowen, M. A.; Ganduri, S. M.; Lindquist, D. J. 1995. An operational
field test of long combination vehicles using ABS and C-dollies. Volume I: final technical report.
Michigan University, Ann Arbor, Transportation Research Institute. 168 p. Sponsor: National
Highway Traffic Safety Administration, Washington, D.C. Report No. UMTRI-95-45-1. UMTRI-
88519
Winkler, C. B.; Bogard, S. E. 1993. Simple predictors of the pegormance of A-trains. Michigan
University, Ann Arbor, Transportation Research Institute. 12 p. Heavy Vehicle Dynamics and
Stability. Warrendale, SAE, 1993. Pp. 145-156 Also published in SAE Transactions, 1993. Volume
102. Warrendale, SAE, 1994. Sponsor: Federal Highway Administration, Washington, D.C. Report
NO. SAE 932995. UMTRI-84991 A13
McFarlane, S.; Sweatman, P.; Dovile, P.; Woodrooffe, J. 1997. The correlation of heavy vehicle
performance measures. Roaduser Research Pty., Ltd. Published by Society of Automotive
Engineers, Warrendale, Pa. 12 p. Heavy Vehicle and Highway Dynamics (SAE-SP-1308).
Warrendale, SAE, 1997. Pp. 2 1-30. Report No. SAE 973 190. UMTRI-90867
Ervin, R. D.; Barnes, M.; MacAdam, C. C.; Scott, R. 1985. Impact of specific geometric features on
truck operations and safety at interchanges. Volume I - technical report. Final report. Michigan
University, Ann Arbor, Transportation Research Institute. 132 p. Sponsor: Federal Highway
Administration, Washington, D.C. Report No. FHWA/RD-8610571UMTRI-85-3311, UMTRI-73167
Freedman, M.; Olson, P. L.; Zador, P. L. 1992. Speed actuated rollover advisory signs for trucks on
highway exit ramps. Insurance Institute for Highway Safety, Arlington, Va.1 Michigan University,
Transportation Research Institute, Ann Arbor. 19 p. UMTRI-84328
McGee, H. W.; Strickland, R. R. 1994. An automatic warning system to prevent truck rollover on
curved ramps. Bellomo-McGee, Inc., Vienna, Va. 6 p. Public Roads, Vol. 57, No. 4, Spring 1994,
pp. 17-22. UMTRI-59109
Kusahara, Y.; Li, X.; Hata, B.; Watanabe, Y. 1994. Feasibility study of active roll stabilizer for
reducing roll angle of an experimental medium-duty truck. Nissan Diesel Motor Company, Ltd.,
Saitama, Japan. 6 p. International Symposium on Advanced Vehicle Control 1994. Proceedings.
Tokyo, Society of Automotive Engineers of Japan, 1994. Pp. 343-348. Report No. SAE 9438501.
UMTRI-86557 A35
Lin, R. C.; Cebon, D.; Cole, D. J. 1994. An investigation of active roll control of heavy road
vehicles. Cambridge University, Engineering Department, England. 14 p. Shen, Z., ed. The
Dynamics of Vehicles on Roads and on Tracks. Proceedings of 13th IAVSD Symposium. Lisse,
Swets and Zeitlinger, 1994. Pp. 308-321. UMTRI-85817 A12
Lin, R. C.; Cebon, D.; Cole, D. J. 1996. Optimal roll control of a single-unit lony. Cambridge
University, England. 11 p. Institution of Mechanical Engineers. Proceedings. Part D: Journal of
Automobile Engineering, Vol. 210, 1996, pp. 45-55. UMTRI-89198
Lin, R. C.; Cebon, D.; Cole, D. J. 1996. Active roll control of articulated vehicles. Cambridge
University, Department of Engineering, England. 27 p. Vehicle System Dynamics, Vol. 26, No. 1,
July 1996, pp. 17-43. UMTRI-60130.
Preston-Thomas, J.; Woodrooffe, J. H. F. 1990. A feasibility study of a rollover warning device for
heavy trucks. National Research Council Canada, Vehicle Dynamics Laboratory, Ottawa, Ontario.
76 p. Sponsor: Transportation Development Centre of Canada, Montreal, Quebec. Report No. TP
10610El TR-VDL-005. UMTRI-82874
Winkler, C., Fancher, P. and Ervin, R. 1998. Intelligent systems for aiding the truck driver in vehicle
control. The University of Michigan Transportation Research Institute. Report No. SAE 1999-01-
1301. 14 p. 1998. Warrendale, Pa.
Palkovics, L.; El-Gindy, M. 1993. Examination of different control strategies of heavy-vehicle
performance. National Research Council Canada, Ottawa, Ontario. 14 p. Ahmadian, M., ed.
-
Advanced Automotive Technologies 1993. [Proceedings]. New York, American Society of
Mechanical Engineers, 1993. Pp. 349-362. UMTRI-86103 A12
Palkovics, L.; Semsey, A,; Ilosvai, L.; Koefalvy, G. 1994. Safety problems and control of a typical
European truckltandem-trailer vehicle configuration. Budapest Technical University, Hungary1
Hungarocamion Road Safety Division, Hungary. 8 p. ISATA International Symposium on
Automotive Technology and Automation, 27th. Proceedings for the Dedicated Conference on Road
 and Vehicle Safety. Croydon, Automotive Automation Ltd., 1994. Pp. 303-310. Sponsor: Ilungairian
National Scientific Research Fund. Report No. 94SF010. UMTRI-87370 A18
60. Palkovics, L.; El-Gindy, M. 1995. Design of an active unilateral brake control system for five-axle
tractor-semitrailer based on sensitivity. National Research Council Canada, Ottawa, Ontario. 34 p.
Vehicle System Dynamics, Vol. 24, No. 10, Dec 1995, pp. 725-758. UMTRI-59877
61. Palkovics, L.; Michelberger, P.; Bokor, J.; Gaspar, P. 1996. Adaptive identification for heavy-tnlck
stability control. Budapest Technical University, Hungary. 18 p. Segel, L., ed. The Dynamics of
Vehicles on Roads and on Tracks. Proceedings of the 14th IAVSD Symposium. Lisse, Swets and
Zeitlinger, 1996. Pp. 502-518. Sponsor: Hungarian Scientific Research and Development Fund.
UMTRI-88885 A18
 APPENDIX A.
NOTESON ACCIDENT-DATA SOURCES

GENERAL ESTIMATES SYSTEM: GES is compiled by the National Center for

Statistics and Analysis (NCSA) within the National Highway Traffic Safety
Administration (NHTSA). The file incorporates data from a probability-based, nationally
representative sample of police-reported crashes. It covers all motor vehicle types,
including medium and heavy trucks. All police-reportable crashes are included.
Approximately 54,000 crashes are sampled each year. The police accident report (PAR)
is the sole source of data. Frequencies based on the GES file reported in the tables in this
report are national estimates, calculated using an appropriate weighting variable. Since
GES is a sample file, estimates are subject to sampling error. The GES file includes data
for every vehicle in a sampled crash whether it was towed due to damage or towed for
some other reason.
FATALITY ANALYSIS REPORTING SYSTEM: FARS is compiled by the National
Center for Statistics and Analysis within NHTSA. The file contains data on a census; of
fatal traffic crashes within the 50 states, the District of Columbia, and Puerto Rico. FARS
includes records for all crashes involving a motor vehicle on a trafficway that resulted in
the death of a vehicle occupant or nonmotorist within 30 days of the crash. Trained
employees within each state code over 100 data elements from a variety of state
documentary sources. These data are then transmitted to a central computerized database
and compiled into the FARS file by NHTSA.
TRUCKS INVOLVED IN FATAL ACCIDENTS: The University of Michigan
Transportation Research Institute produces the TIFA file. TIFA contains detailed
information on all medium and heavy trucks involved in fatal crashes in the United
States, including Alaska and Hawaii. TIFA consists of a random sample of straight trucks
with no trailers and tractor-semitrailers (as recorded in FARS) and all remaining medium
and heavy trucks involved in a fatal crash. The file combines information from the FARS
file, police accident reports, and comprehensive telephone interviews conducted by
UMTRI research staff. TIFA includes most FARS variables, supplemented with a
detailed description of each involved truck collected by the TIFA interview process.
Mississippi does not supply police reports, precluding the TIFA interview process, so
truck configuration is derived from FARS variables for Mississippi cases.
TRUCK AND BUS CRASH FACTBOOK: T&BFB is complied by the Center for
National Truck Statistics at the University of Michigan Transportation Research Institute.
It covers crashes for trucks and buses in reportable crashes. A truck is defined as a
vehicle equipped for carrying property and having two or more axles and six or moire
tires or a vehicle displaying a hazardous material placard. A bus is defined as a vehiicle
designed to carry at least 16 people. Reportable crashes are those with at least one of the
following: a fatality; an injury requiring immediate transportation from the scene folr
medical attention; a towaway. The factbook is based on multiple data source: Motor
Carrier Management Information System, General Estimate System, Fatality Analysis
Reporting System, and Trucks Involved In Fatal Accidents.
 BUREAU OF MOTOR CARRIER SAFETY: Until the end of 1991, the Bureau of
Motor Carrier Safety (BMCS) required all regulated motor carriers in interstate
commerce to report any accidents involving their vehicles which involved death, injury,
or property damage exceeding $4400. Excluded are accidents involving only boarding or
alighting from a stationary vehicle, loading or unloading cargo, or farm-to-market
agricultural transport. Reporting was done by the carriers themselves, not by a policing
agency. Reporting includes description of the vehicle by configuration, body type,
number of axles, length, width, height, gross load, and payload. Accident description is
minimal. Accidents are categorized as collision (with another vehicle) or noncollision
(single vehicle). Only noncollision accidents are further characterized as rollover,
jackknife, and other.
 APPENDIX B.
ANALYSESOF THE INFLUENCE OF ROLL STABILITY
ON ROLLOVER ACCIDENTS

Percent of accidents involving rollover as a function of roll stability

This analysis is based on BMCS accident data (see appendix A) from the years 1987
through 1991. The BMCS data source is the most appropriate for the purpose among
those available in that:
- the file is relatively large, covering all regulated U.S. motor carriers in interstate
commerce;
- the reporting criteria are sufficiently broad as to provide an accident sample
which is relatively unbiased with respect to severity,
- rollover events are identified, although only in single-vehicle accidents,
- the description of the commercial vehicle in the accident is relatively exten~sive.
The disadvantages of the file include the fact that it is limited to interstate carriers. This
probably biases the sample (relative to national norms) toward the better, restricted-
access road network, rural roads, more experienced drivers, and better maintained
vehicles. The data are also self-reported by the carriers, not by a policing agency, and are
thought to probably contain more coding errors than other data files. The reporting
system was terminated at the end of 1991.
Vehicle roll stability is not, of course, reported in the data. However, vehicle
configuration, number of axles, body style, height, width, length, empty weight, cargo
weight, and gross weight are all reported. In order to allow the best possible estimate of
roll stability, this analysis is limited to accidents involving three-axle tractors in
combination with two-axle van semi-trailers with trailer lengths in excess of 12 metlers
(40 feet) and total vehicle tare weights. This defines the most common commercial
vehicle in the U.S. (By far the most common van-trailer lengths in the 1987-1991 time
period were 48 and 45 feet.) Given this constraint, it is then possible to estimate rollover
threshold based on (1) an assumption of typical tare-vehicle properties, including the
relevant tire and suspension properties, and the weights and cg heights of unsprung and
empty sprung masses; (2) a reasonable estimate of a representative height for payload cg;
and (3) the weight data from the accident file. A presentation of the parameters used
appears in [7].12Of primary importance is the choice of the nominal payload cg height of
203 mm (80 inches) applied to all loads. An extensive presentation of the rationale for
this choice is also presented in [7]. As part of this rationale (and to cull obvious reporting
errors from the BMCS data) the analysis is limited to vehicles weighing between 11.4
and 36.4 metric tons (25,000 and 80,000 pounds). The relationship between gross vehicle
weight and roll stability which results appears in figure B-1.

'' The parameters and rationale are, of course, appropriate to vehicles and practices of the 1980s. The
results presented in figure B-1 are also.
I
The analysis is also
Rollover threshold as function of gross weight restricted to single-vehicle
I
accidents (in addition to being
' I n restricted to five-axle tractor-
1 C
c
3 van semitrailer combinations).
-,C 0.7 The BMCS reporting system
.-0
4.- first divides accidents into
.-m>
Y

"collisions7' and
0.6
"noncollisions" (single
vehicle). Further description of
the accident as a rollover,
jackknife, etceteras, is only
provided for the single-vehicle
subset of accidents. Thus the
restriction to single-vehicle
accidents.
10,000 20,000 30,000 Table B-1 presents the
I
Gross vehicle weight, kilograms numbers of single-vehicle
I

Figure B-1. Estimated rollover threshold as a function of gross accidents from the BMCS files
vehicle weight G
of 1987 through 1991 for the
vehicles of interest. Counts are
provided for accidents characterized as rollovers, jackknifes, all other, and the total. The
totals for all five years are also presented. In every case, the counts are presented as a
function of gross vehicle weight in categories spanning 1140 kilograms (2500 pounds).
The summated data for all five years was used to prepare figure B-2. This figure
shows the percentages of
BMCS 1987-1991 single-vehicle accidents
five-axle tractor van-semitrailer (of the subject vehicle)
which were categorized as
rollover, jackknife, or
others, all as a function of
gross vehicle weight. The
plot shows clearly that
heavy vehicles are more
susceptible to rollover,
light vehicles are more
susceptible to jackknife,
and that other accident
types are relatively
insensitive to weight.

10,000 20,000 30,000 40,000

Top of weight class, kilograms

Figure B-2. Types of accidents as a percent of all single-vehicle accidents

42
 Note that percentages for rollover accidents and jackknife accidents are interrelated.
That is, while the physics of vehicle behavior certainly suggests that the rollover
accidents should generally increase with weight because roll stability decreases with
weight, it is also true that some of the relative decrease in rollover for light vehicles could
result simply from an increasing number of jackknife accidents for light vehicles.
That is, the physics of vehicle behavior certainly suggests that rollover accidents
should generally increase with weight because roll stability decreases with weight, but
physics also suggests that jackknife accidents should increase for lighter vehicles because
of issues of brake-force proportioning. Then mathematically, it is clearly possible th,at
some of the relative decrease of rollover for light vehicles which appears in figure B-2
could simply be the result of increasing numbers of jackknife accidents. To properly
examine the influence of physical stability on the tendency to rollover in accident events,
it is appropriate to rernove the influence of the jackknife accidents.
Figure 3 in the main text, then, was produced by (1) using the relationship of figlure
B-1 to determine a representative rollover threshold for the weight categories of table B-
1, and (2) calculating the percentage of rollover accidents in single-vehicle accidents
excluding jackknife accidents.
Rollover accident rate as a function of roll stability
In appendix G of [22], GES data were used to show that, during the 1988-1990 time
period, tractor-semitrailer combinations in the U.S. experienced an average of 8697
rollover accidents per year. The same source shows that this group of vehicles traveled an
average of 53,430 million kilometers per year (33,228 million miles per year). Frornl a
census survey conducted from 1980 to 1985, Campbell et al. produced the distribution of
tractor-semitrailer travel by weight as shown in figure B-3.[231 These data can be
combined with the BMCS data of table B-1 and the stability function of figure B-1 t.o
estimate the influence of
U.S. tractor-semitrailer travel, 1980-1985
rollover threshold on rollover
accident rate of tractor-
semitrailers. The analysis
assumes, of course, that the
various data are compatible
even though they are collected
from various sources covering
somewhat different times (all
nominally in the 1980s,
however) and come from
different vehicle populations. In
particular, it must be assurned
that the distribution of rollover
probablity for single-vehicle
accidents of five-axle tractor-
Gross combination weight, thousands of kilograms van semitrailers used by
Figure B-3. Distribution of travel of tractor-semitrailer interstate carriers (the BMCS
combinations by gross weight
data) is applicable across the broader class of all accidents of all U.S. tractor-semitrailer
combinations.
The analysis proceeds as follows: (1) Estimate that 95 percent of tractor-semitrailer
rollovers occur to vehicles in the weight range covered by table B-I, implying that US.
tractor semitrailers in that range experience a total of 8262 rollovers per year. (2) Use the
five-year-total counts of rollover accidents from table B-1 to obtain the distribution of
8262 rollovers per year by weight. (3) Use the distribution of travel by weight in figure
B-3 to obtain the distribution by weight of the 53,430 million kilometers traveled per year
by tractor-semitrailers. (4) Normalize appropriate rollover yearly rates by travel yearly
rates to obtain rollovers per million kilometers by weight. (The counts from two rollover
categories are summed to match with one travel category. The process cannot be
completed for some of the lightest and some of the heaviest categories.) (5) Translate
weight to stability by the function of figure B-I.
The results of the analysis appear in figure 4 of the main text.
 APPENDIX C.
A PARTIAL
BIBLIOGRAPHY OF
THE LITERATURE ON HEAVY-VEHICLE ROLLOVER

The following are arranged chronologically and then alphabetically by author. All
titles are included in the UMTRI library collection. The final number in each entry is the
UMTRI accession number.

Erz, K. 1964. Ueber die Kippsicherheit von Lastwagen. Teil 1; On the rollover safety of trucks. Part 1.5 p.
ATZ, Vol. 66, NO. 10, Oct 1964, pp. 293-297. UMTRI-11962
Erz, K. 1964. Ueber die Kippsicherheit von Lastwagen. Teil2; On the rollover safety of trucks. Part 2 . 6 p.
ATZ, Vol66, NO. 11, NOV1964, pp. 341-346. UMTRI- 11963
Goerge, E. 1964. Querneigung und Kippgrenze des Sattelkraftfahrzeuges; Lateral sway and maximum
angle of tilt of articulated vehicles. 6 p. ATZ, Vol. 66, No. 12, Dec 1964, pp. 371-376. UMTRI-28038
Dalzell, J. F. 1967. Exploratory studies of liquid behavior in randomly excited tanks: lateral excitation.
Southwest Research Institute, San Antonio, Tex. 50 p. Sponsor: National Aeronautics and Space
Administration, George C. Marshall Space Flight Center, Huntsville, Ala. Report No. 2. UMTRI-48860
Brockington, P. A. C. 1969. Roll stiffness could spell danger. 1 p. Commercial Motor, Vol. 130, No. 3338,
5 Sept 1969, p. 57. UMTliI-11629
Isermann, H. 1970. Overturning limits of articulated vehicles with solid and liquid load; Die Kippgrenze
von Sattelkraftfahrzeugen mit fester und fluessiger Ladung. 70 p. Originally published in Dt. Kraftforsch.
& Strass. -Verk. Tech., No. 200, 1970. UMTRI-15420
Unruh, D. H. 1971. Determination of wheel loader static and dynamic stability. Caterpillar Tractor
Company, Peoria, Ill. 7 p. Report No. SAE 7 10526. UMTRI- 17014
Wilkins, H. A. 1971. The stability of articulated vehicles. Road Research Laboratory, Crowthorne,
England. 5 p. Journal of Automotive Engineering, Vol. 2, No. 3, March 1971, pp. 13-17. UMTRI-502,64
Young, R.; Scheuerman, H. 1971. Test for vehicle rollover procedure. Final report. National Aviation
Facilities Experimental Center, Atlantic City, N.J. 63 p. Sponsor: National Highway Traffic Safety
Administration, Washington, D.C. Report No. 171- 11 DOTMS 800 615. UMTRI- 17060
Holmes, K. E. 1972. Articulated vehicle roll: effect of inverting the coupling. Transport and Road Research
Laboratory, Crowthorne, England. 11 p. Report No. TRRL LR 464. UMTRI-19239
Moncarz, H. T. 1972. Stability and aerodynamic response of an articulated vehicle with roll freedom.
Maryland University, College Park. 109 p. UMTRI-40271
Riley, B. S. 1972. Does roll frequency affect commercial vehicle roll-over? Transport and Road Research
Laboratory, Crowthorne, England. 2 p. Commercial Motor, Vol. 135, No. 3548, 17 March 1972, pp. 42-43.
UMTRI- 19032
1972. Highway accident report; tank-truck combination overturn onto Volkswagen Microbus followed by
fire: U.S. Route 61 1, Moscow, Pennsylvania, September 5, 1971. National Transportation Safety Board,
Washington, D.C. 34 p. Report No. NTSB-HAR-72-6. UMTRI-19968
1973. Highway accident report; multiple-vehicle collision followed by propylene cargo-tank explosioin,
New Jersey Turnpike, Exit 8, September 21, 1972. National Transportation Safety Board, washing tor^,
D.C. 40 p. Report No. NTSB-HAR-73-4. UMTRI-28808
Moncarz, H. T.; Barlow, J. B.; Hawks, R. J. 1973. Stability and cross-wind response of an articulated
vehicle with roll freedom. Maryland University, Baltimore. 26 p. Stephens, H. S., ed. Advances in Road
Vehicle Aerodynamics. 1973. BHRA Fluid Engineering, Cranfield, 1973. Pp. 101-126. UMTRI-28929
A07
Smith, N. P.; Barker, D. 1974. The operational safety of articulated vehicles. First report: literature survey
on articulated vehicle stability and proposals for further research. Motor Industry Research Association,
Lindley, England. 58 p. Sponsor: Great Britain Ministry of Defence, Military Vehicles and Engineering
Establishment, London. Report No. AFC.911. UMTRI-31787
Wagner, W. A. 1974. Liquid rocket metal tanks and tank components - NASA space vehicle design criteria
(chemical propulsion) Rockwell International. 166 p. Sponsor: National Aeronautics and Space
Administration, Lewis Research Center, Cleveland, Ohio. Report No. NASA-SP-8088. UMTRI-48942
Watson, T. H. 1974. Heavy duty truck tandem suspension for onloff highway applications. Oshkosh Truck
Corporation, Oshkosh, Wisc. 8 p. Report No. SAE 740306. UMTRI-29678
1974. Highway accident report; Hoppy's Oil Service, Inc., truck overturn and fire, state route 128,
Braintree, Massachusetts, October 18, National Transportation Safety Board, Washington, D.C. 41 p.
Report No. NTSB-HAR-74-4. UMTRI-30759
Grueters, H. 1975. Ein Hinweis auf parametererregte Wankschwingungen von Kraftfahrzeugen; Parameter
induced rolling vibrations of motor vehicles. 3 p. Automobil-Industrie, 20. Jahrgang, 4-1975, Nov 1975,
pp. 65-67. UMTRI-52721
Miller, D. W. G.; Barter, N. F. 1975. Roll-over of articulated vehicles. Motor Industry Research
Association, Lindley, England. 17 p. Vehicle Safety Legislation - Its Engineering and Social Implications.
Discussion Volume. London, Mechanical Engineering Publications, Ltd., 1975. Pp. 91-107. Sponsor: Great
Britain Department of the Environment, London. Report No. C203173. UMTRI-33685 A10
Strandberg, L.; Nordstroem, 0 . ; Nordmark, S. 1975. Safety problems in commercial vehicle handling,
Statens Vaeg- och Trafikinstitut, Fack, Sweden. 66 p. Commercial Vehicle Braking and Handling
Symposium. Proceedings. Ann Arbor, HSRI, 1975. Pp. 463-528. UMTRI-32725 A16
Tyden, T. 1975. Foerares upplevelse av vaeltningsrisk hos tankbilar en enkaetundersoekning; Driver
apprehension about the overturning risk of tankers. Statens Vaeg- och Trafikinstitut, Stockholm, Sweden.
22 p. Report No. Internrapport 197. UMTRI-31789
Slibar, A,; Troger, H. 1976. Das stationaere Fahrverhalten des Tank-Sattelaufliegerzuges; Dynamic steady
state behaviour of a tractor-semitrailer-system carrying liquid load. Wien Technische Hochschule, Austria.
21 p. Pacejka, H. B., ed. The Dynamics of Vehicles on Roads and on Railway Tracks. Proceedings.
Amsterdam, Swets and Zeitlinger, 1976. Pp. 152-172. UMTRI-40355 A08
Lidstroem, M. 1977. Vaeltningsstabilitet foer tankfordon - med och utan laengsgaaende skvalpskott; Road
tanker overturning - with and without longitudinal baffles. Statens Vaeg- och Trafikinstitut, Linkoeping,
Sweden. 57 p. Report No. 115. UMTRI-37223
Lidstroem, M. 1977. Vaeltningsstabilitet foer tankfordon - med och utan laengsgaaende skvalpskott; Road
tanker overturning - with and without longitudinal baffles. Statens Vaeg- och Trafikinstitut, Linkoeping,
Sweden. 57 p. Report No. 115. UMTRI-37223
1977. Rules for road tank vehicles for flammable liquids. Standards Association of Australia, North
Sydney. 46 p. Report No. AS 2016-1977. UMTRI-46416
Ervin, R. D. 1978. The dynamic stability of fuel-carrying double-tanker trucks in Michigan. Highway
Safety Research Institute, Physical Factors Division, Ann Arbor, Mich. 19 p. HSRI Research Review, Vol.
8, No. 6, May-June 1978, pp. 1-19. Sponsor: Michigan State Office of Highway Safety Planning, Lansing.
UMTRI-5385 1
Ervin, R. D.; Fancher, P. S.; Gillespie, T. D.; Winkler, C. B.; Wolfe, A. 1978. Ad hoc study of certain
safety-related aspects of double-bottom tankers. Final report. Highway Safety Research Institute, Ann
Arbor, Mich. 78 p. Sponsor: Michigan State Office of Highway Safety Planning, Lansing. Report No. UM-
HSRI-78- 18- 1. UMTRI-40219
Favero, J. L.; Choulet, R. 1978. Renversement des poids lourds; Overturning of lorries. 10 p. FISITA.
International Congress. XVIIth. Volume I. Budapest, OMKDK Technoinform, 1978. Pp. 639-648.
UMTRI-41963 A21
Grimault, J. T. 1978. Les ensembles articules et la securite; Articulated vehicles and safety. Organisme
National de Securite Routiere, France. 4 p. Poids Lourd, May 1978, p. 39-44. UMTRI-41735
Kemp, R. N.; Chinn, B. P.; Brock, G. 1978. Articulated vehicle roll stability: methods of assessment and
effects of vehicle characteristics. Transport and Road Research Laboratory, Crowthorne, England. 42 p.
Report No. TRRL LR 788. UMTRI-40397
Lindstroem, M.; Strandberg, L. 1978. Influence from liquid load motion on the lateral stability of road
vehicles. 9 p. FISITA. International Congress. XVIIth. Volume 111. Budapest, OMKDK Technoinform,,
1978. Pp. 1727-1735. UM.TR1-41965 A14
Nordstrom, 0.;Nordmark, S. 1978. Test procedures for the evaluation of the lateral dynamics of
commercial vehicle combinations. Statens Vaeg- och Trafikinstitut, Linkoeping, Sweden. 7 p. Automobil-
Industrie, 23. Jahrgang, Ns. 2, June 1978, pp. 63-69. UMTRI-53893
Ranney, T. A. 1978. Analysis of heavy truck accident data. Calspan Field Services, Inc., Buffalo, N.Y. 144
p. Sponsor: Motor Vehicle Manufacturers Association, Detroit, Mich. Report No. CAL ZQ-6100-V-1.
UMTRI-41135
Strandberg, L. 1978. Lateral stability of road tankers. Volume I - main report. Statens Vaeg- och
Trafikinstitut, Linkoeping, Sweden. 83 p. Sponsor: Statens Trafiksaekerhetsverk, Solna, Sweden;
Transportforskningsdelegationen, Stockholm, Sweden. Report No. 138A. UMTRI-40055
Mallikarjunarao, C.; Fancher, P. S. 1979. Tank trailer stability analysis. Highway Safety Research Institute,
Ann Arbor, Mich. 153 p. Sponsor: Fruehauf Corporation, Detroit, Mich. Report No. UM-HSRI-79-100.
UMTRI-47 197
Neilson, I. D.; Kemp, R. N.; Wilkins, H. A. 1979. Accidents involving heavy goods vehicles in Great
Britain - frequencies and design aspects. Transport and Road Research Laboratory, Crowthorne, England.
14 p. Report No. TRRL SR 470. UMTRI-42055
Sweatman, P.; Little, L. 1979. Articulated vehicle stability - phase 1 review and research proposal.
Australian Road Research Board, Vermont South. 80 p. Report No. AIR 323-1. UMTRI-45152
Ervin, R. D. 1980. Unintended responses of heavy trucks to braking or steering inputs. Highway Safety
Research Institute, Physical Factors Division, Ann Arbor, Mich. 11 p. HSRI Research Review, Vol. 10, No.
5, MarchIApril 1980, pp. 1-11. UMTRI-54591
Ervin, R. D.; Mallikarjunarao, C.; Gillespie, T. D. 1980. Future configuration of tank vehicles hauling
flammable liquids in Michigan. Volume I - technical report. Final report. Highway Safety Research
Institute, Ann Arbor, Mich. 243 p. Sponsor: Michigan Department of State Highways and Transportation,
Lansing. Report No. UM-HSRI-80-73- 1. UMTRI-45258
Ervin, R. D.; Segel, L. 1980. Contrasts in the dynamic properties influencing the pre-crash safety of cars
and heavy trucks. Highway Safety Research Institute, Ann Arbor, Mich. 14 p. American Association for
Automotive Medicine. 24th Conference. Proceedings. Morton Grove, AAAM, 1980. Pp. 273-286.
UMTRI-45664 A23
Krisper, G . ;Reichweger, Y.; Puehringer, P. 1980. Einfluss von Fahrzeugparametern auf die Fahrdynamik
von Laskraftwagen; Effect of vehicle parameters on the dynamics of trucks. 8 p. VDI-Berichte, No. 367,
1980, pp. 115-122. UMTRI-47305
Riley, B. S.; Bates, H. J. 1980. Fatal accidents in Great Britain in 1976 involving heavy goods vehicles.
Transport and Road Research Laboratory, Crowthorne, England. 25 p. Report No. TRRL SR 586. UMTRI-
44788
Choulet, R. 1981. Evolution du guidage et amelioration de la tenue au renversement des vehicules lourds;
The evolution of steering and the improvement of overturning stability of the heavy commercial vehicles.
SERA CD, France. 13 p. Ingenieurs de llAutomobile, No. 10, March 1981, pp. 72-84. UMTRI-54720
Ervin, R. D. 1981. Safer gasoline tankers for Michigan - an interview with Robert D. Ervin. Highway
Safety Research Institute, Ann Arbor, Mich. 11 p. HSRI Research Review, Vol. 11, No. 5, March-April
1 9 8 1 , ~1-11.
~ . UMTRI-54687
Gibson, H. G.; Thorner, B. C.; Thomas, J. W. 1981. Slope stability warning device for articulated tractors.
Published by Patent Office, Washington, D.C. 10 p. Report No. Patent 4,284,987. UMTRI-46832
Heath, W. M. 1981. California tank truck accident survey, February 1, 1980 to January 3 1, 1981. California
Highway Patrol, Enforcement Services Division, Sacramento. 86 p. UMTRI-46881
Nordstroem, 0 . ; Nordmark, S. 1981. Tunga fordonskombinationers manoeveregenskaper vid konstant
hastighet. Slutrapport; Handling characteristics of heavy vehicle combinations at constant speed - final
report. Statens Vaeg- och Trafikinstitut, Linkoeping, Sweden. 83 p. Report No. 234. UMTRI-46831
Rice, R. S. 1981. Heavy truck pilot crash test rollover. Calspan Corporation, Advanced Technology Center,
Buffalo, N.Y. 78 p. Sponsor: National Highway Traffic Safety Administration, Washington, D.C. Report
NO. 6748-V-41 DOTIHS 805 978. UMTRI-46325
Su, T. C.; Lou, J. E.; Flipse, J. E.; Bridges, T. J. 1981. A nonlinear analysis of liquid sloshing in rigid
containers. Final report. Texas A & M University, College Station, Department of Civil Engineering. 641 p.
Sponsor: Transportation Department, Office of University Research, Washington, D.C. Report No. COE-
2401 DOTIRSPAIDMA-5018211. UMTRI-48217
Wolf, B.; Campbell, K. L.; O'Day, J. 1981. Occupant survivability in large-truck crashes; Occupant
survivability in heavy-truck crashes. Highway Safety Research Institute, Ann Arbor, Mich. 34 p. Sponsor:
Motor Vehicle Manufacturers Association, Detroit, Mich. Report No. MVSS 2061 UM-HSRI-81-55.
UMTRI-47016
Campbell, K. L. 1982. Rollover, ejection, and the potential effectiveness of restraints in heavy-truck
occupant fatalities. Michigan University, Ann Arbor, Transportation Research Institute. 6 p. Sponsor:
Motor Vehicle Manufacturers Association, Detroit, Mich.; Western Highway Institute, San Bruno, Calif.;
American Trucking Associations, Inc., Washington, D.C. Report No. SAE 821271. UMTRI-48007
Ervin, R. D.; Nisonger, R. L. 1982. Analysis of the roll stability of cryogenic tankers. Final report.
Highway Safety Research Institute, Ann Arbor, Mich. 136 p. Sponsor: Union Carbide Corporation, Linde
Division, Tarrytown, N.Y. Report No. UM-HSRI-82-32. UMTRI-70907
MacAdam, C. C. 1982. A computer-based study of the yawlroll stability of heavy trucks characterized by
high centers of gravity. Michigan University, Ann Arbor, Transportation Research Institute. 22 p. Also
published in SAE Transactions 1982. Volume 91. Warrendale, SAE, 1983. Sponsor: National Highway
Traffic Safety Administration, Washington, D.C.; Federal Highway Administration, Washington, D.C.
Report No. SAE 821260. UMTRI-47537
Mallikarjunarao, C. 1982. Road tanker design: its influence on the risk and economic aspects of
transporting gasoline in Michigan. Michigan University, Ann Arbor. 331 p. UMTRI-46816
Mallikarjunarao, C.; Ervin, R. D.; Segel, L. 1982. Roll response of articulated motor trucks during steady-
turning maneuvers. Michigan University, Ann Arbor, Transportation Research Institute. 20 p. UMTRI-
47552
Mallikarjunarao, C.; Segel, L. 1982. A study of the directional and roll dynamics of multiple-articulated
vehicles. Highway Safety Research Institute, Ann Arbor, Mich. 16 p. Wickens, A. H., ed. The Dynamics of
Vehicles on Roads and on Tracks. Proceedings. Lisse, Swets and Zeitlinger, 1982. Pp. 81-96. UMTRI-
47875 A07
1982. Highway accident report - Pacific Intermountain Express tractor cargo tank semitrailer Eagle1F.B.
Truck Lines, Inc., tractor lowboy semitrailer, collision and fire, U.S. Route 50, near Canon City, Colorado,
November 14, 1981. National Transportation Safety Board, Washington, D.C. 35 p. Report No. NTSB-
HAR-82-3. UMTRI-47499
Gillespie, T. D.; Ervin, R. D. 1983. Comparative study of vehicle roll stability. Phase V. Final report.
Michigan University, Ann Arbor, Transportation Research Institute. 46 p. Sponsor: Federal Highway
Administration, Washington, D.C. Report No. UMTRI-83-25. UMTRI-70908
Ervin, R. D. 1983. The influence of size and weight variables on the roll stability of heavy duty trucks.
Michigan University, Ann Arbor, Transportation Research Institute. 28 p. Report No. SAE 831 163.
UMTRI-73623
Winkler, C. B.; Fancher, P. S.; MacAdam, C. C. 1983. Parametric analysis of heavy duty truck dynamic
stability. Volume I - technical report. Final report. Michigan University, Ann Arbor, Transportation
Research Institute. 170 p. Sponsor: National Highway Traffic Safety Administration, Washington, D.C.
Report No. UMTRI-83- 13-11DOTNS 806 4 11. UMTRI-48533
Winkler, C. B.; Nisonger, R. L. 1983. Steering and suspension system - descriptive parameters used in
analyzing the braking and handling of heavy trucks. Volume 4. Second edition. Final report. Michigan
University, Ann Arbor, Transportation Research Institute. 217 p. Sponsor: National Highway Traffic Safety
Administration, Washington, D.C. Report No. UMTRI-83-16. UMTRI-48602
1983. Multiple vehicle collisions and fire, Caldecott Tunnel, near Oakland, California, April 7, 1982;
Highway accident report: multiple vehicle collisions and fire, Caldecott Tunnel, near Oakland, California,
April 7, 1982. National Transportation Safety Board, Washington, D.C. 46 p. Report No. NTSB-HAR-83-
01. UMTRI-48232
Ervin, R. D. 1984. Issues related to the usage of a tilt table for measuring the roll stability characteristics of
heavy-duty truck combinations. Final report. Michigan University, Ann Arbor, Transportation Research
Institute. 20 p. Sponsor: Motor Vehicle Manufacturers Association, Detroit, Mich. Report No. UMTRI-84-
3 1. UMTRI-71980
Ervin, R. D.; Fancher, P. S.; Gillespie, T. D. 1984. An overview of the dynamic performance properties of
long truck combinations. Michigan University, Ann Arbor, Transportation Research Institute. 69 p.
Sponsor: Federal Highway Administration, Washington, D.C. Report No. UMTRI-84-26. UMTRI-71!381
Mai, L.; Sweatman, P. 1984. Articulated vehicle stability - phase 11: tilt tests and computer models.
Australian Road Research Board, Vermont South. 166 p. Report No. AIR 323-2. UMTRI-70995
Soret, P. 1984. Handling and road holding of tractor-semi-trailer units: improvement of rollover threshold.
Published by Motor Industry Research Association, Nuneaton, England. 20 p. Report No. MIRA
Translation No. 22/84. UMTRI-72240
Sweatman, P. 1984. Rollover. Australian Road Research Board, Victoria. 5 p. Truck and Bus
Transportation, Vol. 48, No. 10, Oct 1984, pp. 24-26,28. UMTRI-72184
Sweatman, P.; Mai, Id. 1984. Articulated vehicle stability - the role of vehicle design. 12 p. FISITA
Congress. XXth. [Volume 2. Proceedings.] Warrendale, Pa., SAE, May 1984. Pp. 2.241-2.252. Report No.
SAE 845055. UMTlXI-70439 A05
Tidbury, G. H. 1984. International Journal of Vehicle Design, Vol. 5, Nos. 112, Jan 1984, pp. 67-85.
Sponsor: Great Britain Department of Transport, Vehicle Standards and Engineering Division, London;
HCB-Angus, Southampton, England. UMTRI-55506
Ervin, R. D.; Barnes, M.; MacAdam, C. C.; Scott, R. 1985. Impact of specific geometric features on truck
operations and safety at interchanges. Volume I - technical report. Final report. Michigan University, .4nn
Arbor, Transportation Research Institute. 132 p. Sponsor: Federal Highway Administration, Washington,
D.C. Report No. FHWAIKD-8610571UMTRI-85-3311. UMTRI-73167
Ervin, R. D.; Barnes, M.; Wolfe, A. 1985. Liquid cargo shifting and the stability of cargo tank trucks.
Volume I - executive summary. Final report. Michigan University, Ann Arbor, Transportation Resear~zh
Institute. 9 p. Sponsor: Federal Highway Administration, Bureau of Motor Carrier Safety, Washington,
D.C. Report No. UMTRI-85-35/1. UMTRI-73 164
Bedard, J. T. 1986. Vehicle rollover threshold evaluation. Quebec Centre de Recherche Industrielle,
Canada. 84 p. Vehicle Weights and Dimensions Study, Volume 12, 1986. Sponsor: Canroad Transportation
Research Corporation, Ottawa, Ontario, Canada. UMTRI-77344
Billing, J. R. 1986. Demonstration test program: five, six and seven axle tractor semitrailers. Ontario
Ministry of Transportation and Communications, Downsview, Canada. 100 p. Vehicle Weights and
Dimensions Study, Volume 4, 1986. Sponsor: Canroad Transportation Research Corporation, Ottawa,
Ontario, Canada. UMTRX-77336
Billing, J. R. 1986. Summary of tests of baseline vehicle performance. Ontario Ministry of Transportation
and Communications, Downsview, Canada. 87 p. Vehicle Weights and Dimensions Study, Volume 3,
1986. Sponsor: Canroad Transportation Research Corporation, Ottawa, Ontario, Canada. Report No. (2V-
86-12. UMTRI-77335
Billing, J. R.; Mercer, W. 1986. Demonstration of baseline vehicle performance: 45 ft semitrailer. Ontario
Ministry of Transportation and Communications, Downsview, Canada. 42 p. Vehicle Weights and
Dimensions Study, 1986. Sponsor: Canroad Transportation Research Corporation, Ottawa, Ontario,
Canada. Report No. CV-86-02. UMTRI-77350
Billing, J. R.; Mercer, W.; Stephenson, W. R. 1986. Procedures for test of baseline and additional vehicles.
Ontario Ministry of Transportation and Communication, Downsview, Canada. 82 p. Vehicle Weights and
Dimensions Study, 1986. Sponsor: Canroad Transportation Research Corporation, Ottawa, Ontario,
Canada. Report No. CV-86-01. UMTRI-77349
Bordfeld, G.; Junker, H. 1986. Entwicklung einer schwingungstechnisch optimierten Doppelachse. Teil2:
Prototypentwicklung und Erprobung; Development of a double axle with optimised vibrations. Part 2:
prototype development and testing. Hannover Technische Universitaet, Institut fuer Kraftfahrwesen,
Germany FR. 6 p. Automobil-Industrie, 31. Jahrgang, No. 1, Feb 1986, pp. 63-68. UMTRI-56193
Carsten, 0 . ; Campbell, K. L. 1986. Overview of the University of Michigan Transportation Research
Institute large-truck survey program. Michigan University, Ann Arbor, Transportation Research Institute.
16 p. UMTRI-73293
Delisle, G . ;Pearson, J. R. 1986. Investigating articulated vehicle roll stability using a tilt table device.
Quebec Centre de Recherche Industrielle, Canada1 Roads and Transportation Association of Canada,
Ottawa, Ontario. 243 p. Vel icle Weights and Dimensions Study, Volume 7, 1986. Sponsor: Canroad
Transportation Research Corporation, Ottawa, Ontario, Canada. UMTRI-77339
Ervin, R. D. 1986. The dependence of truck roll stability on size and weight variables. Michigan
University, Ann Arbor, Transportation Research Institute. 17 p. International Journal of Vehicle Design,
Vol. 7, No. 516, Sept 1986, pp. 192-208. UMTRI-56488
Ervin, R. D. 1986. The importance of tire cornering properties to the dynamic behavior of heavy-duty
trucks. Michigan University, Ann Arbor, Transportation Research Institute. 36 p. Schuring, D. J., ed.
Cornering and Handling Characteristics of Tires. Rubber Division Symposia. Volume 11. Akron, Akron
University, 1986. Pp. 137-172. UMTRI-89496 A02
Ervin, R. D.; Guy, Y. 1986. The influence of weights and dimensions on the stability and control of heavy-
duty trucks in Canada. Volume I - technical report. Final report. Michigan University, Ann Arbor,
Transportation Research Institute. 277 p. Vehicle Weights and Dimensions Study, Volume 1, 1986.
Sponsor: Canroad Transportation Research Corporation, Ottawa, Ontario, Canada. Report No. UMTRI-86-
35A. UMTRI-73754
Guang, X. 1986. Study of stability of articulated vehicles on side slopes: a discussion about the two-stage
theory of roll overturn. China, People's Republic, Corps of Engineers. 12 p. International Society for
Terrain-Vehicle Systems. First Asian-Pacific Conference. Proceedings. China Academic Publishers, Aug
1986. Pp. 264-274. UMTRI-76037
Carsten, 0. 1987. Safety implications of truck configuration. Michigan University, Ann Arbor,
Transportation Research Institute. 10 p. Transportation Research Record, No. 1111, 1987, pp. 17-26.
UMTRI-56791 A03
Clarke, R. M.; Leasure, W. A., Jr.; Radlinski, R. W.; Smith, M. 1987. Heavy truck safety study. Prepared in
response to Section 216 of the Motor Carrier Safety Act of 1984. Final report. National Highway Traffic
Safety Administration, Research and Development, Washington, D.C. 216 p. Report No. DOTIHS 807 109.
UMTRI-76132
Ervin, R. D.; MacAdam, C. C.; Barnes, M. A. 1987. Truck control problems posed by the design of
highway ramps. Michigan University, Ann Arbor, Transportation Research Institute1 Washtenaw County
Road Commission, Ann Arbor, Mich. 14 p. Vehicle Highway Infrastructure: Safety Compatibility.
Warrendale, Society of Automotive Engineers, Feb 1987. Pp. 29-42 Also published in SAE Transactions
1987. Volume 96. Warrendale, SAE, 1988. Report No. SAE 870071. UMTRI-74692 A02
Ervin, R. D.; Mathew, A. 1987. Stability of tank truck combinations on curved road segments in the
Yukon. Final report. Michigan University, Ann Arbor, Transportation Research Institute. 104 p. Sponsor:
Yukon Community and Transportation Services, Whitehorse, Canada. Report No. UMTRI-87-9. UMTRI-
74372
Fancher, P. S.; Mathew, A. 1987. A vehicle dynamics handbook for single-unit and articulated heavy
trucks. Final report. Michigan University, Ann Arbor, Transportation Research Institute. 367 p. Sponsor:
National Highway Traffic Safety Administration, Washington, D.C. Report No. UMTRI-87-27. UMTRI-
76322
Komatsu, K. 1987. Non-linear sloshing analysis of liquid in tanks with arbitrary geometries. Japan National
Aerospace Laboratories, Tokyo. 15 p. International Journal of Nonlinear Mechanics, Vol. 22, No. 3, 1987,
pp. 193-207. UMTRI-76467
Stedtnitz, W.; Appel, H. 1987. Analysis of heavy-freight vehicle and tank-truck accidents. Berlin
Technische Universitaet, Institut fuer Kraftfahrzeuge, Germany FR. 20 p. Symposium on the Role of
Heavy Freight 'Vehicles in Traffic Accidents. Volume 2. Ottawa, Roads and Transportation Association of
Canada, 1987. Pp. 3-3 - 3-22. UMTRI-76118 A01
Tso, Y. 1987. Interim report on articulated vehicle stability - phase I11 full-scale dynamic tests. Australian
Road Research Board, Vermont South. 31 p. Report No. AIR-323-4. UMTRI-76415
Winkler, C. B. 1987. Experimental determination of the rollover threshold of four tractor-semitrailer
combination vehicles. Final report. Michigan University, Ann Arbor, Transportation Research Institute. 65
p. Sponsor: Sandia Laboratories, Albuquerque, N.M. Report No. UMTRI-87-31. UMTRI-76285
Baker, C. J. 1988. High sided articulated road vehicles in strong cross winds. Nottingham University,
Department of Civil Engineering, England. 19 p. Journal of Wind Engineering and Industrial
Aerodynamics, Vol. 31, 1988, pp. 67-85. UMTRI-78089
Ervin, R. D.; Mathew, A. 1988. Reducing the risk of spillage in the transportation of chemical wastes by
truck. Final report. Michigan University, Ann Arbor, Transportation Research Institute. 25 p. Sponsor:
Rohm and Haas, Bristol, Pa. Report No. UMTRI-88-28. UMTRI-77525
Furleigh, D. D.; Vanderploeg, M. J.; Oh, C. Y. 1988. Multiple steered axles for reducing the rollover risks
of heavy articulated trucks;. Iowa State University, Ames. 7 p. Report No. SAE 881866. UMTRI-77613
Laird, L. A. 1988. Measurement of heavy vehicle suspension roll-stability properties, and a method to
evaluate overall stability performance. PACCAR Technical Center, Mount Vernon, Wash. 46 p. Report No.
SAE 881869. UMTKI-77611
Miki, S. 1988. Method for evaluating stability and handling of a truck considering body torsional rigidity.
Hino Motors Company, Ltd., Tokyo, Japan. 10 p. Report No. SAE 881870. UMTRI-77612
Stedtnitz, W.; Appel, H. 1988. Experimental and analytical simulation of liquid sloshing in tank trucks.
Berlin Technische Universitaet, Institut fuer Kraftfahrzeuge, Germany FR. 11 p. FISITA. Congress.
XXIInd. Automotive Systems Technology: the Future. Volume 11. Technical Papers. [Proceedings.]
Warrendale, SAE, 1988. Pp. 2.137-2.147. Report No. SAE 8851 10. UMTRI-78279 A07
Sweatman, P. F.; Glynn, L.; George, R. M., eds. 1988. Truck designers sprung? Symposium on Heavy
Vehicle Suspension Characteristics. Proceedings. Australian Road Research Board, Vermont South,
Victoria. 479 p. Report No. ARRMS 871172. UMTRI-79852
Sweatman, P.; Tso, Y. 1988. The influence of suspension characteristics on the rollover stability of
articulated vehicles. Australian Road Research Board, Vermont South, Victoria. 19 p. Sweatman, P. F.,
Glynn, L., and George, R. M., eds. Truck Designers Sprung? Symposium on Heavy Vehicle Suspension
Characteristics. Proceedings. Vermont South, A R M , 1988. Pp. 179-197. UMTRI-79852 A05
Winkler, C. B. 1988. Improving the dynamic performance of multi-trailer vehicles: a study of innovative
dollies. Michigan University, Ann Arbor, Transportation Research Institute. 19 p. UMTRI Research
Review, Vol. 18, No. 5, March-April 1988, pp. 1-19. UMTRI-56985
Woodrooffe, J. H. F.; LeBlanc, P. A. 1988. Suspensions and vehicle stability as determined by the
Canadian Heavy Vehicle Weights and Dimensions Study. National Research Council Canada, Ottawa,
Ontario. 14 p. Sweatman, P. F., Glynn, L., and George, R. M., eds. Truck Designers Sprung? Symposium
on Heavy Vehicle Suspension Characteristics. Proceedings. Vermont South, A R M , 1988. Pp. 199-212.
UMTRI-79852 A06
Ajluni, K. K. 1989. Rollover potential of vehicles on embankments, sideslopes, and other roadside features.
Federal Highway Administration, Office of Safety and Traffic Operations Research and Development,
McLean, Va. 7 p. Public Roads, Vol. 52, No. 4, March 1989, pp. 107-1 13. UMTRI-57206
Dickson-Simpson, J. 1989. Stability - a tanker-safety fundamental. 2 p. Transport Engineer, June 1989, pp.
34-35. UMTRI-57294
Fancher, P. S.; Campbell, K. L.; Blower, D. F. 1989. Vehicle design implications of the Turner proposal.
Michigan University, Ann Arbor, Transportation Research Institute. 20 p. SAE Transactions 1989. Volume
98. Warrendale, SAE, 1990. Sponsor: Great Lakes Center for Truck Transportation Research, Ann Arbor,
Mich.; Transportation Department, University Transportation Centers Program, Washington, D.C. Report
NO. GLCTTR 06-89/21 SAE 892461. UMTRI-79116
Fancher, P. S.; Mathew, A.; Campbell, K. L.; Blower, D. F.; Winkler, C. B. 1989. Turner truck handling
and stability properties affecting safety. Volume I - technical report. Final report. Michigan University,
Ann Arbor, Transportation Research Institute. 206 p. Sponsor: National Cooperative Highway Research
Program, Washington, D.C. Report No. UMTRI-89-11-1. UMTRI-78447
Firestine, M.; McGee, H.; Toeg, P. 1989. Improving truck safety at interchanges. Summary report. Walcoff
and Associates, Alexandria, Va. 36 p. Sponsor: Federal Highway Administration, Office of
Implementation, McLean, Va. Report No. FHWAAP-891024. UMTRI-79739
Klingenberg, B.; Rossow, G.; Jacobsen, R. 1989. FACT - the FreightlinerIHeil Advanced Concept Truck.
Freightliner Corporation, Portland, Ore. 27 p. Report No. SAE 892462. UMTRI-79151
Rakheja, S.; Sankar, S.; Ranganathan, R. 1989. Influence of tank design factors on the rollover threshold of
partiallly filled tank vehicles. Concordia University, Montreal, Quebec, Canada. 14 p. Sponsor: Canada,
Transport Canada, Transportation Development Centre, Montreal, Quebec. Report No. SAE 892480.
UMTRI-78980
Ranganathan, R.; Rakheja, S.; Sankar, S. 1989. Kineto-static roll plane analysis of articulated tank vehicles
with arbitrary tank geometry. Concordia University, Montreal, Quebec, Canada. 23 p. International Journal
of Vehicle Design, Vol. 10, No. 1, 1989, pp. 89-1 11. Sponsor: Canada, Transport Canada, Transportation
Development Centre, Montreal, Quebec. UMTRI-57246
Ranganathan, R.; Rakheja, S.; Sankar, S. 1989. Steady turning stability of partially filled tank vehicles with
arbitrary tank geometry. Concordia University, Montreal, Quebec, Canada. 9 p. Journal of Dynamic
Systems, Measurement and Control, Vol. 111, No. 3, Sept 1989, pp. 481-489. Sponsor: Canada, Transport
Canada, Transportation Development Centre, Montreal, Quebec. UMTRI-78745
Sankar, S.; Rakheja, S.; Ranganathan, R. 1989. Directional response of partially filled tank vehicles,
Concordia University, Montreal, Quebec, Canada. 11 p. Report No. SAE 892481. UMTRI-78979
Sankar, S.; Rakheja, S.; Ranganathan, R.; Sabounghi, L. 1989. On the stability of heavy articulated liquid
tank vehicles. Concordia University, Montreal, Quebec, Canada. 16 p. International Symposium on Heavy
Vehicle Weights and Dimensions. Second. Proceedings. Volume 2. [1989]. Pp. 53-69. Sponsor: Canada,
Transport Canada, Ottawa, Ontario. UMTRI-84446 A02
Tso, Y.; Sweatman, P. 1989. The dynamic stability testing of articulated vehicles. Australian Road
Research Board, Vermont South. 5 p. SAE Australasia, Vol. 49, No. I , JantFeb 1989, pp. 29-33. UMTRI-
57207
Becher, H. 0. 1990. Untersuchungen zum Wankverhalten von Nutzfahrzeugen mit adaptiven Fahrwerken;
Investigations into the roll behaviour of commercial vehicles with adaptive chassis. Hannover Universitaet,
Germany FR. 10 p. Automobil Industrie, 35. Jahrgang, No. 2, April 1990, pp. 135-144. UMTRI-57659
El-Gindy, M.; Woodrooffe, J. H. F. 1990. Study of rollover threshold and directional stability of log
hauling trucks. National Research Council Canada, Division of Mechanical Engineering, Ottawa, Ontario.
240 p. Report No. TR-VDL-0021 NRCC No. 3 1274. UMTRI-80157
Goehring, E. 1990. Aktiv und passive Sicherheit bei Nutzfahrzeugen; Active and passive safety of
commercial vehicles. Mercedes-Benz AG, Stuttgart, Germany FR. 14 p. Automobil Industrie, 35. Jahrgang,
No. 4, July 1990, pp. 337-350. UMTRI-57729
Harwood, D. W; Mason, J. M.; Glauz, W. D.; Kulakowski, B. T.; Fitzpatrick, K. 1990. Truck
characteristics for use in highway design and operation. Volume I: research report. Final report. Midwest
Research Institute, Kansas City, Mo. 261 p. Sponsor: Federal Highway Administration, Office of Safety
and Traffic Operations Research and Development, McLean, Va. Report No. FHWA-RD-89-226. UMTRI-
80716
Hasegawa, S.; Kusahara, Y.; Watanabe, Y. 1990. Influence of vehicle body torsional stiffness on vehicle
roll characteristics of medium-duty trucks. Nissan Diesel Motor Company, Ltd., Ageo, Japan. 7 p. Total
Rakheja, S.; Ranganathan, R.; Sankar, S. 1992. Field testing and validation of directional dynamics model
of a tank truck. Concordia University, Montreal, Quebec, Canada1 Missouri University, Columbia,
Department of Mechanical and Aerospace Engineering. 25 p. International Journal of Vehicle Design, Vol.
13, No. 3, 1992, pp. 251-275. Sponsor: Canada, Transport Canada, Transportation Development Centre,
Montreal, Quebec; Canada, Formation de Chercheurs et 1'Aide a la Recherche of Quebec Government.
UMTRI-58432
Sankar, S.; Ranganathan, R.; Rakheja, S. 1992. Impact of dynamic fluid slosh loads on the directional
response of tank vehicles. Concordia University, Montreal, Quebec, Canada1 Missouri University,
Columbia. 20 p. Vehicle System Dynamics, Vol. 21, No. 6, 1992, pp. 385-404. Sponsor: Natural Sciences
and Engineering Research Council of Canada, Ottawa, Ontario; Canada, Formation de Chercheurs et 1'Aide
a la Recherche of Quebec Government; Transportation Development Centre of Canada, Quebec. UMTRI-
58633
Winkler, C. B. 1992. A primer on the mechanics of commercial vehicle rollover - annotated slides for an
oral presentation. Michigan University, Ann Arbor, Transportation Research Institute. 17 p. Report No.
SAE 922425. UMTRI-83393
Winkler, C. B.; Fancher, P. S. 1992. A rationale for regulating roll stability of combination vehicles.
Michigan University, Ann Arbor, Transportation Research Institute. 8 p. Cebon, D., and Mitchell, K., eds.
Heavy Vehicles and Roads - Technology, Safety and Policy. Proceedings of the Third International
Symposium on Heavy Vehicle Weights and Dimensions. London, Thomas Telford, 1992. Pp. 323-330.
UMTRI-84 153 A30
Winkler, C. B.; Fancher, P. S.; Bareket, Z.; Bogard, S. E.; Johnson, G.; Karamihas, S. M.; Mink, C. E.
1992. Heavy vehicle size and weight: test procedures for minimum safety performance standards. Final
technical report. Michigan University, Ann Arbor, Transportation Research Institute. 118 p. Sponsor:
National Highway Traffic Safety Administration, Washington, D.C. Report No. UMTRI-92-131 DOTMS
807 855. UMTRI-82343
Winkler, C. B.; Karamihas, S. M.; Bogard, S. E. 1992. Roll-stability performance of heavy-vehicle
suspensions. Michigan University, Ann Arbor, Transportation Research Institute. 13 p. Suspensions and
Their Relationship to Vehicle Roll Stability. Warrendale, SAE, 1992. Pp. 1-13 Also published in SAE
Transactions 1992. Volume 101. Warrendale, SAE, 1993. Report No. SAE 922426. UMTRI-83581 A01
Woodrooffe, J. H.; El-Gindy, M. 1992. Application of handling and roll stability performance measures for
determining a suitable tractor wheelbase. National Research Council Canada, Ottawa, Ontario. 6 p. Cebon,
D., and Mitchell, K., eds. Heavy Vehicles and Roads - Technology, Safety and Policy. Proceedings of the
Third International Symposium on Heavy Vehicle Weights and Dimensions. London, Thomas Telford,
1992. Pp. 30-35. UMTRI-84153 A05
1992. Suspensions and their relationship to vehicle roll stability. Society of Automotive Engineers,
Warrendale, Pa. 100 p. Report No. SAE SP-940. UMTRI-83581
Ahn, K. W.; Kim, C. S." 1993. Dynamic stability of a medium duty truck depending upon the location of
a heavy object. Asia Motors. 7 p. Report No. SAE 931952. UMTRI-85026
Carr, G. W.; Rose, M. J. 1993. Cross-wind stability of vehicles on bridges. Motor Industry Research
Association, Lindley, England. 9 p. Environmental and Physical Effects on Vehicle Safety. Seminar Thirty
Nine Papers. England, Institution of Mechanical Engineers, 1993. Pp. 17-25. Report No. C462/39/088.
UMTRI-865 14 A02
Clover, C.; McLaughlin, J.; Bernard, J.; Vardeman, S. 1993. Sensitivity analysis of a tilt table test
procedure. Iowa State University, Ames, Center for Emerging Manufacturing Technology. 68 p. Sponsor:
American Automobile Manufacturers Association, Detroit, Mich. UMTRI-85403
Das, N. S.; Suresh, B. A,; Wambold, J. C. 1993. Estimation of dynamic rollover threshold of commercial
vehicles using low speed experimental data. Engineering Global Solutions, Inc.1 Pennsylvania
Transportation Institute, University Park. 11 p. Heavy Vehicle Dynamics and Stability. Warrendale, SAE,
1993. Pp. 85-95. Report No. SAE 932949. UMTRI-84991 A08
Fancher, P. 1993. Heavy vehicle rollover. Michigan University, Ann Arbor, Transportation Research
Institute. 16 p. Heavy Vehicle Rollovers: A Safety Issue. TOPTEC. [Proceedings]. Day Two. Warrendale,
SAE, 1993. Pp. 43-58. UMTRI-84660 A01
Karnopp, D.; Hibbard, R. 1993. Steering control for roll model damping augmentation in tall road vehicles.
California University, Davis. 9 p. Heavy Vehicle Dynamics and Stability. Warrendale, SAE, 1993. Pp. 65-
73. Report No. SAE 932947. UMTRI-84991 A07
McGee, H.; Joshua, S.; Hughes, W.; Strickland, R.; Bareket, Z.; Fancher, P. 1993. Feasibility of an
automatic truck warning system. Final report. Bellomo-McGee, Inc., Vienna, Va.1 Michigan Universipy,
Ann Arbor, Transportation Research Institute. 82 p. Sponsor: Federal Highway Administration, Office of
Safety and Traffic Operations Research and Development, McLean, Va. Report No. FHWA-RD-93-039.
UMTRI-84184
Nordstroem, 0. 1993. Stability, steerability and braking performance of heavy duty vehicles: a review of
experimental and theoretical research and regulation proposals by VTI in Sweden. Statens Vaeg- och
Trafikinstitut, Linkoeping, Sweden. 29 p. Heavy Vehicle Systems, Vol. 1, No. 1, 1993, pp. 34-62.
UMTRI-592 15
Ranganathan, R. 1993. Rollover threshold of partially filled tank vehicles with arbitrary tank geometry.
Missouri University, Columbia, Department of Mechanical and Aerospace Engineering. 4 p. Institution of
Mechanical Engineers. Proceedings. Part D: Journal of Automobile Engineering, Vol. 207, No. D3, 15193,
pp. 241-244. UMT!LI-85810
Ranganathan, R.; Ying, Y.; Miles, J. B. 1993. Analysis of fluid slosh in partially filled tanks and their
impact on the directional response of tank vehicles. Missouri University, Columbia. 7 p. Heavy Vehicle
Dynamics and Stability. Warrendale, SAE, 1993. Pp. 39-45. Report No. SAE 932942. UMTRI-84991 A03
Ranganthan, R.; Rakheja, S.; Sankar, S. 1993. The effects of vehicle configurations and tank design fs~ctors
on directional dynamics of tank vehicles. Missouri University, Columbia/ Concordia University, Montreal,
Quebec, Canada. 20 p. Canadian Society for Mechanical Engineering Transactions, Vol. 17, No. 4B, 1993,
pp. 923-942. UMTRI-87092 A08
Sweatman, P. F. 1993. Overview of dynamic performance of the Australian heavy vehicle fleet. Road User
Research Pty., Ltd., Williamstown, Victoria, Australia. 80 p. Sponsor: Australian National Road Transport
Commission, Melbourne, Victoria. Report No. Technical Working Paper No. 7. UMTRI-85114
Timoney, E. P.; Timoney, S. S. 1993.Off-road tests of an independently sprung airport fire truck and a
handling comparison with beam axle vehicles. Timoney Research Ltd., County Meath, Ireland1 University
College of Dublin, Ireland. 24 p. Vehicle Ride and Handling International Conference. Proceedings,
London, Institution of Mechanical Engineers, 1993. Pp. 145-168. Report No. C466/013/93. UMTRI-85137
A12
1993. Heavy vehicle rollovers - a safety issue. TOPTEC. [Proceedings]. Day one. Society of Automoltive
Engineers, Warrendale, Pa. 250 p. UMTRI-84659
1993. Heavy vehicle rollovers - a safety issue. TOPTEC. [Proceedings]. Day Two. Society of Automotive
Engineers, Warrendale, Pa.196 p. UMTRI-84660
1993. Mechanics of heavy-duty trucks and truck combinations. Michigan University, Ann Arbor,
Transportation Research Institute. Published by Michigan University, Ann Arbor, Engineering
Conferences. 1125 p. UMTRI-85216
Duffin, A. R. 1994. The stability of road tankers for the carriage of hazardous goods. Frazer-Nash
Consultancy, Ltd., Surrey, England. 8 p, ISATA International Symposium on Automotive Technology and
Automation, 27th. Proceedings for the Dedicated Conference on Road and Vehicle Safety. Croydon,
Automotive Automation Ltd., 1994. Pp. 285-292. Report No. 94SF004. UMTRI-87370 A16
El-Gindy, M.; Tong, L.; Tabarrok, B. 1994. Frequency response analysis of Canadian logging trucks.
National Research Council Canada, Ottawa, Ontario1 Victoria University, British Columbia, Canada. 25 p.
Vehicle System Dynamics, Vol. 23, No. 5, July 1994, pp. 325-349. UMTRI-59327
Kusahara, Y.; Li, X.; Hata, B.; Watanabe, Y. 1994. Feasibility study of active roll stabilizer for reducing
roll angle of an experimental medium-duty truck. Nissan Diesel Motor Company, Ltd., Saitama, Japan. 6 p.
International Symposium on Advanced Vehicle Control 1994. Proceedings. Tokyo, Society of Autoniotive
Engineers of Japan, 1994. Pp. 343-348. Report No. SAE 9438501. UMTRI-86557 A35
Leonard, J. D., 11. 1994. Alternative approach to estimation of critical cornering speed. Georgia Institute of
Technology, Atlanta. 19 p. Journal of Transportation Engineering, Vol. 120, No. 3, MayIJune 1994, pp.
478-496. UMTRI-593 18
Lin, R. C.; Cebon, D.; Cole, D. J. 1994. An investigation of active roll control of heavy road vehicles.
Cambridge University, Engineering Department, England. 14 p. Shen, Z., ed. The Dynamics of Vehicles on
Roads and on Tracks. Proceedings of 13th IAVSD Symposium. Lisse, Swets and Zeitlinger, 1994. Pp. 308-
321. UMTRI-85817 A12
McGee, H. W.; Strickland, R. R. 1994. An automatic warning system to prevent truck rollover on curved
ramps. Bellomo-McGee, Inc., Vienna, Va. 6 p. Public Roads, Vol. 57, No. 4, Spring 1994, pp. 17-22.
UMTRI-59 109
Ranganathan, R.; Ying, Y.; Miles, J. B. 1994. Development of a mechanical analogy model to predict the
dynamic behavior of liquids in partially filled tank vehicles. Missouri University, Columbia. 8 p. Heavy
Vehicle Dynamics and Simulation in Braking, Steering, and Suspension Systems. Warrendale, SAE, 1994.
Pp. 79-86. Report No. SAE 942307. UMTRI-86616 A04
Sankar, S.; Surial, S. 1994. A sensitivity analysis approach for fast estimation of rollover stability of heavy
articulated vehicles during steady state turning. Concordia University, Montreal, Quebec, Canada. 22 p.
Heavy Vehicle Systems, Vol. 1, No. 3, 1994, pp. 282-303. Sponsor: Natural Sciences and Engineering
Research Council of Canada, Ottawa, Ontario; Canada, Formation de Chercheurs et 1'Aide a la Recherche
of Quebec Government. UMTRI-59385
Wang, Z.; Rakheja, S.; Cunzhen, S. 1995. Influence of partition location on the braking performance of a
partially-filled tank truck. Concordia University, Montreal, Quebec, Canada1 Jilin University of
Technology, Changchun, China, People's Republic. 10 p. New Developments in Axle, Steering,
Suspension, and Chassis Technology. Warrendale, Society of Automotive Engineers, 1995. Pp. 75-84.
UMTRI-88292 A08
Lin, R. C.; Cebon, D.; Cole, D. J. 1996. Active roll control of articulated vehicles. Cambridge University,
Department of Engineering, England. 27 p. Vehicle System Dynamics, Vol. 26, No. 1, July 1996, pp. 17-
43. UMTRI-60130
Lin, R. C.; Cebon, D.; Cole, D. J. 1996. Optimal roll control of a single-unit lony. Cambridge University,
England. 11 p. Institution of Mechanical Engineers. Proceedings. Part D: Journal of Automobile
Engineering, Vol. 210, 1996, pp. 45-55. UMTRI-89198
Palkovics, L.; Michelberger, P.; Bokor, J.; Gaspar, P. 1996. Adaptive identification for heavy-truck
stability control. Budapest Technical University, Hungary. 18 p. Segel, L., ed. The Dynamics of Vehicles
on Roads and on Tracks. Proceedings of the 14th IAVSD Symposium. Lisse, Swets and Zeitlinger, 1996.
Pp. 502-518. Sponsor: Hungarian Scientific Research and Development Fund. UMTRI-88885 A18
Popov, G.; Sankar, S.; Sankar, T. S. 1996. Shape optimization of elliptical road containers due to liquid
load in steady-state turning. Concordia University, Montreal, Quebec, Canada. 19 p. Vehicle System
Dynamics, Vol. 25, No. 3, Mar 1996, pp. 203-221. Sponsor: Canada, Formation de Chercheurs et 1'Aide a
la Recherche of Quebec Government; Natural Sciences and Engineering Research Council of Canada,
Ottawa, Ontario. UMTRI-60015
Ranganathan, R.; Yang, Y. S. 1996. Impact of liquid load shift on the braking characteristics of partially
filled tank vehicles. Missouri University, Columbia. 18 p. Vehicle System Dynamics, Vol. 26, No. 3, Sept
1996, pp. 223-240. UMTRI-60211
Rompe, K.; Heuser, G. 1996. THESEUS: Tankfahrzeuge mit hoechsterreichbarer Sicherheit durch
experimentelle Unfallsimulation; THESEUS: achieving maximum possible road transport tanker safety by
means of experimental accident simulation. 8 p. ATZ, 98. Jahrgang, No. 3, May 1996, pp. 154-161.
UMTRI-60038
Dahlberg, E.; Vaagstedt, N.-G. 1997. The advantages of a simple approach modelling heavy vehicle
handling. Royal Institute of Technology, Stockholm (Sweden)/ Scania (Sweden) Published by Society of
Automotive Engineers, Warrendale, Pa. 10 p. Report No. SAE 973264. UMTRI-91463
Liu, P.; Rakheja, S.; Ahmed, A. K. W. 1997. Detection of dynamic roll instability of heavy vehicles for
open-loop rollover control. Concordia University, Montreal, Quebec (Canada) Published by Society of
Automotive Engineers, Warrendale, Pa. 10 p. Sponsor: Canada, Formation de Chercheurs et 1'Aide a la
Recherche of Quebec; Natural Sciences and Engineering Research Council of Canada, Ottawa, Ontario.
Report No. SAE 973263. UMTRI-91464
McFarlane, S.; Sweatman, P.; Dovile, P.; Woodrooffe, J. 1997. The correlation of heavy vehicle
performance measures. Roaduser Research Pty., Ltd. Published by Society of Automotive Engineers,
Warrendale, Pa. 12 p. Heavy Vehicle and Highway Dynamics (SAE-SP-1308). Warrendale, SAE, 199:l.
Pp. 21-30, Report No. SAE 973190. UMTRI-90867
Ervin, R.; Winkler, C.; Fancher, P.; Hagan, M.; Krishnaswami, V.; Zhang, H.; Bogard, S. 1998. Two alztive
systems for enhancing dynamic stability in heavy truck operations. Michigan University, Ann Arbor,
Transportation Research Institute. 216 p. Sponsor: National Highway Traffic Safety Administration,
Washington, D.C. Report No. UMTRI-98-39. UMTRI-91200
Strickland, R.; McGee, H. 1998. Evaluation of prototype automatic truck rollover warning systems.
Bellomo-McGee, Inc., Vienna, Va. 87 p. Sponsor: Federal Highway Administration, Office of Safety and
Traffic Operations, Washington, D.C. Report No. FHWA-RD-97-124. UMTRI-90723