OQASlstaaltlawsc.docx

O
Q
A
S
l
s
t
a
a
l
t
l
a
w
s
c
i
c
s
c
t
m
u
e

s
p
c
R
W
d
c
T
v
L
A
(
s
o
K
O
54
A
RIGINAL ARTICLE
uantification of Lumbar Stability by Using 2 Different
bdominal Activation Strategies
ylvain G. Grenier, PhD, Stuart M. McGill, PhD
i
i
w

c
v
h
u
m
p
e
s
b
b
g
t
i
c
d
a
d
r
t
m
t
a
m
e
w
c
k
a
s
i
o
[
p
m

o
c
n
v
F
d
e
a
s
e
v
p
s
a
e
i
l
ABSTRACT. Grenier SG, McGill SM. Quantification of
umbar stability by using 2 different abdominal activation
trategies. Arch Phys Med Rehabil 2007;88:54-62.
Objective: To determine whether the abdominal hollowing
echnique is more effective for lumbar spine stabilization than
full abdominal muscle cocontraction.
Design: Within-subject, repeated-measures analysis of vari-
nce was used to examine the effect of combining each of 4
oading conditions with either the hollow or brace condition on
he dependent variables of stability and compression. A simu-
ation was also conducted to assess the outcome of a person
ctivating just the transversus abdominis during the hollow.
Setting: Laboratory.
Participants: Eight healthy men (age range, 20�33y).
Interventions: Electromyography and spine kinematics

ere recorded during an abdominal brace and a hollow while
upporting either a bilateral or asymmetric weight in the hands.
Main Outcome Measures: Spine stability index and lumbar
ompression were calculated.
Results: In the simulation “ideal case,” the brace technique
mproved stability by 32%, with a 15% increase in lumbar
ompression. The transversus abdominis contributed .14% of
tability to the brace pattern with a less than 0.1% decrease in
ompression.
Conclusions: Whatever the benefit underlying low-load
ransversus abdominis activation training, it is unlikely to be
echanical. There seems to be no mechanical rationale for
sing an abdominal hollow, or the transversus abdominis, to
nhance stability. Bracing creates patterns that better enhance
tability.
Key Words: Abdominal muscles; Back injuries; Low back
ain; Motor skills; Rehabilitation; Spine.
© 2007 by the American Congress of Rehabilitation Medi-
ine and the American Academy of Physical Medicine and
ehabilitation
ITH A GOAL TO DECREASE low back pain (LBP),
stabilizing exercises, along with modifications of other
aily activities, have been shown to be effective in randomized
linical trials.1,2 However, the source of this effect is uncertain.
his study assessed the influence of different abdominal acti-
ation strategies on mechanical stability in the lumbar spine.
umbar spine stability is an important issue, especially given

From the Spine Biomechanics Laboratory, Department of
Kinesiology, Faculty of
pplied Health Sciences, University of Waterloo, ON, Canada.
Supported by the Natural Sciences and Engineering Research
Council of Canada
grant no. RGPIN36516-98).
No commercial party having a direct financial interest in the
results of the research
upporting this article has or will confer a benefit upon the
author(s) or upon any
rganization with which the author(s) is/are associated.
Reprint requests to Sylvain G. Grenier, PhD, Biomechanics,
Ergonomics and
inesiology Laboratory, School of Human Kinetics, Laurentian
University, Sudbury,
N P3E 2C6, Canada, e-mail: [email protected]
s
0003-9993/07/8801-10692$32.00/0
doi:10.1016/j.apmr.2006.10.014
rch Phys Med Rehabil Vol 88, January 2007
ts potential link to mechanisms of injury and associated clin-
cal efforts directed toward enhancing stability in patients.3 The
ay in which patients activate their abdominal muscles is
entral to the stability theme. For example, findings that trans-
ersus abdominis is recruited later, in some LBP sufferers,
ave led to speculation that it is related to an unhealthy or
nstable spine.4 The strategy to recruit the transversus abdo-
inis, through the abdominal hollowing technique, has been
roposed as an effective way to increase stability.5 Richardson
t al6 investigated the effect of bracing and hollowing on

acroiliac joint laxity in a nonfunctional task. They found that
oth improved stiffness but concluded that hollowing was
etter. However, the muscle resting levels differed between
roups. Until now, most of the supporting evidence, for the
ransversus abdominis being an important contributor to stabil-
ty, has been indirect and qualitative.
Motivation to focus on the transversus abdominis in the
linic has been provided by Hodges and Richardson5 who
ocumented delays of the transversus abdominis during rapid
rm movements in some people who have a history of low back
isorders, although this result has not been confirmed in recent
eports.7-9 It has also been suggested9 that, although the direc-
ion of limb movement does not affect preactivation time, the
agnitude of the arm movement perturbation does. Although
his does not provide direct evidence linking the transversus
bdominis to stability, it was suggested that this is evidence of
otor control deficit in chronic back pain patients. Additional
vidence of other types of motor control deficiencies in those
ith LBP include larger delays of onset in several torso mus-
les when the entire torso is moved quickly,10 inhibition of
nee extensors,11 and perturbed gluteal patterns while walking
nd inability to simultaneously breathe heavily and maintain
pine stability.12
It appears that the transversus abdominis may be activated
ndependently of other abdominal muscles at very low levels
f challenge (1%�2% of maximum voluntary contraction
MVC]).6 But at higher levels of activation, when people
erform tasks requiring spine stability, the transversus abdo-
inis has also been shown to be a synergist of internal
blique.13 Some clinical trials, testing the “stabilization exer-

ise,” have shown success in addressing LBP.1,2 However,
one of these studies have showed a direct link to the trans-
ersus abdominis and the mechanism for efficacy is not known.
or example, Vezina and Hubley-Kozey14 have measured ab-
ominal wall activation levels during abdominal hollowing
xercises and reported that none of their subjects were able to
ctivate only the transversus abdominis. Furthermore, David-
on and Hubley-Kozey15 showed that, as the demand of an
xercise progression increased, the abdominal muscles con-
erged, such that all muscles were activating to the same
ercentage of maximum at the highest exercise level. A recent
tudy16 suggests that hollowing and attempts to specifically
ctivate the transversus abdominis reduced the efficacy of the
xercise. One could argue that there is no quantitative evidence
dentifying the transversus abdominis as an important stabi-
izer, although our own clinical efficacy study17 has shown that
tabilization exercises are effective for patients with LBP.
N
t
g
Q
c
s
a
c
i
s
E
d

(
r
a
a
o
C
w
d
(
s
t
a
c
(
a
e
p
t
J
a
t
a
a
i
F
s
l
o
D
d
r
m

55ABDOMINAL ACTIVATION AND LUMBAR STABILITY,
Grenier
evertheless, it is not possible to make legitimate claims as to
he stabilizing role of any specific muscles, only that the
eneral approach was effective.
uantification of Stability
The definition of stability is a critical issue. Because clini-
ians generally do not have the technical capability to quantify
tability, kinematic “indicators” of instability have evolved. In
nd of itself, this is not a problem, except that these indicators
an be disguised with appropriate muscle recruitment. Stabil-
ty, as assessed in this study, is defined as the ability of the
pinal column to survive an applied perturbation (known as
uler column stability). If the work done (input energy or
isturbance) is greater than the potential energy of the column
energy stored in disks, ligaments, muscles, tendons), equilib-
ium will not be regained. From a clinical perspective, there is
subtle, but important, difference between excessive motion
nd instability; excessive motion does not imply instability,
nly the potential for instability (for a detailed description, see
ig 1. Stability analysis. This
chematic depicts the process
eading up to the calculation
f stability. Abbreviations:
M, distribution moment; dL/
t, derivative of length with
espect to time; EMG, electro-
yography.
holewicki and McGill18). The model used in this experiment, w
hich quantifies stability in this mechanical sense, is fully
escribed elsewhere,19 although an overview is provided here
fig 1).

In the context of lumbar stability, the effect of the transver-
us abdominis on the spine is still inconclusive. Isolated con-
raction of the transversus abdominis beyond very low levels of
ctivation has not been measured because all abdominal mus-
les become involved at more functional levels of activation
5%�20%).15 An investigation of specific abdominal muscle
ctivation strategies and individual muscles’ role in that strat-
gy should provide insight to the clinical decision-making
rocess. The hollowing technique has been developed as a
ransversus abdominis motor pattern retraining technique by
ull and Richardson20 to address the motor disturbances. It
ppears that others have assumed that it should be used as a
echnique to enhance spine stability. This article evaluates 2
bdominal activation strategies, hollowing and bracing, which
re techniques used clinically to improve lumbar spine stabil-
ty. Given the clinical issue stated previously and our previous
ork21,22 examining the stabilizing role of many torso muscles,
Arch Phys Med Rehabil Vol 88, January 2007
w
s
t
o
a
m
t
i
a
u

s
t
a
s
m
i
m
t
i
D
w
l
a
T
t
w
c
w
a
b
A
a
m
1
9
N
m
a
e
6
p
v

E
t
t
t
s
e
l
a
1
j
p
e
i
c
s
(
c
l
l
s
g
d
i
f
i
t
i
s
t
s
d
d

a
t
t
i
e
2
w
a
a
d
a
t
o
b
f
g
t
t
l
b
v
i
H
a
a
a
w
w
c
a
s
r
r
(
1
h

e
o
l
i
t
m
c
r
t
56 ABDOMINAL ACTIVATION AND LUMBAR STABILITY,
Grenier
A
e were motivated to quantify the mechanical impact of the 2
trategies on lumbar stability. It was therefore hypothesized
hat the abdominal hollowing strategy would be inferior to that
f a bracing strategy for enhancing stability. Spine stiffness is
lways stabilizing (summarized in McGill23). The question
otivating this study therefore is as follows: is the isolated
ransversus abdominis recruitment associated with the abdom-
nal hollowing technique a more effective stabilizer than a full
bdominal girdle cocontraction?
METHODS
In the present study, the comprehensive lumbar spine model
sed to quantify stability19 was enhanced to include a repre-
entation of transversus abdominis. Because pilot work showed
hat none of our subjects could perform an ideal “hollow”6 (ie,
ctivating only the transversus abdominis and internal oblique),
imulations were also conducted to artificially activate the
uscles in an “ideal” way. This was done, together with real,
n vivo data collection, with the understanding that subjects

ay have had imperfect technique. In this way, we were able
o evaluate “perfect hollowing” and bracing, as well as the
mperfect clinical reality.
ata Collection
The study proceeded with 2 components in parallel. The first
as the in vivo data collection, and the second was the simu-
ation. During the in vivo task, torso muscle electromyographic
ctivity and spine kinematics were recorded from each subject.
hese data were then input to the biomechanic model to de-
ermine spine stability. In the simulation, a single subject’s data
ere modified to reflect the anatomic and electromyographic
hanges associated with an “ideal” hollow and brace. In this
ay, several combinations of muscle activation were tested to
ssess their effect on stability.
In vivo: subjects. Data were collected from 8 healthy (no
ack pain in the past year) men between the ages of 20 and 33.
ll subjects provided informed consent, and the study was
pproved by the university ethics committee. Subjects had a
ean height � standard deviation (SD) of 1.82�0.06m (range,
.73�1.88m), a mean weight of 79.8�11.5kg (range, 60.5�
3.6kg), and a mean age of 23.8�4.33 years (range, 20�33y).
one had any experience with the transversus abdominis recruit-
ent training.
In vivo: kinematics. Spine kinematics were recorded with
3Space Isotrak unit,a which measured lumbar flexion and
xtension, lateral bend, and axial twist at a sample rate of

0Hz. The 3Space electromagnetic field source was strapped in
lace over the sacrum, and a sensor was worn over the T12
ertebrae, isolating lumbar motion. The 3Space unit returns
uler angles. These were adjusted for anatomic relevance to
he spine’s orthopedic axes. Flexion and extension corresponds
o rotation about y, lateral bend to rotation about x, and twist
o rotation about z.24
In vivo: electromyographic activity. Electromyographic
ignals were obtained by using Ag-Ag/Cl Meditrace surface
lectrodes,b in a bipolar configuration, spaced 25mm apart, in
ine with muscle fiber directions.25 The signals were then
mplifiedc (12-bit analog-to-digital conversion; sampling rate,
024Hz; frequency response, 10�1000Hz; common mode re-
ection ratio, 115dB at 60Hz; input impedance, �10G�). As
er the previously validated electrode positions,19 7 muscles on
ach side of the body were recorded for a total of 14 muscles,
ncluding the rectus abdominis (2cm lateral to the umbili-
us), internal oblique (caudal to the anterior superior iliac
pine and medial to the inguinal ligament), external oblique a
15cm lateral to the umbilicus), latissimus dorsi over the mus-
le belly (15cm lateral to T9), thoracic erector spinae (5cm
ateral to T9 over the muscle belly), lumbar erector spinae (3cm
ateral to L3), and the multifidus (2cm lateral to L5, angled
lightly with superior electrode more medial). The electromyo-
raphic signals were normalized to the amplitudes measured
uring the MVC procedure after rectification and low-pass filter-
ng at 2.5Hz. In addition to preparing the electromyography for
orce estimation, this process also removed electrocardiographic
nterference.

MVCs, for normalization, were performed in 2 separate
rials.26 First, the abdominal flexors were contracted maximally
n a seated position, leaning backward at 45°. Then, the exten-
ors were contracted maximally in a Biering-Sorensen posi-
ion.27 In both cases, the experimenter provided sufficient re-
istance for a maximal isometric contraction. Once this was
one, the hollowing and bracing strategies were explained and
emonstrated.
In vivo: protocol. Subjects were asked to hollow their
bdomen without sucking in their belly. They were told that
hey should be able to breathe normally. Assuming that the
ransversus abdominis was synergistic in its activation with the
nternal oblique,13 the subjects practiced until they were able to
asily achieve the required internal oblique activation target of
0% of MVC, as displayed on an oscilloscope. This compares
ith a range of approximately 12% of MVC of external oblique
ctivation in hollowing and 32% of MVC of external oblique
ctivation in bracing measured by Richardson et al.6 A brace is
ifferent from the hollow in several ways; the brace involves
ctivation of all abdominal muscles to a level that increases
orso stiffness. This does not mean that no motion can occur,
nly that the motion is controlled. Furthermore, abdominal
racing causes the back extensors to become active, which
urther enhances spine stiffness. Initially, the electromyo-
raphic patterns were verified with oscilloscope, and observa-
ions of initial muscle contraction with ultrasound ensured that
he activation patterns reproduced the technique reported in the
iterature.6 The ultrasound probe was placed at the rectus
order so that the 3 layers of the abdominal wall could be
iewed. In the current study, there were considerable variations
n the recruitment patterns used to achieve the hollowing.
ollowing trials were collected when subjects could show, on
n oscilloscope, a decrease in external oblique and rectus

bdominis activity, along with an increase in internal oblique
ctivity, although this increase was often minimal. The testing
as performed in an anatomically neutral standing posture,
ith a 10-kg load in either or both hands, depending on the
ondition. Over a period of 25 seconds, subjects were then
sked to relax for 5 seconds, “hollow” the abdomen for 5
econds, relax for 5, “brace” the abdomen for 5 seconds, and
elax for the final 5 seconds. These trials were repeated, in
andom fashion, 3 times each with (1) no load in the hands
bilateral no lift), (2) with 10kg in each hand (bilateral lift), (3)
0kg in the right hand only, and finally with (4) 10kg in the left
and only. Different load conditions were used to target the
ffect of asymmetric activation, as well as to address the issue
f increased intervertebral stiffness with higher compression
oads. The kinematic and electromyographic data were then
nput to the custom stability model. Note that, for this part of
he experiment, because transversus abdominis activity was not
easured, it was assumed to be a synergist of the anterior and
audal portions of internal oblique. Thus, the internal oblique
ecording was used to drive the transversus abdominis in
he model. Two studies have shown this to be a realistic
ssumption.13,28
D
b
p
w
d
e

a
f
i
m
s
e
h
a
t
p
l
w
t
b
a
i
w
r
c
(
f
m
w
c
m
u
s
m
m
t
s
i

t
v
m
v
a
e
a
t
o
o
b
w
m
e
T
a
a
“
n
a
1
2
3
4
a
v
a
c
T
w

2
a
t
m
m
f
a
i
F
s
d
a
Grenier
ata-Collection Procedures
Quantification of stability. Spine stability was calculated
y computing the potential energy of the spinal system. For the
urpose of summing potential energy, the muscle and tendons
ere considered as linear springs, whereas the intervertebral
isks were considered as torsional springs. The elastic potential
nergy was then summed in each rotational degree of freedom,
nd the work done by the external load was then subtracted
rom this. The second derivative of the potential energy matrix,
f greater than or equal to 0, indicated a stable system.19 This
ethod of quantifying stability has been applied to mechanical
tructures and validated repeatedly in civil and mechanical
ngineering.29 Its application to the osteoligamentous spine
as also been validated.30-32 Specific components of the model
re described as follows.
Passive contribution to potential energy. The skeleton of
he model consisted of 5 lumbar vertebrae between a rigid

elvis and sacrum and a rigid ribcage. The vertebrae were
inked by torsional springs representing the vertebral disks,
hich generated passive stiffness and allowed rotation but no
ranslation (about 3 orthogonal axes). Three-dimensional lum-
ar motion, measured with the 3Space Isotrak was proportion-
lly distributed among all vertebrae.33 The stiffnesses, includ-
ng stiffness of the disks, ligaments, fascia, skin, and viscera,34
ere also distributed among the vertebrae.19,33
Active contribution of potential energy. In addition to the
estorative moment of the passive tissues, the muscles also
ontributed a restorative moment to balance the external load
fig 2). The muscle force and stiffness estimates were obtained
rom a distribution-moment muscle model that used electro-
yographic and muscle length as input.35 The muscle forces
ere then applied through 118 muscle fascicles to the skeletal
omponents such that the moment they created balanced the
oment generated by the external load. Stability calculations
sed all of these variables as input, together with the muscle
tiffnesses calculated by the distribution-moment model.36
Modifications to the model. Anatomic improvements were
ade to better represent the application of transversus abdo-
inis forces on the spine. Specifically, the fascial attachment of
ig 2. A schematic representation of a simplified spine motion
egment to show the concept of linear (muscles, ligaments, ten-
fi
ons) and torsional (intervertebral disks) springs working to
sustain
n applied load.
he transversus abdominis on the lumbar vertebrae was repre-

ented with 10 fascicles bilaterally on the 5 segments (2 orig-
nating on the posterior tip of the lumbar spinous processes and
he other 2 originating on the transverse process of the lumbar
ertebrae). To capture the line of action of the fascial attach-
ents on the posterior spinal processes, the 10 fascicles con-
erged on a point 60cm directly lateral of L5. This arrangement
lso closely approximated the experimental findings of Tesh
t al37 that the compression cosine of the lateral transversus
bdominis force was 39% of its resultant magnitude. Because
ransversus abdominis activity was not measured, the stiffness
f the transversus abdominis was calculated by using internal
blique activity magnitude because these have been shown to
e synergistic. The length of the transversus abdominis fibers
as calculated based on the real hoop-like architecture of the
uscle. This architecture was intended to simply evaluate the
ffects of the muscles recognizing their attachment to the spine.
here were no assumptions made regarding the role of intra-
bdominal pressure to stiffen the abdominal wall.
Simulations. The objective of the simulation trials was to
rtificially adjust abdominal muscle activity levels to imitate
ideal” hollowing and bracing strategies because subjects could
ot completely isolate, or preferentially activate, any single
bdominal muscle to the extent required.
A total of 1 each of 5 types of simulations was performed.
. Simulation of the hollowing strategy: the transversus abdo-
minis and internal oblique electromyographic signals were
adjusted to an activity level at 20% of MVC, whereas the
rectus abdominis and external oblique were adjusted to 2%
of MVC. The activity levels in the back extensor muscles
were simply those measured.

. Simulation of the bracing strategy: all abdominal electro-
myographic signals were replaced by activity at 20% of
MVC. The activity levels in the back extensor muscles were
simply those measured.
. Simulation of the bracing strategy: the stabilizing effect of
the transversus abdominis was evaluated, and isolated, with
a sensitivity test in which transversus abdominis activation
was zeroed during a bracing simulation (bracing with no
transversus abdominis).
. Simulation of the bracing strategy: the procedure in simu-
lation 3 was repeated by zeroing each abdominal muscle
pair, similar to the muscle “knockout” approach of
Cholewicki and VanVliet.38
A. A right and left external oblique set to 0% of MVC.
B. A right and left internal oblique set to 0% of MVC.
C. A right and left rectus abdominis set to 0% of MVC.
In addition to these trials, across conditions, a sensitivity
nalysis was also conducted to assess the effect of the trans-
ersus abdominis alone on stability. First, only the transversus
bdominis was set to 0% of MVC. This was compared with the
ondition in which all abdominal muscles were left untouched.
hen, all abdominals except for the transversus abdominis
ere set to 0, whereas the transversus abdominis was set to
0% of MVC and then to 100% of MVC so that the transversus
bdominis was the only active muscle. In addition to modifying
he muscle activity of selected muscles in all simulations, the
oment arm of rectus abdominis (and consequently the attach-
ents of internal and external oblique) was shortened by 5cm,
or the hollowing simulations, to mimic the “drawing in” of the
bdomen (figs 3, 4). These artificially modified data were then
nput to the spine stability model as described earlier and in

gure 1.
S
c
t
w
p
i
d
t
2
p
b
l
w
t
a
p
s
p
w
i
t
n

h
(
f
n
b
r
t
t
2
r
t
l
t
s
o
s
s
t
u
t
w
F
p
t
o
t
u
g
“
s
s
F
d

w
i
N
d
P
s
a
Grenier
A
tatistical Analyses
The dependent variables from the in vivo trials were spine
ompression and the stability index, which is a general indica-
or of the stability of the equilibrium state of the spine.19
In vivo analysis. Using the R statistical package,39,d a
ithin-subject, repeated-measures analysis of variance was
erformed to examine the effect of combining each of 4 load-
ng conditions with the hollow or brace condition on the
ependent variables of stability and compression. In this study
ig 3. Because the moment is calculated as the product of force
and
erpendicular distance, moving the muscle attachment (A) closer
to
he joint (J) (as occurs during hollowing) shortens the moment
arm
f rectus abdominis. This has a large effect on the resulting
poten-
ial energy, and, consequently, stability as well. Stability of a
col-
mn with guy wires (or a spine with muscles) is a function of

eometry (the location of the support attachments), stiffness, the
balance” in stiffness in all directions, and column
imperfections,
uch as curvature. Hollowing reduces the potential energy and
tability by narrowing the geometric base.
ig 4. In the hollowing simulation, the moment arm of rectus ab-
ominis was reduced by 5cm (“B” in left panel), when compared
a
ith the bracing condition (right panel), to account for the change
n anatomic geometry. “A” defines the relaxed condition.
hen, the independent variable of “muscle activity pattern” had
conditions, hollowing and bracing, whereas the second inde-
endent variable of “load type” had 4 conditions: (1) load in
oth hands, (2) no load, (3) right-hand load, and (4) left-hand
oad. The Tukey post hoc honestly significant difference test
as used to investigate differences.
Simulation analysis. Simulation trials were compared to
he reference bracing trial, with the “muscle knockout” model,
s well as with the simulated hollowing condition by using a
aired t test. The change in (dependent) compression and
tability index between bracing and hollowing trials was ex-
ressed as a percentage change.
% change �
BRC � HLW
BRC
� 100 (1)
here BRC is bracing and HLW is hollowing.

RESULTS
Bracing stability was always greater than hollowing stabil-
ty, and asymmetric loads always produced greater stability
han symmetric loads. In in vivo trials, stability differed sig-
ificantly for hollowing and bracing conditions (bracing �
ollowing, P�.001) and between the loading conditions
P�.009) (table 1). Figure 5 shows a hollow-brace composite
or 1 subject showing this difference. Compression values were
ot different, either between hollowing and bracing (P�.54) or
etween loading conditions (P�.095). One of the subject’s
esults were excluded because the abdominal recruitment pat-
erns were not consistent with what had been requested for the
rials (ie, hollowing of the abdomen was not achieved with a
0% of MVC internal oblique activation). Post hoc testing
evealed that spine stability, in the no-load condition, was less
han both right- and left-side loads but not less than the bilateral
oad condition.
The simulation data supported the in vivo data, showing that
he hollowing was not as effective as bracing for increasing
tability in the lumbar spine. In fact, bracing improved stability
ver hollowing by 32%, with only a 15% increase in compres-
ion. Figure 6 shows the difference in stability, whereas figure 7
hows the difference in compression for 1 in vivo subject. In
able 2, 1 subject’s data were used for the simulation. The
nmodified hollow data were compared, for that subject, with
he brace data. The various simulations were also compared
ith the full brace trial. When removing only the transversus
Table 1: Mean Stability and Compression Value From In Vivo
Data Trials
Trial
Mean Stability

Index (Nm/rad)
Mean Compression
(N) at L4-5
Hollowing: no load 474.6�85.4 1866.5�451.4
Hollowing: 2-hand load 495.6�70.8 1929.2�360.0
Hollowing: left-hand load 517.7�55.8 2003.0�264.6
Hollowing: right-hand load 533.4�57.7 2041.7�286.6
Bracing: no load 511.3�39.5 1911.0�189.0
Bracing: 2-hand load 527.3�59.8 1989.1�247.9
Bracing: left-hand load 555.0�70.1 2060.4�223.1
Bracing: right-hand load 546.1�59.7 2008.6�266.1
OTE. Values are mean � SD. In the case of stability, there were
ifferences between muscle activity patterns (bracing �
hollowing,
�.001) and between the loading conditions (P�.009). The
compres-
ion values had no differences, either between loads or between
ctivation types.
bdominis from the bracing pattern (see table 2, Simulation
b
.
m
o
w
(
v
l
s
s

F
t
l
n
g
s
t
t
m
o
t
m
f
u
a
I
T
l
e
i
f
f
t
i
o
h
n
t
S
t
F
c

a
c
c
fi
F
t
t
a
i
F
u
w
a
o
M
A
N
t
t
o
E
t
l
Grenier
racing: no transversus abdominis), stability decreased by
14% with a less than 0.1% decrease in compression. Further-
ore, the importance of the transversus abdominis relative to
ther abdominal muscles in affecting stability was very small
hen assessed by a “muscle-knockout” approach40 (P�.01)

fig 8).
DISCUSSION
Is the abdominal hollowing technique and its specific trans-
ersus abdominis recruitment pattern a more effective stabi-
izer than a full abdominal girdle cocontraction? These data
uggest that it is not. The bracing strategy provided greater
tability than hollowing in both the simulation and in vivo data.
urthermore, for our subjects, the ability to activate just the
ransversus abdominis at functional levels was extremely chal-
enging, if not impossible, as evidenced by all other abdomi-
als’ electromyographic activity not being silent. This sug-
ested that the attempt to hollow would, in effect, result in
ome degree of bracing. The simulations showed that the
ransversus abdominis had virtually no effect on stability. Al-
hough Hodges et al41 have shown that the transversus abdo-
inis does contribute stiffness to the spine in a porcine model,
ur sensitivity tests (see fig 8) show that its relative contribu-
ig 5. A composite of 2 trials of in vivo subject data (a 2-hand
load
rial chosen at random) shows that bracing had greater stability
han an attempted hollowing. The ideal hollowing was not
chieved. Ideal hollowing was tested in the simulations. Note the
nitial overactivation when trying to establish the hollow.
1 2 3 4 5
0
100
200

300
400
500
600
700
800
900
1000
1100
Stb Index HLW
Stb Index BRC
Time (s)
S
ta
bi
lit
y
In
de
x
(N

m
/r
ad
)
ig 6. Lumbar stability increased by a significant margin (32%)
nder the simulated bracing condition. The hollow was conducted
ith 20% activation in both the lateral oblique and the
transversus
bdominis and 2% activation in both rectus abdominis and
external
blique. This was considered feasible after observing our
subjects.
ost of the stability was obtained through the internal oblique.
bbreviations: BRC, bracing; HLW, hollowing; Stb, Stability.
A
a
ion is very small when compared with the other abdominal
uscles. Assessment of this relative performance was absent
rom their experiment. Performing a hollowing pattern by
sing just the transversus abdominis greatly reduced stability,
s opposed to all other abdominal muscles being active (fig 9).
n addition, the compression-stability ratio favored bracing.
he fact that the bracing strategy was more effective at stabi-
izing than the hollowing strategy should not be interpreted as
vidence to diminish abdominal hollowing as a tool for train-
ng, or retraining the recruitment of the transversus abdominis
rom a motor control perspective, because this muscle does

orm a component of the abdominal girdle. However, many
herapists and coaches, in our experience, recommend “draw-
ng in” of the abdomen in an effort to increase stability. Based
n our results, this appears to be misdirected. In fact, 2 studies
ave recently concluded that focusing on general activity42 and
onspecific exercise16 is more beneficial for both pain reduc-
ion and functionality.
tudy Limitations
Several limitations should be addressed to guide interpreta-
ion of these results. The transversus abdominis in the model
1 2 3 4 5
0
250
500
750
1000
1250
1500
1750
2000
2250
2500
2750
3000

3250
3500
Cmp HLW
Cmp BRC
Time (s)
C
om
pr
es
si
on
(
N
)
ig 7. Although the external load component contributes to
lumbar
ompression equally among conditions, compression due to
muscle
ctivation differed between the simulation bracing and hollowing
ondition by as much as 500N. Note that the hollowing here in-
luded both the internal oblique and transversus abdominis as in
gure 6. Abbreviation: Cmp, compression.
Table 2: One Subject’s Data Were Taken and Modified
for the Simulation
Compared With Bracing
Stability %

Change
Compression
% Change
In vivo: no load hollowing 7.2 2.3
In vivo: 2-hand load hollowing 6.0 3.0
In vivo: left-hand load hollowing 6.7 2.8
In vivo: right-hand load hollowing 2.3 �1.7
Simulation hollowing 32.5 15.3
Simulation bracing: no TA .14 .00
Simulation bracing: no EO 16.5 11.0
Simulation bracing: no IO 32.5 12.7
Simulation bracing: no RA 12.6 10.6
OTE. The unmodified hollow data were compared, for that
subject,
o the brace data. The various simulations were also compared
with
he full brace trial. The percentage increase in stability of
bracing
ver hollow is clear, especially for the ideal simulation of
bracing.
ach condition was compared with pure bracing trial. Note that in
he subject data, compression increased under right-hand load
hol-
owing conditions, although only by 1.7%.
bbreviations: EO, external oblique; IO, internal oblique; RA,
rectus
bdominis; TA, transversus abdominis.

w
h
s
i
t
t
g
o
o
s
t
b
s
t
l
h
d
p
w
e
t
r
g
o
a
r
t
e
b
t
I
i
p

f
m
l
v
u
e
p
f
f
s
3
a
t
d
s
t
i
a
a
e
u
o
w
p
y
v
a
i
a
e
r
s
a
F

u
p
r
m
c
l
i
r
w
w
H
F
w
e
t
1
a
s
a
s
d
v
a
s
(
t
Grenier
A
as driven by the internal oblique activation profile because it
as been shown to share a synergistic activation for neutral
tatic postures, such as those tested in this study.13,43 The way
n which the transversus abdominis was modeled was an at-

empt to capture its force, moment, and stabilizing effects on
he spine. The architecture of the transversus abdominis sug-
ests a limited capability to influence lumbar stability on its
wn. In contrast, the oblique vertical “wrap” of the internal
blique, for example, has a much higher potential to directly
tabilize and compress the spinal column. Also, although a
ransversus abdominis muscle was modeled, the interaction
etween the transversus abdominis and intra-abdominal pres-
ure was not accounted for. Intra-abdominal pressure is linked
o abdominal wall activity, which stiffens and stabilizes the
umbar spine. However, at least 1 study44 suggests that a
ollowing strategy is unlikely to generate a greater intra-ab-
ominal pressure than full bracing. This being the case, if
ressure does affect stability, even more stability would result
hen bracing as compared with hollowing. The work of Tesh
t al45 contributed some insight as to the modeling of the
ransversus abdominis, its link with abdominal pressure, and
esistance to lateral bending moments in full flexion. The force
enerated by our transversus abdominis equivalent was on the
rder of 50N for a 20% MVC. They also report that intra-
bdominal pressure may contribute as much as 40% of the
estorative moment in lateral bending. Our own work shows
hat intra-abdominal pressure contributes significant stiffness,
specially in the neutral posture.44 We speculate that there may
e a binding of the abdominal muscle layers, when contracted
ogether, that creates more stiffness than the sum of the parts.
f this is the case, then the transversus abdominis would be an
mportant member of the “abdominal team” but no more im-
ig 8. Stability index in simulation trial. A single, 2-hand load
trial
as processed, with 4 sensitivity tests, to isolate the role of
differ-
nt muscles. The bottom trace is actually 2 conditions using the

ransversus abdominis as the only active abdominal muscle (at
00% and 20% of MVC). Whether the transversus abdominis is
ctive at 20% or 100% of MVC had little effect and those 2 plots
are
uperimposed. The top trace is actually 2 conditions: (1) with all
bdominal muscles active in a brace pattern (at whatever level
the
ubject had activated them) and (2) with only the transversus ab-
ominis removed. Only 1 line is visible because removing the
trans-
ersus abdominis has almost no effect, and its plot overlays the
full
ctivation plot. The variability in the stability plot, including
what
tability is present, results from variable activity in the extensor
and other modeled) muscles. Abbreviations: abs, abdominals;
TrA,
ransversus abdominis.
ortant than any of the other muscles.
h
H
In this study, the stability response was measured only with
ully anticipated loads in isometric, neutral postures, for sym-
etric and asymmetric loads. It is possible that for a sudden
oad, the combination of a pressurization response and trans-
ersus abdominis activation might maintain sufficient stability
ntil the remainder of the torso muscles are recruited. How-
ver, we remain skeptical of this possibility because the re-
orted 30-ms delay for transversus abdominis onset is just a
raction of the 130-ms electromechanical delay inherent in the
orce production for trunk muscle.46 In other words, a loss of
tability is not likely to occur within the window between
0ms, the time at which the transversus abdominis turns on,

nd 130ms, the time at which it begins to produce force. It is
herefore suggested that, although transversus abdominis onset
elay5 may be statistically significant, it is not mechanically
ignificant. In any case, recent work done with cocontraction of
he abdominals and its effect on stability, supports our find-
ngs.47 A general cocontraction of the abdominal wall, bal-
nced against antagonists and the load, is most effective in
ttaining and maintaining lumbar stability. There were consid-
rable variations in the recruitment patterns, shown by subjects,
sed to achieve a hollowing. This is normal and is an advantage
f our modeling approach, which is sensitive to the individual
ays people activate muscle. The simulation was able to re-
roduce the ideal. Furthermore, this study was performed on
oung healthy males with no LBP. Injury, LBP, or population
ariability may result in different muscle recruitment patterns
nd stability profiles. Again, this possible weakness in the
n vivo data was addressed via the simulations in which muscle
ctivation was under complete control of the experimenters to
lucidate the differences between ideal techniques and in vivo
eality.
The simulations were important, for interpretation of this
tudy, in 2 ways. First, the difficulty that participants had in
chieving a hollowing recruitment pattern necessitated some
ig 9. Stability index with abdominal muscles removed from sim-
lation. This histogram shows the stability, for actual subjects,
erforming the brace and an attempted hollow (shown at the far
ight). Then, the right and left abdominal muscle pairs were re-
oved from the analysis, each in turn, and the stability was recal-
ulated. Although the removal of the transversus abdominis made
ittle difference to stability, internal oblique was the most
important
n affecting stability, followed by external oblique and finally by
ectus abdominis. Abbreviations: Brc, normal brace; BrEO, brace

ith external oblique; BrIO, brace with internal oblique; BrRA,
brace
ith rectus abdominis; BrTA, brace with transversus abdominis;
lw, normal hollow; HwEO, hollow with external oblique; HwIO,
ollow with internal oblique; HwRA, hollow with rectus
abdominis;
wTA, hollow with transversus abdominis.
w
u
a
b
t
a
e
p
h
r
m
c
t
o
b
e
e
o
l
m
a
v
c
p

t
h
b
g
i
s
p
i
t
p
w
o
t
l
l
r
f
b
c
1
1
1
1
1
1
1

1
1
1
2
2
2
2
2
2
2
Grenier
ay of testing the “ideal” hollowing pattern. Second, the sim-
lations confirmed that, even if an ideal hollowing pattern was
chieved, it would still fall short of the stability provided by
racing. This is the case regardless of the fact that in this study
he transversus abdominis was not measured; the simulations
ssumed that it was optimally active. Recent work by Howarth
t al48 showed that the magnitude of the stability index is
roportional to the level of risk of becoming unstable (of
aving the stability index go below 0). Nonetheless, it is still a
elative measure so that it is not possible to compare the
agnitude of the index between people. It should only be used
omparatively within a subject, and even then, only within a
esting session. For this reason, the participants acted as their

wn controls. Finally, this work examined Euler column sta-
ility; it is possible that the abdominal muscles work differ-
ntly to buttress shear or pelvic instability. The mechanical
ffect on lumbar stability, of either the transversus abdominis
r of hollowing, has not, to our knowledge, been reported in the
iterature before this. It must be clearly stated that this is a
echanical analysis of the role of the transversus abdominis
nd other abdominal muscles. Therefore, whatever role trans-
ersus abdominis isolation training has in rehabilitation, it
annot be explained by a mechanically based stability princi-
le. If spinal stiffness is the ultimate goal in a training program,
hen bracing of the abdominal muscles is a clear winner over
ollowing. In reality, a single focus on either strategy may not
e optimal (or even possible) for functional tasks that show a
reat diversity in load and velocity.
CONCLUSIONS
This biomechanically based assessment suggests that brac-
ng of the abdominal muscles provides greater lumbar spine
tability than hollowing. According to our simulations, the
otential of the transversus abdominis to enhance stability, on
ts own, appears to be very limited. The inability to isolate the
ransversus abdominis in a functional context may be a moot
oint because in healthy men bracing increases spine stability
ith minimal increase in spine compression loads. Muscles
ther than the transversus abdominis contribute relatively more
o avoiding an unstable spine. Whatever the benefit underlying
ow-load transversus abdominis activation training, it is un-
ikely to be mechanical. There seems to be no mechanical
ationale for using stabilization exercises to enhance a hollow
or stabilization purposes; rather a brace creates patterns that
etter enhance stability.
Acknowledgment: We thank Jay Green for assistance in data

ollection.
References
1. Hides JA, Jull GA, Richardson CA. Long-term effects of
specific
stabilizing exercises for first-episode low back pain. Spine
2001;
26:E243-8.
2. O’Sullivan PB, Phyty GD, Twomey LT, Allison GT.
Evaluation
of specific stabilizing exercise in the treatment of chronic low
back pain with radiologic diagnosis of spondylolysis or spon-
dylolisthesis. Spine 1997;22:2959-67.
3. McGill SM. Low back disorders: evidence-based prevention
and
rehabilitation. Champaign: Human Kinetics. 2002.
4. Hodges PW. Is there a role for transversus abdominis in
lumbo-
pelvic stability? Man Ther 1999;4:74-86.
5. Hodges PW, Richardson CA. Inefficient muscular
stabilization of
the lumbar spine associated with low back pain. A motor control
evaluation of transversus abdominis. Spine 1996;21:2640-50.
6. Richardson CA, Snijders CJ, Hides JA, Damen L, Pas MS,
Storm J.
The relation between the transversus abdominis muscles,
sacroil-
iac joint mechanics, and low back pain. Spine 2002;27:399-405.

7. Allison GT, Henry SM. The influence of fatigue on trunk
muscle
responses to sudden arm movements, a pilot study. Clin
Biomech
(Bristol, Avon) 2002;17:414-7.
8. Newcomer KL, Jacobson TD, Gabriel DA, Larson DR, Brey
RH,
An KN. Muscle activation patterns in subjects with and without
low back pain. Arch Phys Med Rehabil 2002;83:816-21.
9. Hodges PW, Richardson CA. Altered trunk muscle
recruitment in
people with low back pain with upper limb movement at
different
speeds. Arch Phys Med Rehabil 1999;80:1005-12.
0. Radebold A, Cholewicki J, Panjabi MM, Patel TC. Muscle re-
sponse pattern to sudden trunk loading in healthy individuals
and
in patients with chronic low back pain. Spine 2000;25:947-54.
1. Suter E, Lindsay D. Back muscle fatigability is associated
with
knee extensor inhibition in subjects with low back pain. Spine
2001;26:E361-6.
2. McGill S, Grenier S, Bluhm M, Preuss R, Brown S, Russell
C.
Previous history of LBP with work loss is related to lingering
deficits in biomechanical, physiological, personal, psychosocial
and motor control characteristics. Ergonomics 2003;46:731-46.
3. Juker D, McGill S, Kropf P, Steffen T. Quantitative
intramuscular
myoelectric activity of lumbar portions of psoas and the abdom-

inal wall during a wide variety of tasks. Med Sci Sports Exerc
1998;30:301-10.
4. Vezina MJ, Hubley-Kozey CL. Muscle activation in
therapeutic
exercises to improve trunk stability. Arch Phys Med Rehabil
2000;81:1370-9.
5. Davidson KL, Hubley-Kozey CL. Trunk muscle responses to
demands of an exercise progression to improve dynamic spinal
stability. Arch Phys Med Rehabil 2005;86:216-23.
6. Koumantakis GA, Watson PJ, Oldham JA. Trunk muscle
stabili-
zation training plus general exercise versus general exercise
only:
randomized controlled trial of patients with recurrent low back
pain. Phys Ther 2005;85:209-25.
7. Hicks GE, Fritz JM, Delitto A, McGill SM. Preliminary
develop-
ment of a clinical prediction rule for determining which patients
with low back pain will respond to a stabilization exercise pro-
gram. Arch Phys Med Rehabil 2005;86:1753-62.
8. Cholewicki J, McGill SM. Mechanical stability of the in vivo
lumbar spine: implications for injury and chronic low back pain.
Clin Biomech (Bristol, Avon) 1996;11:1-15.
9. McGill SM. Low back stability: from formal description to
issues
for performance and rehabilitation. Exerc Sport Sci Rev
2001;29:
26-31.
0. Jull GA, Richardson CA. Motor control problems in patients

with
spinal pain: a new direction for therapeutic exercise. J
Manipula-
tive Physiol Ther 2000;23:115-7.
1. Kavcic N, Grenier S, McGill SM. Quantifying tissue loads
and
spine stability while performing commonly prescribed low back
stabilization exercises. Spine 2004;29:2319-29.
2. McGill SM, Grenier S, Kavcic N, Cholewicki J. Coordination
of
muscle activity to assure stability of the lumbar spine. J
Electro-
myogr Kinesiol 2003;13:353-9.
3. McGill SM. Ultimate back fitness and performance.
Waterloo:
Wabuno Publishers; 2004.
4. McGill S, Cholewicki J, Peach JP. Methodological
considerations
for using inductive sensors (3SPACE ISOTRAK) to monitor 3-D
orthopaedic joint motion. Clin Biomech (Bristol, Avon)
1997;12:
190-4.
5. Basmajian JV, De Luca CJ. Muscles alive: their functions re-
vealed by electromyography. 5th ed. Baltimore: Williams &
Wilkins; 1985.
6. McGill SM. Electromyographic activity of the abdominal and
low
back musculature during the generation of isometric and
dynamic
axial trunk torque: implications for lumbar mechanics. J Orthop

Res 1991;9:91-103.
2
2
2
3
3
3
3
3
3
3
3
3
3
4

4
4
4
4
4
4
4
4
a
b
c
d
Grenier
A
7. Biering-Sorensen F. Physical measurements as risk indicators
for
low back trouble over a one year period. Spine 1984;9:106-19.
8. Cresswell AG, Thorstensson A. Changes in intra-abdominal
pres-
sure, trunk muscle activation and force during isokinetic lifting
and lowering. Eur J Appl Physiol Occup Physiol 1994;68:315-
21.

9. Simitses GJ. An introduction to the elastic stability of
structures.
Malabar: RE Krieger; 1986.
0. Lucas DB, Bresler B. Stability of the ligamentous spine. San
Francisco: Univ California, Biomechanics Laboratory; 1961;
Technical report no. 40.
1. Crisco JJ III, Panjabi MM, Yamamoto I, Oxland T. Euler
stability
of the human ligamentous lumbar spine. Part II: Experiment.
Clin
Biomech 1992;7:27-32.
2. Crisco JJ III, Panjabi MM. The intersegmental and
multisegmental
muscles of the lumbar spine. A biomechanical model comparing
lateral stabilizing potential. Spine 1991;16:793-9.
3. White AA, Panjabi M. Clinical biomechanics of the spine.
To-
ronto: Lippincott; 1978.
4. McGill S, Seguin J, Bennett G. Passive stiffness of the
lumbar
torso in flexion, extension, lateral bending, and axial rotation.
Effect of belt wearing and breath holding. Spine 1994;19:696-
704.
5. Kavcic N, Grenier S, McGill SM. Determining the stabilizing
role
of individual torso muscles during rehabilitation exercises.
Spine
2004;29:1254-65.

6. Ma S, Zahalak GI. A distribution-moment model of
energetics in
skeletal muscle. J Biomech 1991;24:21-35.
7. Tesh KM, Evans JH, Shaw Dunn J, O’Brien JP. The
contribution
of skin, fascia, and ligaments to resisting flexion and the lumbar
spine. In: Whittle M, Harris D, editors. Biomechanical measure-
ment in orthopaedic practice. Oxford: Clarendon Pr; 1985. p
179-87.
8. Cholewicki J, VanVliet JJ IV. Relative contribution of trunk
muscles to the stability of the lumbar spine during isometric
exertions. Clin Biomech (Bristol, Avon) 2002;17:99-105.
9. R Development Core Team. R: a language and environment
for
statistical computing. Vienna: R Foundation for Statistical
Com-
puting; 2006.
0. Panjabi MM, Lydon C, Vasavada A, Grob D, Crisco JJ 3rd,
Dvorak J. On the understanding of clinical instability. Spine
1994;19:2642-50.
1. Hodges P, Kaigle Holm A, Holm S, et al. Intervertebral
stiffness
of the spine is increased by evoked contraction of transversus
abdominis and the diaphragm: in vivo porcine studies. Spine
2003;28:2594-601.
2. Hurwitz EL, Morgenstern H, Chiao C. Effects of recreational
physical activity and back exercises on low back pain and psy-

chological distress: findings from the UCLA Low Back Pain
Study. Am J Public Health 2005;95:1817-24.
3. Cresswell AG. Responses of intra-abdominal pressure and
abdom-
inal muscle activity during dynamic trunk loading in man [pub-
lished erratum in: Eur J Appl Physiol 1993;67:97]. Eur J Appl
Physiol Occup Physiol 1993;66:315-20.
4. Grenier SG. Stabilization strategies of the lumbar spine in
vivo
[PhD dissertation]. Waterloo: Faculty of Applied Health
Sciences,
Univ Waterloo; 2002. p 170.
5. Tesh KM, Shaw Dunn J, Evans JH. The abdominal muscles
and
vertebral stability. Spine 1987;12:501-8.
6. van Dieen JH, Thissen CE, van de Ven AJ, Toussaint HM.
The
electro-mechanical delay of the erector spinae muscle: influence
of rate of force development, fatigue and electrode location. Eur
J Appl Physiol Occup Physiol 1991;63:216-22.
7. Granata KP, Orishimo KF, Sanford AH. Trunk muscle
coactiva-
tion in preparation for sudden load. J Electromyogr Kinesiol
2001;11:247-54.
8. Howarth SJ, Allison AE, Grenier SG, Cholewicki J, McGill
SM.
On the implications of interpreting the stability index: a spine
example. J Biomech 2004;37:1147-54.
Suppliers

. Polhemus, 40 Hercules Dr, PO Box 560, Colchester, VT
05446.
. Graphic Controls Corp, Medical Products Div, PO Box 1274,
Buffalo, NY 14240.
. Bortec Biomedical, 239 Springborough Wy, Calgary, AB T3H
5M8, Canada.
. R statistical package. Available at: http://www.R-project.org.
Ac-
cessed October 6, 2006.
http://www.R-project.orgQuantification of Lumbar Stability by
Using 2 Different Abdominal Activation
StrategiesQuantification of StabilityMETHODSData
CollectionIn vivo: subjectsIn vivo: kinematicsIn vivo:
electromyographic activityIn vivo: protocolData-Collection
ProceduresQuantification of stabilityPassive contribution to
potential energyActive contribution of potential
energyModifications to the modelSimulationsStatistical
AnalysesIn vivo analysisSimulation
analysisRESULTSDISCUSSIONStudy
LimitationsCONCLUSIONSAcknowledgmentReferencesSupplie
rs
Chapter 5
From Building a Model toAdaptive Robust Decision Making
Using Systems Modeling
Erik Pruyt
Abstract Starting from the state-of-the-art and recent evolutions
in the field of system dynamics modeling and simulation, this
chapter sketches a plausible near term future of the broader field
of systems modeling and simulation. In the near
termfuture,differentsystemsmodelingschoolsareexpectedtofurthe
rintegrateand accelerate the adoption of methods and techniques
from related fields like policy analysis, datascience,

machinelearning, andcomputerscience.Theresultingfuture state
of the art of the modeling field is illustrated by three recent
pilot projects. Each of these projects required further
integration of different modeling and simulation approaches and
related disciplines as discussed in this chapter.These examples
also
illustratewhichgapsneedtobefilledinordertomeettheexpectationso
frealdecision makers facing complex uncertain issues.
5.1 Introduction
Many systems, issues, and grand challenges are characterized by
dynamic complexity, i.e., intricate time evolutionary behavior,
often on multiple dimensions of interest. Many dynamically
complex systems and issues are relatively well known,
buthavepersistedforalongtimeduetothefactthattheirdynamiccomp
lexitymakes them hard to understand and properly manage or
solve. Other complex systems and issues—especially rapidly
changing systems and future grand challenges—are largely
unknown and unpredictable. Most unaided human beings are
notoriously bad at dealing with dynamically complex issues—
whether the issues dealt with are persistent or unknown. That is,
without the help of computational approaches, most
humanbeingsareunabletoassesspotentialdynamicsofcomplexsyste
msandissues, and are unable to assess the appropriateness of
policies to manage or address them.
Modelingandsimulationisafieldthatdevelopsandappliescomputati
onalmethods to study complex systems and solve problems
related to complex issues. Over the past half century, multiple
modeling methods for simulating such issues and for advising
decision makers facing them have emerged or have been further
developed. Examples include system dynamics (SD) modeling,
discrete event simulation (DES), multi-actor systems modeling
(MAS), agent-based modeling (ABM), and complex adaptive
systems modeling (CAS).All too often, these developments have
taken place in distinct fields, such as the SD field or theABM
field, developing into separate “schools,” each ascribing

dynamic complexity to the complex underlying
mechanismstheyfocuson,suchasfeedbackeffectsandaccumulation
effectsinSDor heterogenousactor-
specific(inter)actionsinABM.Theisolateddevelopmentwithin
separate traditions has limited the potential to learn across
fields and advance faster
andmoreeffectivelytowardsthesharedgoalofunderstandingcomple
xsystemsand supporting decision makers facing complex issues.
Recent evolutions in modeling and simulation together with the
recent explosive
growthincomputationalpower,data,socialmedia,andotherevolutio
nsincomputer science have created new opportunities for model-
based analysis and decision making. These internal and external
evolutions are likely to break through silos of old, open up new
opportunities for social simulation and model-based decision
making, and stir up the broader field of systems modeling and
simulation. Today, different modeling approaches are already
used in parallel, in series, and in mixed form, and several
hybrid approaches are emerging. But not only are different
modeling traditions being mixed and matched in multiple ways,
modeling and simulation fields have also started to adopt—or
have accelerated their adoption of—useful methods
andtechniquesfromotherdisciplinesincludingoperationsresearch,
policyanalysis, data analytics, machine learning, and computer
science. The field of modeling and simulation is consequently
turning into an interdisciplinary field in which various modeling
schools and related disciplines are gradually being integrated.
In practice, the blending process and the adoption of
methodological innovations have just started. Although some
ways to integrate systems modeling methods and many
innovationshavebeendemonstrated,
furtherintegrationandmassiveadoptionarestill awaited.
Moreover, other multi-methods and potential innovations are
still in an experimental phase or are yet to be demonstrated and
adopted.
Inthischapter,someofthesedevelopmentswillbediscussed,apicture

ofthenear future state of the art of modeling and simulation is
drawn, and a few examples of
integratedsystemsmodelingarebrieflydiscussed.TheSDmethodisu
sedtoillustrate these developments. Starting with a short
introduction to the traditional SD method in Sect. 5.2, some
recent and current innovations in SD are discussed in Sect. 5.3,
resulting in a picture of the state of modeling and simulation in
Sect. 5.4. A few
examplesarethenbrieflydiscussedinSect.5.5toillustratewhatthesed
evelopments couldresultinandwhatthefuturestate-of-the-
artofsystemsmodelingandsimulation could look like. Finally,
conclusions are drawn in Sect. 5.6.
5.2 System Dynamics Modeling and Simulation of Old
System dynamics was first developed in the second half of the
1950s by Jay W.
Forresterandwasfurtherdevelopedintoaconsistentmethodbuiltons
pecificmethodological choices1. It is a method for modeling and
simulating dynamically complex systems or issues characterized
by feedback effects and accumulation effects. Feedback means
that the present and future of issues or systems, depend—
through a chain of causal relations—on their own past. In SD
models, system boundaries are set broadly enough to include all
important feedback effects and generative mechanisms.
Accumulation relates not only to building up real stocks—of
people, items, (infra)structures, etc.,—but also to building up
mental or other states. In SD models, stock variables and the
underlying integral equations are used to group largely
homogenous persons/items/... and keep track of their aggregated
dynamics over time. Together, feedback and accumulation
effects generate dynamically complex behavior both inside SD
models and—so it is assumed in SD—in real systems. Other
important characteristic of SD are (i) the reliance on relatively
enduring
conceptualsystemsrepresentationsinpeople’sminds,akamentalmo

dels(Doyleand Ford 1999, p. 414), as prime source of “rich”
information (Forrester 1961; Doyle and Ford 1998); (ii) the use
of causal loop diagrams and stock-ﬂow diagrams to represent
feedback and accumulation effects (Lane 2000); (iii) the use of
credibility and ﬁtness for purpose as main criteria for model
validation (Barlas 1996); and (iv)
theinterpretationofsimulationrunsintermsofgeneralbehaviorpatte
rns,akamodes of behavior (Meadows and Robinson 1985). In
SD, the behavior of a system is to be explained by a dynamic
hypothesis, i.e., a causal theory for the behavior (Lane 2000;
Sterman 2000). This causal theory is
formalizedasamodelthatcanbesimulatedtogeneratedynamicbehavi
or.Simulating the model thus allows one to explore the link
between the hypothesized system structure and the time
evolutionary behavior arising out of it (Lane 2000). Not
surprisingly, these characteristics make SD particularly useful
for dealing
withcomplexsystemsorissuesthatarecharacterizedbyimportantsys
temfeedback effects and accumulation effects. SD modeling is
mostly used to model core system structures or core structures
underlying issues, to simulate their resulting behavior,
andtostudythelinkbetweentheunderlyingcausalstructureofissuesa
ndmodelsand
theresultingbehavior.SDmodels,whicharemostlyrelativelysmalla
ndmanageable,
thusallowforexperimentationinavirtuallaboratory.Asaconsequenc
e, SDmodels are also extremely useful for model-based policy
analysis, for designing adaptive policies (i.e., policies that
automatically adapt to the circumstances), and for testing
theirpolicyrobustness(i.e.,whethertheyperformwellenoughacross
alargevariety of circumstances).
In terms of application domains, SD is used for studying many
complex social– technical systems and solving policy problems
in many application domains, for example, in health policy,

resource policy, energy policy, environmental policy, housing
policy, education policy, innovation policy, social–economic
policy, and other public policy domains. But it is also used for
studying all sorts of business dynamics problems, for strategic
planning, for solving supply chain problems, etc. At the
inception of the SD method, SD models were almost entirely
continuous,
i.e.,systemsofdifferentialequations,butovertimemoreandmoredis
creteandother noncontinuous elements crept in. Other
evolutionary adaptations in line with ideas from the earliest
days of the field, like the use of Group Model Building to elicit
mental models of groups of stakeholders (Vennix 1996) or the
use of SD models as engines for serious games, were also
readily adopted by almost the entire field. But slightly more
revolutionary innovations were not as easily and massively
adopted. In other words, the identity and appearance of
traditional SD was well established bythemid-1980sanddoes—
atfirstsight—notseemtohavechangedfundamentally since then.
5.3 Recent Innovations and Expected Evolutions
5.3.1 Recent and Current Innovations
Looking in somewhat more detail at innovations within the SD
field and its adoption of innovations from other fields shows
that many—often seemingly more revolutionary—innovations
have been introduced and demonstrated, but that they have not
been massively adopted yet.
Forinstance,intermsofquantitativemodeling,systemdynamicistsha
veinvested
inspatiallyspecificSDmodeling(RuthandPieper1994;Struben2005
;BenDorand Kaza2012),individualagent-
basedSDmodelingaswellasmixedandhybridABMSDmodeling(Cas
tilloandSaysal2005;Osgood2009;Feolaetal.2012;Rahmandad and
Sterman 2008), and micro–macro modeling (Fallah-Fini et al.
2014). Examples of recent developments in simulation setup and
execution include model calibration
andbootstrapping(Oliva2003;Dogan2007),differenttypesofsampli
ng(Fiddaman

2002;Ford1990;Clemsonetal.1995;IslamandPruyt2014),multi-
modelandmultimethodsimulation(PruytandKwakkel2014;
Moorlag2014), anddifferenttypesof optimization approaches
used for a variety of purposes (Coyle 1985; Miller 1998;
Coyle1999;GrahamandAriza1998;Hamaratetal.2013,2014).Rece
ntinnovations in model testing, analysis, and visualization of
model outputs in SD include the
developmentandapplicationofnewmethodsforsensitivityanduncer
taintyanalysis (Hearne 2010; Eker et al.2014), formal model
analysis methods to study the link between structure and
behavior (Kampmann and Oliva 2008, 2009; Saleh et al. 2010),
methods for testing policy robustness across wide ranges of
uncertainties
(Lempertetal.2003),statisticalpackagesandscreeningtechniques(F
ordandFlynn 2005; Taylor et al. 2010), pattern testing and time
series classification techniques
(YücelandBarlas2011;Yücel2012;SuculluandYücel2014;Islaman
dPruyt2014), and machine learning techniques (Pruyt et al.
2013; Kwakkel et al. 2014; Pruyt et al. 2014c). These methods
and techniques can be used together with SD models to identify
root causes of problems, to identify adaptive policies that
properly address
theserootcauses,totestandoptimizetheeffectivenessofpoliciesacro
sswideranges of assumptions (i.e., policy robustness), etc. From
this perspective, these methods
andtechniquesareactuallyjustevolutionaryinnovationsinlinewithe
arlySDideas. And large-scale adoption of the aforementioned
innovations would allow the SD
field,andbyextensionthelargersystemsmodelingfield,tomovefrom
“experiential art” to “computational science.” Most of the
aforementioned innovations are actually integrated in particular
SD approaches like in exploratory system dynamics modelling
and analysis (ESDMA), which is an SD approach for studying
dynamic complexity under deep uncertainty. Deep uncertainty
could be defined as a situation in which analysts do not know or

cannotagreeon(i)anunderlyingmodel,(ii)probabilitydistributions
ofkeyvariables
andparameters,and/or(iii)thevalueofalternativeoutcomes(Lemper
tetal.2003).It
isoftenencounteredinsituationscharacterizedbyeithertoolittleinfo
rmationortoo much information (e.g., conflicting information or
different worldviews). ESDMA
isthecombinationofexploratorymodelingandanalysis(EMA),akaro
bustdecision making, developed during the past two decades
(Bankes 1993; Lempert et al. 2000;
Bankes2002;Lempertetal.2006)andSDmodeling.EMAisaresearch
methodology
fordevelopingandusingmodelstosupportdecisionmakingunderdee
puncertainty.
Itisnotamodelingmethod,inspiteofthefactthatitrequirescomputati
onalmodels. EMA can be useful when relevant information that
can be exploited by building computational models exists, but
this information is insufficient to specify a single model that
accurately describes system behavior (Kwakkel and Pruyt
2013a). In such situations, it is better to construct and use
ensembles of plausible models since ensembles of models can
capture more of the un/available information than any
individualmodel(Bankes2002).Ensemblesofmodelscanthenbeuse
dtodealwith model uncertainty, different perspectives, value
diversity, inconsistent information, etc.—in short, with deep
uncertainty.2
InEMA(andthusinESDMA),theinfluenceofaplethoraofuncertainti
es,including method and model uncertainty, are systematically
assessed and used to design policies: sampling and multi-
model/multi-method simulation are used to generate
ensemblesofsimulationrunstowhichtimeseriesclassificationandm
achinelearning techniques are applied for generating insights.
Multi-objective robust optimization
(Hamaratetal.2013,2014)isusedtoidentifypolicyleversanddefinep
olicytriggers, and by doing so, support the design of adaptive
robust policies. And regret-based approaches are used to test

policy robustness across large ensembles of plausible runs
(Lempert et al. 2003). EMA and ESDMA can be performed with
TU Delft’s
EMA workbench software, which is an open source tool3 that
integrates multimethod, multi-model, multi-
policysimulationwithdatamanagement, visualization, and
analysis.
Thelatterisjustoneoftherecentinnovationsinmodelingandsimulati
onsoftware and platforms: online modeling and simulation
platforms, online ﬂight simulator
andgamingplatforms,andpackagesformakinghybridmodelshavebe
endeveloped too. And modeling and simulation across platforms
will also become reality soon: the eXtensible Model Interchange
LanguagE (XMILE) project (Diker and Allen 2005; Eberlein
and Chichakly 2013) aims at facilitating the storage, sharing,
and combination of simulation models and parts thereof across
software packages and
acrossmodelingschoolsandmayeasetheinterconnectionwith(real-
time)databases,
statisticalandanalyticalsoftwarepackages,andorganizationalinfor
mationandcommunicationtechnology(ICT)infrastructures.Noteth
atthisisalreadypossibletoday with scripting languages and
software packages with scripting capabilities like the
aforementioned EMA workbench.
5.3.2 Current and Expected Evolutions
Threecurrentevolutionsareexpectedtofurtherreinforcethisshiftfro
m“experiential art” to “computational science.” The ﬁrst
evolution relates to the development of “smarter” methods,
techniques, and tools (i.e., methods, techniques, and tools that
provide more insights and deeper understanding at reduced
computational cost). Similar to the development of formal
model analysis techniques that smartened the traditional SD
approach, new methods, techniques, and tools are currently
being developed to smarten modeling and simulation approaches
that rely on “brute force” sampling, for example, adaptive

output-oriented sampling to span the space of possible dynamics
(Islam and Pruyt
2014)orsmartermachinelearningtechniques(Pruytetal.2013;Kwak
keletal.2014; Pruyt et al. 2014c) and time series classification
techniques (Yücel and Barlas2011; Yücel 2012; Sucullu
andYücel 2014; Islam and Pruyt 2014), and (multi-objective)
robust optimization techniques (Hamarat et al. 2013, 2014).
Partly related to the previous evolution are developments relates
to “big data,”
datamanagement,anddatascience.AlthoughtraditionalSDmodelin
gissometimes calleddata-
poormodeling,itdoesnotmeanitis,norshouldbe.SDsoftwarepackag
es allow one to get data from, and write simulation runs to,
databases. Moreover, data are also used in SD to calibrate
parameters or bootstrap parameter ranges. But more could be
done, especially in the era of “big data.” Big data simply refers
here to much more data than was until recently manageable. Big
data requires data science techniques to make it manageable and
useful. Data science may be used in
modelingandsimulation(i)toobtainusefulinputsfromdata(e.g.,fro
mreal-timebig data sources), (ii) to analyze and interpret model-
generated data (i.e., big artificial
data),(iii)tocomparesimulatedandrealdynamics(i.e.,formonitorin
gandcontrol), and (iv) to infer parts of models from data (Pruyt
et al. 2014c). Interestingly, data science techniques that are
useful for obtaining useful inputs from data may also be made
useful for analyzing and interpreting model-generated data, and
vice versa. Online social media are interesting sources of real-
world big data for modeling and
simulation,bothasinputstomodels,tocomparesimulatedandrealdyn
amics,andto
informmodeldevelopmentormodelselection.Therearemanyapplica
tiondomains in which the combination of data science and
modeling and simulation would be beneficial. Examples, some
of which are elaborated below, include policy making with

regard to crime fighting, infectious diseases, cybersecurity,
national safety and security, financial stress testing, energy
transitions, and marketing. Another urgently needed innovation
relates to model-based empowerment of
decisionmakers.Althoughexistingflightsimulatorandgamingplatfo
rmsareusefulfor developing and distributing educational flight
simulators and games, and interfaces
canbebuiltinSDpackages,usingthemtodevelopinterfacesforreal-
worldreal-time
decisionmakingandintegratingthemintoexistingICTsystemsisdiffi
cultandtime consuming. In many cases, companies and
organizations want these capabilities inhouse, even in their
boardroom, instead of being dependent on analyses by external
or internal analysts. The latter requires user-friendly interfaces
on top of (sets of) models possibly connected to real-time data
sources. These interfaces should allow for experimentation,
simulation, thoroughly analysis of simulation results, adaptive
robust policy design, and policy robustness testing.
5.4 Future State of Practice of Systems Modeling and
Simulation
Theserecentevolutionsinmodelingandsimulationtogetherwithther
ecentexplosive
growthincomputationalpower,data,socialmedia,andotherevolutio
nsincomputer
sciencemayheraldthebeginningofanewwaveofinnovationandadopt
ion,moving the modeling and simulation field from building a
single model to simultaneously simulating multiple models and
uncertainties; from single method to multi-method and hybrid
modeling and simulation; from modeling and simulation with
sparse data to modeling and simulation with (near real-time) big
data; from simulating and analyzing a few simulation runs to
simulating and simultaneously analyzing wellselected
ensembles of runs; from using models for intuitive policy
testing to using models as instruments for designing adaptive
robust policies; and from developing educational flight
simulators to fully integrated decision support. For each of the

modeling schools, additional adaptations could be foreseen too.
In case of SD, it may for example involve a shift from
developing purely endogenous to largely endogenous models;
from fully aggregated models to sufﬁciently spatially explicit
and heterogenous models; from qualitative participatory
modeling
to quantitative participatory simulation; and from using SD to
combining problem structuring and policy analysis tools,
modeling and simulation, machine learning techniques, and
(multi-objective) robust optimization. Adoption of these recent,
current, and expected innovations could result in the future state
of the art4 of systems modeling as displayed in Fig. 5.1. As
indicated by (I) in Fig. 5.1, it will be possible to simultaneously
use multiple hypotheses (i.e.,
simulationmodelsfromthesameordifferenttraditionsorhybrids),fo
rdifferentgoals
includingthesearchfordeeperunderstandingandpolicyinsights,exp
erimentationin avirtuallaboratory,future-
orientedexploration,robustpolicydesign,androbustness testing
under deep uncertainty. Sets of simulation models may be used
to represent differentperspectivesorplausibletheories,
todealwithmethodologicaluncertainty, or to deal with a plethora
of important characteristics (e.g., agent characteristics,
feedback and accumulation effects, spatial and network effects)
without necessarily
havingtointegratetheminasinglesimulationmodel.Themainadvant
agesofusing multiple models for doing so are that each of the
models in the ensemble of models remains manageable and that
the ensemble of simulation runs generated with the
ensemble of models is likely to be more diverse which allows
for testing policy robustness across a wider range of plausible
futures. Some of these models may be connected to real-time or
near real-time data streams, and some models may even be
inferred in part with smart data science tools from data sources

(see (II) in Fig. 5.1). Storing the outputs of these simulation
models in databases and applying data science techniques may
enhance our understanding, may generate policy insights, and
may allow for testing policy robustness across large
multidimensional uncertainty spaces (see (III) in Fig. 5.1). And
userfriendly interfaces on top of these interconnected models
may eventually empower policy makers, enabling them to really
do model-based policy making.
Note,however,thattheintegratedsystemsmodelingapproachsketch
edinFig.5.1 may only suit a limited set of goals, decision
makers, and issues. Single model simulation properly serves
many goals, decision makers, and issues well enough for multi-
model/multi-method, data-rich, exploratory, policy-oriented
approaches not to be required. However, there are most
certainly goals, decision makers, and issues that do.
5.5 Examples
Although all of the above is possible today, it should be noted
that this is the current state of science, not the state of common
practice yet. Applying all these methods and techniques to real
issues is still challenging, and shows where innovations are
most needed. The following examples illustrate what is possible
today as well as what the most important gaps are that remain to
be filled. The first example shows that relatively simple systems
models simulated under deep uncertainty allow for generating
useful ensembles of many simulation runs. Using methods and
techniques from related disciplines to analyze the resulting
artificialdatasetshelpstogenerateimportantpolicyinsights.Andsim
ulationofpolicies
acrosstheensemblesallowstotestforpolicyrobustness.Thisfirstcase
nevertheless shows that there are opportunities for multi-method
and hybrid approaches as well as for connecting systems models
to real-time data streams. The second example extends the first
example towards a system-of-systems
approachwithmanysimulationmodelsgeneratingevenlargerensem
blesofsimulation runs. Smart sampling and scenario discovery
techniques are then required to reduce the resulting data sets to

manageable proportions. The third example shows a recent
attempt to develop a smart model-based decision-support system
for dealing with another deeply uncertain issue. This example
shows that it is almost possible to empower decision makers.
Interfaces with
advancedanalyticalcapabilitiesaswellaseasierandbetterintegratio
nwithexisting ICT systems are required though. This example
also illustrates the need for more advanced hybrid systems
models as well as the need to connect systems models to real-
time geo-spatial data.
5.5.1 Assessing the Risk, and Monitoring, of New Infectious
Diseases
Thefirstcase,whichisdescribedinmoredetailin(PruytandHamarat2
010;Pruytet al. 2013), relates to assessing outbreaks of new flu
variants. Outbreaks of new (variants of) infectious diseases are
deeply uncertain. For example, in the first months
afterthefirstreportsabouttheoutbreakofanewfluvariantinMexicoan
dtheUSA,
muchremainedunknownaboutthepossibledynamicsandconsequenc
esofthispossible epidemic/pandemic of the new flu variant,
referred to today as new influenza A(H1N1)v.Table 5.1 shows
that more and better information became available over
time,butalsothatmanyuncertaintiesremained.However,evenwithth
eseremaining uncertainties, it is possible to model and simulate
this flu variant under deep uncertainty, for example with the
simplistic simulation model displayed in Fig. 5.2, since flu
outbreaks can be modeled. Simulating this model thousands of
times over very wide uncertainty ranges for each of the
uncertain variables generates the 3D cloud of potential
outbreaks displayed in Fig. 5.3a. In this figure, the worst flu
peak (0–50 months) is displayed on the X-axis, the infected
fraction during the worst flu peak (0–50%) is displayed on the
Y-axis, and the cumulative number of fatal cases in the Western
world (0– 50.000.000)isdisplayedontheZ-
axis.This3Dplotshowsthatthemostcatastrophic outbreaks are

likely to happen within the first year or during the first winter
season followingtheoutbreak.
Usingmachinelearningalgorithmstoexplorethisensemble of
simulation runs helps to generate important policy insights
(e.g., which policy levers to address). Testing different variants
of the same policy shows that adaptive policies outperform their
static counterparts (compare Fig. 5.3b and c). Figure 5.3d
finally shows that adaptive policies can be further improved
using multi-objective robust optimization.
However,takingdeepuncertaintyseriouslyintoaccountwouldrequir
esimulating more than a single model from a single modeling
method: it would be better to
simultaneouslysimulateCAS,ABM,SD,andhybridmodelsunderdee
puncertainty and use the resulting ensemble of simulation runs.
Moreover, near real-time geospatial data (from twitter, medical
records, etc.) may also be used in combination with simulation
models, for example, to gradually reduce the ensemble of
modelgenerated data. Both suggested improvements would be
possible today.
5.5.2 Integrated Risk-CapabilityAnalysis under Deep
Uncertainty
The second example relates to risk assessment and capability
planning for National Safety and Security. Since 2001, many
nations have invested in the development of all-hazard
integrated risk-capability assessment (IRCA) approaches. All-
hazard IRCAsintegratescenario-basedriskassessment,
capabilityanalysis, andcapabilitybased planning approaches to
reduce all sorts of risks—from natural hazards, over technical
failures to malicious threats—by enhancing capabilities for
dealing with
them. Current IRCAs mainly allow dealing with one or a few
specific scenarios for a limited set of relatively simple event-
based and relatively certain risks, but not for dealing with a
plethora of risks that are highly uncertain and complex,

combinationsofmeasuresandcapabilitieswithuncertainanddynami
ceffects, anddivergent opinions about degrees of
(un)desirability of risks and capability investments.
Thenextgenerationmodel-
basedIRCAsmaysolvemanyoftheshortcomingsof
theIRCAsthatarecurrentlybeingused.
Figure5.4displaysanextgenerationIRCA for dealing with all
sorts of highly uncertain dynamic risks. This IRCA approach,
described in more detail in Pruyt et al. (2012), combines EMA
and modeling and simulation, both for the risk assessment and
the capability analysis phases. First, risks—like outbreaks of
new flu variants—are modeled and simulated many times
acrosstheirmultidimensionaluncertaintyspacestogenerateanense
mbleofplausible risk scenarios for each of the risks. Time series
classification and machine learning techniques are then used to
identify much smaller ensembles of exemplars that are
representativeforthelargerensembles.Theseensemblesofexemplar
sarethenused as inputs to a generic capability analysis model.
The capability analysis model is subsequently simulated for
different capabilities strategies under deep uncertainty (i.e.,
simulating the uncertainty pertaining to their effectiveness) over
all ensembles
ofexemplarstocalculatethepotentialofcapabilitiesstrategiestoredu
cetheserisks.
Fig. 5.4 Model-based integrated risk-capability analysis (IRCA)
Finally, multi-objective robust optimization helps to identify
capabilities strategies that are robust. Not only does this
systems-of-systems approach allow to generate thousands of
variants per risk type over many types of risks and to perform
capability analyses across all sorts of risk and under
uncertainty, it also allows one to find sets of capabilities that
are effective across many uncertain risks. Hence, this integrated
model-basedapproachallowsfordealingwithcapabilitiesinanall-
hazardwayunder deep uncertainty. This approach is currently
being smartened using adaptive output-oriented sampling

techniques and new time-series classification methods that
together help to identify the largest variety of dynamics with
the minimal amount of simulations. Covering the largest variety
of dynamics with the minimal amount of exemplars is
desirable,forperformingautomatedmulti-
hazardcapabilityanalysisovermanyrisks is—due to the nature of
the multi-objective robust optimization techniques used—
computationally very expensive. This approach is also being
changed from a multimodel approach into a multi-method
approach. Whereas, until recently, sets of SD models were used;
there are good reasons to extend this approach to other types of
systemsmodelingapproachesthatmaybebettersuitedforparticularri
sksor—using multipleapproaches—
helptodealwithmethodologicaluncertainty.Finally,settings of
some of the risks and capabilities, as well as exogenous
uncertainties, may also be fed with (near) real-world data.
5.5.3 Policing Under Deep Uncertainty
The third example relates to another deeply uncertain issue,
high-impact crimes (HIC).An SD model and related tools (see
Fig. 5.5) were developed some years ago
inviewofincreasingtheeffectivenessofthefightagainstHIC,moresp
ecificallythe
fightagainstrobberyandburglary.HICsrequireasystemicperspectiv
eandapproach:
Thesecrimesarecharacterizedbyimportantsystemiceffectsintimea
ndspace,such as learning and specialization effects, “waterbed
effects” between different HICs and precincts, accumulations
(prison time) and delays (in policing and jurisdiction),
preventive effects, and other causal effects (ex-post preventive
measures). HICs are also characterized by deep uncertainty:
Most perpetrators are unknown and even though their archetypal
crime-related habits may be known to some extent at some
pointintime,
accuratetimeandgeographicallyspecificpredictionscannotbemade.
At the same time, is part of the HIC system well known and is a
lot of real-world information related to these crimes available.

ImportantplayersintheHICsystembesidesthepoliceand(potential)
perpetrators are potential victims (households and shopkeepers),
partners in the judicial system (the public prosecution service,
the prison system, etc.). Hence, the HIC system is dynamically
complex, deeply uncertain, but also data rich, and contingent
upon external conditions.
Themaingoalsofthispilotprojectweretosupportstrategicpolicymak
ingunder deep uncertainty and to test and monitor the
effectiveness of policies to ﬁght HIC. The SD model (I) was
used as an engine behind the interface for policy makers (II) to
explore plausible effects of policies under deep uncertainty and
identify realworld pilots that could possibly increase the
understanding about the system and
effectivenessofinterventions(III),toimplementthesepilots(IV),an
dmonitortheir outcomes (V). Real-world data from the pilots
and improved understanding about the functioning of the real
system allow for improving the model.
Today, a lot of real-world geo-spatial information related to
HICs is available online and in (near) real time which allows to
automatically update the data and model, and hence, increase its
value for the policy makers. The model used in this project was
an ESDMA model. That is, uncertainties were included by
means of sets of plausible assumptions and uncertainty ranges.
Although this could already be argued to be a multi-model
approach, hybrid models or a multi-method approach would
really be needed to deal more properly with systems, agents,
and spatial characteristics. Moreover, better interfaces and
connectors to existing ICT systems and databases would also be
needed to turn this pilot into a real decision-support system that
would allow chiefs of police to experiment in a virtual world
connected to the real world, and to develop and test adaptive
robust policies on the spot.
5.6 Conclusions
Recent and current evolutions in modeling and simulation
together with the recent explosive growth in computational

power, data, social media, and other evolutions in computer
science have created new opportunities for model-based
analysis and decision making. Multi-method and hybrid
modeling and simulation approaches are being developed to
make existing modeling and simulation approaches appropriate
for dealing
withagentsystemcharacteristics,spatialandnetworkaspects,deepu
ncertainty,and otherimportantaspects.
Datascienceandmachinelearningtechniquesarecurrently
beingdevelopedintotechniquesthatcanprovideusefulinputsforsim
ulationmodels as well as for building models. Machine learning
algorithms, formal model analysis methods, analytical
approaches, and new visualization techniques are being
developedtomakesenseofmodelsandgenerateusefulpolicyinsights.
Andmethodsand tools are being developed to turn intuitive
policy making into model-based policy design. Some of these
evolutions were discussed and illustrated in this chapter.
Itwasalsoarguedandshownthateasierconnectorstodatabases,
tosocialmedia,
toothercomputerprograms,andtoICTsystems,aswellasbetterinterf
acingsoftware need to be developed to allow any systems
modeler to turn systems models into real decision-support
systems. Doing so would turn the art of modeling into the
computational science of simulation. It would most likely also
shift the focus of attention from building a model to using
ensembles of systems models for adaptive robust decision
making.
References
Bankes SC (1993) Exploratory modeling for policy analysis.
Operat Res 41(3):435–449
BankesSC(2002)Toolsandtechniquesfordevelopingpoliciesforco
mplexanduncertainsystems. Proc NatlAcad Sci U SA
99(3):7263–7266
[email protected]
5 From Building a Model toAdaptive Robust Decision Making
Using Systems Modeling 91

BarlasY(1996)Formalaspectsofmodelvalidityandvalidationinsyst
emdynamics.SystDynRev 12(3):183–210 BenDor TK, Kaza N
(2012)A theory of spatial system archetypes. Syst Dyn Rev
28(2):109–130
BryantBP,LempertRJ(2010)Thinkinginsidethebox:aparticipatory
,computer-assistedapproach to scenario discovery. Technol
Forecast Soc Change 77(1):34–49 Castillo D, SayselAK (2005)
Simulation of common pool resource field experiments: a
behavioral model of collective action. Ecol Econ 55(3):420–436
ClemsonB,TangY,PyneJ,UnalR(1995)Efficientmethodsforsensiti
vityanalysis.SystDynRev 11(1):31–49 Coyle GR (1985) The use
of optimization methods for policy design in a system dynamics
model. Syst Dyn Rev 1(1):81–91 Coyle GR (1999) Simulation
by repeated optimisation. J Operational Res Society 50:429–438
DikerVG,Allen RB (2005) XMILE: towards an XML
interchange language for system dynamics models. Syst Dyn
Rev 21(4):351–359 Dogan G (2007) Bootstrapping for
confidence interval estimation and hypothesis testing for
parameters of system dynamics models. Syst Dyn Rev
23(4):415–436 Doyle JK, Ford DN (1998) Mental models
concepts for system dynamics research. Syst Dyn Rev 14:3–29
Doyle JK, Ford DN (1999) Mental models concepts revisited:
some clarifications and a reply to Lane. Syst Dyn Rev 15:411–
415 Eberlein RL, Chichakly KJ (2013) XMILE: a new standard
for system dynamics. Syst Dyn Rev 29(3):188–195 Eker S,
Slinger JH, Daalen EC, Yücel G (2014) Sensitivity analysis of
graphical functions. Syst Dyn Rev 30(3):186–205 Feola G,
Gallati JA, Binder CR (2012) Exploring behavioural change
through an agent-oriented system dynamics model: the use of
personal protective equipment among pesticide applicators in
Colombia. Syst Dyn Rev 28(1):69–93 Fallah-
FiniS,RahmandadR,ChenHJ,WangY(2014)Connectingmicrodyna
micsandpopulation distributions in system dynamics models.
Syst Dyn Rev 29(4):197–215
FiddamanTS(2002)Exploringpolicyoptionswithabehavioralclima
te–economymodel.SystDyn Rev 18(2):243–267. Ford A (1990)

Estimating the impact of efficiency standards on the uncertainty
of the northwest electric system. Oper Res 38(4):580–597 Ford
A, Flynn H (2005) Statistical screening of system dynamics
models. Syst Dyn Rev 21(1): 273–303 Forrester JW (1961)
Industrial dynamics. MIT Press, Cambridge (Waltham, MA:
Pegasus Communications.AND Productivity Press: Portland,
OR) Forrester JW (1968) Principles of systems.Wright-Allen
Press, Cambridge Forrester JW (1969) Urban dynamics. MIT
Press, Cambridge Forrester JW (1991) The beginning of system
dynamics. D-Note D-4165–1 Forrester JW (2007) System
dynamics—a personal view of the first fifty years. Syst Dyn Rev
23(2–3):345–358
GrahamAK,ArizaCA(2003)Dynamic,hardandstrategicquestions:u
singoptimizationtoanswer a marketing resource allocation
question. Syst Dyn Rev 19(1):27–46 Hamarat C, Kwakkel JH,
Pruyt E (2013) Adaptive robust design under deep uncertainty.
Technol Forecast Soc Change 80(3):408–418 Hamarat C,
Kwakkel JH, Pruyt E, Loonen ET (2014) An exploratory
approach for adaptive policymaking by using multi-objective
robust optimization. Simul Model Pract Theory 46:25–39
HearneJW(2010)Anautomatedmethodforextendingsensitivityanal
ysistomodelfunctions. Nat Resour Model 23(2):107–120
HomerJB(2012)Modelsthatmatter:selectedwritingsonsystemdyna
mics1985–2010.Grapeseed Press, Barrytown
[email protected]
92 E. Pruyt
Islam T, Pruyt E (2014) An adaptive sampling method for
examining the behavioral spectrum of long-term metal scarcity.
In: Proceedings of the 32nd international conference of the
System Dynamics Society. System Dynamics Society, Delft,
The Netherlands Kampmann CE, Oliva R (2008) Structural
dominance analysis and theory building in system dynamics.
Syst Res Behav Sci 25(4):505–519 Kampmann CE, Oliva R
(2009) Analytical methods for structural dominance analysis in
system dynamics. In: Meyers R (ed) Encyclopedia of
complexity and systems science. Springer, New York, pp 8948–

8967 Kwakkel JH,AupingWL, Pruyt E (2013) Dynamic scenario
discovery under deep uncertainty: the future of copper. Technol
Forecast Soc Change 80(4):789–800
KwakkelJH,AupingWL,PruytE(2014)Comparingbehavioraldyna
micsacrossmodels: thecase
ofcopper.In:Proceedingsofthe32ndinternationalconferenceoftheS
ystemDynamicsSociety. System Dynamics Society, Delft, The
Netherlands
KwakkelJH,PruytE(2013a)Usingsystemdynamicsforgrandchallen
ges:theESDMAapproach. Syst Res Behav Sci
doi:10.1002/sres.2225 Kwakkel JH, Pruyt E (2013b)
Exploratory modeling and analysis, an approach for model-
based foresight under deep uncertainty. Technol Forecast Soc
Change 80(3):419–431 Lane DC (2000) Diagramming
conventions in system dynamics. J Opl Res Soc 51(2):241–245
LempertRJ,GrovesDG,PopperSW,BankesSC(2006)Ageneral,
analyticmethodforgenerating robust strategies and narrative
scenarios. Manag Sci 52(4): 514–528 Lempert RJ, Popper SW,
Bankes SC (2003) Shaping the next one hundred years: new
methods forquantitative, long-termpolicyanalysis.
RANDreportMR-1626,TheRANDPardeeCenter, Santa Monica,
CA Lempert RJ, Schlesinger ME (2000) Robust strategies for
abating climate change. Climat Change 45(3–4):387–401
LogtensT,PruytE,GijsbersG(2012)SocietalagingintheNetherland
s:exploratorysystemdynamics modeling and analysis. In:
Proceedings of the 30th international conference of the System
Dynamics Society. System Dynamics Society, St.-Gallen, CH
Meadows DH, Robinson JM (1985)The electronic Oracle.
Computer models and social decisions. Wiley, Chichester
MillerJ(1998)Activenonlineartests(ANTs)ofcomplexsimulation
models.ManagSci44(6):820– 830 Moorlag R (2014) Exploring
the effects of shale gas development on natural gas markets: a
multimethodapproach.In:Proceedingsofthe32ndinternationalconf
erenceoftheSystemDynamics Society. System Dynamics
Society, Delft, The Netherlands Oliva R (2003) Model
calibration as a testing strategy for system dynamics models.

Eur J Operat Res 151(3):552–568 Osgood N (2009) Lightening
the performance burden of individual-based models through
dimensional analysis and scale modeling. Syst Dyn Rev
25(2):101–134 Pruyt E (2013) Small system dynamics models
for big issues: triple jump towards real-world complexity. TU
Delft Library, Delft. http://simulation.tbm.tudelft.nl/ Pruyt E,
Cunningham SC, Kwakkel JH, de Bruijn JA (2014) From data-
poor to data-rich: system dynamics in the era of big data. In:
Proceedings of the 32nd international conference of the System
Dynamics Society. System Dynamics Society, Delft, The
Netherlands Pruyt E, Hamarat C (2010) The inﬂuenza
A(H1N1)v pandemic: an exploratory system dynamics
approach.In:Proceedingsofthe28thinternationalconferenceoftheS
ystemDynamicsSociety. System Dynamics Society, Seoul,
Korea PruytE,HamaratC,KwakkelJH(2012)Integratedrisk-
capabilityanalysisunderdeepuncertainty: an integrated ESDMA
approach. In: Proceedings of the 30th international conference
of the System Dynamics Society. System Dynamics Society, St.-
Gallen Pruyt E, Kwakkel JH (2014) Radicalization under deep
uncertainty: a multi-model exploration of activism, extremism,
and terrorism. Syst Dyn Rev 30:1–28 doi:10.1002/sdr.1510
Core Training: Stabilizing the Confusion
Mark D. Faries and Mike Greenwood, PhD, CSCS,* D; FNSCA
Baylor University, Waco, Texas
© National Strength and Conditioning Association
Volume 29, Number 2, pages 10–25
Keywords: core; core musculature; core strength; core stability;
lumbo-pelvic-hip complex; spinal stability; functional training;
transversus abdominus; multifidus
R

ecently, an infomercial promised
its audience that the advertised
piece of exercise equipment
would give anyone an “attractive core.”
There are unfortunate, and at times hu-
morous, misconceptions associated
with core muscle training and the idea
of someone having an attractive trans-
versus abdominis or attractive multi-
fidus muscle. To the trained eye, it
was apparent that these advertisers
were referring to the potential chis-
eled appearance of the rectus abdo-
minis and perhaps external obliques,
but similar to many misperceptions,
these individuals did not have a com-
plete understanding of what the core
truly is. At times, the same confusion
is noted in the exercise physiology, fit-
ness, and strength and conditioning
professions. The confusion runs from
the specific anatomy of the core with
regard to defining what it truly is, to
whether particular exercises are de-
signed to enhance core strength or
core stability, to the definition of core
exercise, to its separation from func-
tional exercise, and finally to the ef-
fects of core training on performance
outcomes. Many times this confusion
is as simple as pure semantics and/or
differences in terminology, but in any
case, the confusion does more to di-
vide the misunderstood topic than it

does to combine research areas and
training strategies. Using available re-
search, this article attempts to educate
the readership on an extremely popu-
lar, but controversial, topic. In hopes
of eliminating much of the confusion
associated with the core musculature,
the specific intent of this article is to
provide an idea of where current core
research resides, thereby enabling di-
rection for future scientific research
and application in the strength and
conditioning fields.
Core Strength Versus
Core Stability
It is wise to begin this section by describ-
ing a general foundational overview of
the core, and then discuss the differ-
ences between core strength and core
stability. The “core musculature” can be
defined generally as the 29 pairs of mus-
cles that support the lumbo-pelvic-hip
complex in order to stabilize the spine,
pelvis, and kinetic chain during func-
tional movements (26). The core is also
commonly referred to as the “power-
house” or the foundation of all limb
movement (1). These muscles are theo-
rized to create this foundation for move-
ment through muscle contraction that
provides direct support and increased
intra-abdominal pressure to the inher-
ently unstable spine (10, 25, 33, 61). To
ensure stability of the spine in order to
produce force and to prevent injury,

trunk muscles must have sufficient
strength, endurance, and recruitment
patterns (10).
“Strength,” in reference to this article,
can be defined as the ability of a muscle
to exert or withstand force. Active con-
trol of spine stability, in this case, is
achieved through the regulation of this
force in the surrounding muscles (16).
When instability is present, there is a
s u m m a r y
Confusion exists regarding what the
core musculature is, how it is evalu-
ated, how it is trained, and how it is
applied to functional performance.
The core musculature is divided into
2 systems, local (stabilization) and
global (movement), with distinction
between core-strength, core-stabili-
ty, and functional exercises.
10 April 2007 • Strength and Conditioning Journal

failure to maintain correct vertebral
alignment, or, in other words, a failure
in the musculature to apply enough
force to stabilize the spine. So, “stabili-
ty” describes the ability of the body to
control the whole range of motion of a
joint so there is no major deformity,
neurological deficit, or incapacitating
pain (51, 53). In general, the goal of the
core musculature is to stabilize the spine
during functional demands, because the
body wants to maximize this stability (1,
16). This level of stability and kinematic
response of the trunk is determined by
the mechanical stability level of the
spine and the reflex response of the
trunk muscles prior to force being ap-
plied to the body (16). Limb movement
provides exertional force onto the spine,
where the magnitude of reactive forces is
proportional to the inertia of the limb
(35, 37), whereas coactivation of the ag-
onistic and antagonistic trunk muscles
work to stiffen the lumbar spine to in-
crease its stability (16). There is a con-
cern, which is discussed later, that too
much strength or force from core mus-
culature actually can cause greater insta-
bility if it is not directed correctly. Also,
there is evidence to support that en-
durance is the more important training
variable when it comes to the core mus-
culature (46, 47).
With a general understanding of the goal
of the core musculature to stabilize the

OQASlstaaltlawsc.docx

Recommended

Recommended

More Related Content

Similar to OQASlstaaltlawsc.docx

Similar to OQASlstaaltlawsc.docx (20)

More from vannagoforth

More from vannagoforth (20)

Recently uploaded

Recently uploaded (20)

OQASlstaaltlawsc.docx