PRADS 2007

Dynamic Stability Criteria Based on the Simulation of Full Scale Ac-

cidents
Author Name(s): Stefan Krueger, Florian Kluwe
Hamburg University of Technology (TUHH), Institute of Ship Design and Ship Safety
Hamburg, Germany
Abstract
It has become obvious that modern ships suffer from
problems related to their seakeeping-behavior, which is
mainly related to large amplitude roll motion in head
and following seas.
As these effects are not covered by the existing intact
stability criteria, an additional concept was developed.
This new concept allows us to quantify the risk of the
occurrence of large roll angles by calculating a capsiz-
ing index.
The question left open so far was, how much stability
would be required for a certain ship to meet a distinct
safety level against capsizing.
This paper presents an approach addressing this ques-
tion based on the analysis of some real capsizing acci-
dents analyzed by applying various criteria proposed in
the past, in comparison to the new intact stability index.
For this purpose our simulation code, ROLLS, was
applied, taking into account relevant effects such as
cargo shift, additional heeling moments and water in-
gress.
Keywords
Ship safety; parametric rolling; intact stability; capsiz-
ing; probability; stability criteria; seakeeping perform-
ance
Introduction
In a concerted effort to address shipping accidents, a
number of capsizing criteria have been proposed in the
last decades, either based on model tests and simula-
tions or on empirical observations. A brief introduction
to a selection of these criteria introduced by German
research groups is given in the first section of this paper.
All presented criteria are intended to reduce the capsize
risk of ships in heavy weather. Most of these criteria do
not take into account dynamic effects of ships traveling
in a rough seaway.
New techniques such as numerical motion simulations
in the time domain have improved our knowledge on
the phenomena and the situations in which ships are
endangered with respect to large roll angles. Today this
increased knowledge allows us to address exactly those
dynamic aspects lacking in most of the older criteria.
This concern seems to be absolutely necessary, as mod-
ern hull designs seem to be even more endangered by
phenomena like parametric roll than traditional designs.
Moreover the mean ship size and speed have increased
in the last decades, also contributing to the fact that the
current intact stability rules are not able to guarantee a
sufficient safety level for all ships.
Therefore, a new concept was developed called Insuffi-
cient Stability Event Index {ISEI} based on a large
database of ships simulated in various sea states. An
introduction to this new approach is given in the second
part of this document.
To get an idea of the capabilities the new index has with
respect to ship safety in heavy seas, a number of real
capsizing accidents were re-investigated. The loading
condition of the ship present at the time of the accident,
which can always be clearly identified as not safe,
was analyzed with a set of intact stability criteria includ-
ing the new index. Finally an attempt was made to iden-
tify the stability increase necessary to omit the individ-
ual accidents. All three examples of accidents chosen
and clearly related to insufficient intact stability are
presented in the third part of this document. More acci-
dents are currently being investigated at TUHH, but no
final results are available thus far.
Overview on Selected Capsizing Criteria
Empirical Criteria Related to Righting Levers
The aim of these criteria is to ensure sufficient safety of
ships in heavy weather by identifying significant, stabil-
ity related characteristics of the ships lever arm curves.
Wendelss concept of Balancing Righting and Heeling
Levers:
Wendel and his group developed a concept whereby the
stability of ships should be evaluated on the basis of an
individual balance of righting and heeling levers (Arndt
(1960) . The dynamic effects of capsizing as such are
disregarded in this concept, but the stability reduction is
taken into account by using the mean value of the crest
and trough condition lever arms instead of the stillwater
righting lever, which is questionable from todays point
of knowledge. The theoretical background of Wendel's
concept is described in Wendel (1954) or Arndt (1960)
.The German Navys stability standard BV1033 is based
on this criterion.

The C-Factor Concept for Container Vessels Larger
than 100m in Length:
With the introduction of container vessels the average
beam-to-depth-ratio of the world merchant fleet grew
significantly from ca. 1.60 in 1960 to ca. 1.9 in 1980.
An increased beam-to-depth ratio leads to larger initial
stability, whereas added form stability is significantly
reduced. Therefore, Blume and Wagner carried out a
number of model tests for container vessels. Based on
the results, Blume tried to establish a criterion for the
minimum stability of vessels in rough weather, (Blume
and Hattendorf, 1987a). The findings lead to the devel-
opment of the C-factor concept, which enhances the
original Rahola-criteria.
For example, this task is done by replacing the static
requirement for the righting lever at 30 degrees being
larger or equal 0.2 m according to Rahola by the con-
stant value divided by C, where C is calculated as fol-
lows (Blume, 1987):

L C
C
KG
T
B
D T
C
WP
B
100
2

=

[1]

Here, T denotes the draft, D a modified depth including
hatches, KG is the center of gravity above base line. C
B

and C
WP
denote the block- and the waterline-coefficient,
respectively.
The C-factor today is part of the IMO Code on Intact
Stability for certain types of vessels above 100m in
length, but as the overall code, it is not mandatory. Fi-
nally the problem still remains that the C-factor is re-
lated to the still water righting lever curve, which is not
sufficiently representative for seakeeping problems.
Capsizing Criteria Derived from Model Tests or Simu-
lations
The Kastner/Roden Criterion for a Minimum GM to
Prevent Pure Loss Failures:
Based on model tests carried out on the inland lake
Ploen in Germany by Kastner (1962), a method was
developed to determine a minimum GM required to
prevent the vessel from capsizing in rough weather. The
authors observed the interesting phenomenon that a
clear limiting GM seemed to exist, distinguishing be-
tween ships being safe or unsafe with respect to capsiz-
ing. The criterion is based on the probability density
function for the time to capsize determined during the
model tests. The authors then asked for a cumulated
probability of 95% for the complement event not cap-
sized. Then the time interval T
k
is determined which
the ship must survive to fulfill the requirement given
above. Assuming that the ship always capsizes in the
largest wave a
k
occurring during T
k
the capsizing prob-
ability is linked to the probability of occurrence of that
wave.
Now a maximum wave height a
k
can be determined
which has lead to the capsizing in a specific situation,
e.g. during a model test. Now, assuming a probability
for a non-capsize, a related wave height a
nk
the ship
needs to survive in order to be sufficiently safe can be
determined in the same way. The authors then con-
cludes that the GM- value of the vessel must be in-
creased by the ratio which is defined by the these two
wave heights:
nk
k
nk
k
a
a
GM
GM
=

[2]
This ratio is somewhat doubtful from todays point of
knowledge as the assumptions made clearly fail in the
case where the GM gets close to zero.

Soedings Concept of Simulating Rare Events by Artifi-
cially Amplified Wave Heights:
In principle event probabilities can be determined sim-
ply by counting them during model tests or numerical
simulations. But, as extreme events (e.g. capsizing) are
rare, it is difficult to determine significant values for
capsizing probabilities during model tests and numerical
simulations due to the limited duration and the resulting
small number of occurrences.
Therefore, Soeding and Tonguc (1986) suggest the
simulations be run in artificially high waves. By assum-
ing Rayleigh-distributed amplitudes, the capsizing prob-
ability can be extrapolated to the actual wave height of
interest by the following relationship:
25 . 1 ) ln(
25 . 1 ) ln(
2
2
+
+
=
act
sim
act
sim
p
p
H
H

[3]

Here H denotes the actual (act) or the simulated (sim)
wave height, respectively. P denotes the capsizing prob-
ability, using the same indices. However, the proposed
criterion does not provide a procedure to determine the
enlargement factor for the wave height. Additionally the
concept does not include any threshold values for the
capsizing probability.

The Blume-Criterion
Blume and Hattendorff (1987b) developed this criterion
to evaluate the ship safety with respect to capsizing in
following and stern quartering seas by model tests. For
each run during the model test the maximum roll angle
is registered. Then the residual area E
R
below the still
water lever arm is calculated, limited by the maximum
roll angle and the point of vanishing stability (see Fig.
1). If the ship capsizes during the run, E
R
is set to zero.
Finally a ship is regarded as safe against capsizing if it
fulfills the following requirement:
0 3 > s ER [4]

Here
R E denotes the residual area averaged by all runs,
s represents the standard deviation of E
R
. By this a sta-
bility limit, represented by either a minimum GM or by
a limiting maximum wave height can be determined.

Fig. 1: Residual area below the righting leer curve
The Insufficient Stability Event Index (ISEI)
After some incidents related to parametric rolling with
container vessels become known at the end of the last
decade, a German research group was established to
develop dynamic stability criteria, which should be
based on numerical simulations. The simulation code
ROLLS, originally developed by Kroeger (1987) and
Petey (1988), was chosen to serve as the basis for the
evaluation of seakeeping related problems,. The code
was validated and further enhanced by Cramer and
Krueger (2005). A research program was established in
which a large number of model tests for different mod-
ern hull forms were carried out with tailored wave se-
quences to validate the simulation code. It was con-
cluded that the simulation code was able to predict all
relevant phenomena related to the problem of insuffi-
cient stability in waves with sufficient accuracy. There-
fore, it was decided to develop a concept for minimum
stability based exclusively on numerical motion simula-
tions. Based on the numerical simulations, the following
main findings were made or confirmed:
Both model tests and simulations confirmed
that critical situations endangering the ship
with respect to large roll amplitudes are ob-
served in head as well as following seas.
No capsizing events were found in beam seas
at zero speed.
The most dangerous scenarios appeared to be
those where the ship was traveling in following
and stern quartering seas.
In head and head-quartering seas, large rolling
angles were observed, but capsizing usually did
not occur. This is due to the fact that critical
resonances are connected to relatively low val-
ues of GM in following seas, and to high GM
values in head seas. The model tests were con-
ducted close to potentially critical resonances.
Unlike the expectations by previous authors, wave-
lengths significantly shorter than the ship length could
endanger the vessel, whereas wavelengths significantly
larger than ship length did not initiate large roll ampli-
tudes.
In contradiction to previous criteria, it was decided to
determine all possible scenarios that may lead to a dan-
gerous situation, but not to quantify just how dangerous
a specific situation actually is. When defining limiting
stability values, it is of importance to assess the prob-
ability of a specific loading condition being dangerous
or not for the vessel. For this application it is not of
practical interest to get the exact capsizing rate during
the simulation, but it is singularly important to know if
the ship did fail. Based on this, the concept is aimed
towards determining long-term probabilities rather
short-term probabilities. Thus, the concept requires a
methodology to distinguish between being safe or un-
safe for a ship in a specific situation without counting
the actual up-crossing rates.
Given that such a methodology is available, the total
long term probability for a dangerous situation happen-
ing in a specific loading condition can be defined, then
by the insufficient stability event index (ISEI), which is
defined by the following equation (see also Krueger and
Kluwe (2006)):
1 3 / 1 1 3 / 1
0 0
2
0
1 3 / 1
) , , , (
) , (
1 3 / 1
max
min
dT dH d dv v T H p
T H p ISEI
s s dang
T H
v
v v
sea
s

= = =
=
[5]

Here p
sea
denotes the probability of occurrence of a
specific sea state defined by the significant wave height
H
1/3
and the characteristic (peak) period T
1
, whereas
p
dang
represents the probability for the actual loading
condition leading to a dangerous situation under the
condition of a specific sea state.
The two-dimensional probability density function is
calculated from a scatter table presented by Soeding
(2001). Taking the discrete values from the scatter table
for each of the intervals for H
1/3
and T
1
, the integration
of equation [5] can easily be transformed into a summa-
tion of the respective values.
The probability that the actual loading condition leads to
a dangerous situation in the seastate given by H
1/3
and
T
1
then can be written as follows:
) , , (
) (
) , , , ( ) , , , (
1 3 / 1
1 3 / 1 1 3 / 1

T H v p
p
v T H p v T H p
s v
s fail s dang

=

[6]

In this equation, p

() denotes the probability the ship is

traveling at a course of -degrees relative to the domi-
nating wave propagation. It is assumed that p

() is
independent from the actual values of H
1/3
and T
1
. p

()
can be taken from full-scale observations (see Krueger,
Hinrichs, Kluwe and Billerbeck (2006)). Then
p
v
(H
1/3
,T
1
,,v
s
) denotes the probability that the ship is
traveling at a speed of v
s
knots. As p

() is selected
independently from the seastate, p
v
(,v
s
|H
1/3
,T
1
,) is a
conditional probability depending on all four parame-
ters, as not all speeds are physically possible in a spe-
cific situation. Krueger, Hinrichs, Kluwe and Billerbeck
(2006) determine the maximum possible ship speed in
the given environmental conditions at full engine output
and the minimum speed at engine idle speed from sys-
tematic propulsion calculations. Within the range of
possible speeds [v
min
,v
max
] the probability of occurrence
is assumed equally distributed as more accurate data is
lacking.
The failure probability p
fail
(H
1/3
,T
1
,,v
s
) is determined
from the time series of the numerical simulation by
applying the Blume-criterion mentioned above. Given
the loading condition fulfills the Blume-Criterion in the
actual situation, p
fail
(H
1/3
,T
1
,,v
s
) is set to 0, which
means that the loading condition is sufficiently safe for
the given conditions. In case the Blume-criterion fails
for the current situation, p
fail
(H
1/3
,T
1
,,v
s
) is set to 1. This
equation means that decision is taken only between
safe and unsafe by setting the failure probability to
0 or 1, respectively.
All situations in which the failure criterion is set to 1
contribute to the overall long-term probability. Formally
this does not deliver a correct capsizing probability,
which is the reason that the result is called capsizing
index. Yet taking into account the practical considera-
tions, it seems to be more important for us to identify
dangerous situations than to determine the exact failure
rate in a specific situation that is known to be danger-
ous.
Furthermore, it should be noted that our method explic-
itly treats head sea and following sea cases only. There-
fore, we restrict the contributing courses to a 45-degree
sector of encounter angles, port and starboard in head
and following seas. Consequently, it is then useful to
split the ISEI in a head sea and a following sea index.
The ISEI then can be written as follows:
)) ( , ) ( ), ( ) ( (
)) ( (
)) ( ), ( (
)) ( , ) ( ), ( ) ( (
)) ( (
)) ( ), ( (
1 3 / 1
1 1 1
1 3 / 1
1 3 / 1
1 1 1
1 3 / 1
1 3 / 1 , ,
1 3 / 1 , ,
k i T j H l v p
k p
i T j H p
k i T j H l v p
k p
i T j H p
ISEI ISEI ISEI
v
N
i
N
j j
N
k
N
l
sea
v
N
i
N
j j
N
k
N
l
sea
head following
T H
Bl
h h v
T H
Bl
f f v

=
+ =


= = = =
= = = =
[7]

In the formula, the summation on the limiting wave
heights starts at j
Bl
, which is the smallest significant
wave height for the given significant period T
1
where
p
fail
equals 1. The encounter angles run from =-/4 to
=+/4 for the following sea cases and from =3/4 to
=5/4 for head seas. The speed summation runs from
the minimum speed possible in that condition to the
maximum speed possible. The indices h and f indicate
head and following seas, respectively.
For practical applications, it is useful to find those com-
binations of H
1/3
, T
1
, and v
s
, which represent the limit
between safe and unsafe. This solution can be most
efficiently achieved by finding the limiting significant
wave height for a given combination of parameters T
1
,
and v
s
according to the Blume-criterion. In cases where
the Blume-criterion does not deliver suitable results
typically due to large angles of vanishing stability, the
occurrence of a certain maximum roll angle may be
simultaneously taken into account.. The more conserva-
tive value is taken for the decision between safe and
unsafe. The results may be plotted in the form of
polar diagrams as presented in Fig. 2. Each polar dia-
gram presents the limiting wave heights for a specific
significant period (or the related significant deep water
wave length), giving an overview about critical situa-
tions (see Cramer and Krueger (2005) and Krueger
(2002)). Typically the simulations, with a duration of
10000 seconds in real time, are repeated five times, each
with different wave realizations.

Fig. 2: Graphical visualization of dangerous scenarios by
the limiting significant wave height according to the
Blume-criterion
The ISEI-concept allows the identification of ship de-
signs and ship types, which are vulnerable for insuffi-
cient stability events in following or head seas. At this,
the ISEI-concept takes into account all relevant phe-
nomena occurring in head and following seas that may
endanger the vessel with respect to minimum stability.
Unfortunately, there is no limiting value for the ISEI
thus far making it difficult to actually apply the concept
with respect to the determination of minimum stability
requirements. In order to define threshold values for the
ISEI-concept, Krueger and Kluwe (2006) suggested
analyzing the safety levels for a large number of exist-
ing ships by using the ISEI as the quantitative meas-
urement.
The Simplified ISEI
To evaluate a ships safety by applying the ISEI concept
requires expert knowledge with respect to numerical
seakeeping simulations while being relatively time con-
suming. Therefore, it would be very useful to calculate
the capsizing index without the necessity of applying
time domain simulations to obtain p
fail
. Currently, we
are working on a concept, which replaces the simula-
tion-based calculation procedure for p
fail
by a simplified
approximation method, mainly based on lever arm al-
terations between wave crest and wave trough condi-
tions. A first, but in some aspects unsatisfying, approach
is presented in Krueger and Kluwe (2006).
Investigation of Real Capsize Accidents
In order to test and calibrate the newly developed con-
cept, a number of real capsize accidents are currently re-
investigated by TUHH. The criteria described in the
first section of this paper are applied to the operating
condition of the ship at the time of the accident, which
always clearly is an unsafe situation. Secondly the
criteria are used to estimate a loading condition where
the respective vessel can be considered as safe. For
both conditions the ISEI is calculated in order to dem-
onstrate whether the new concept is able to distinguish
between safe and unsafe loading conditions. Three ex-
amples for this work are given below. More cases are
currently being investigated.
The Capsizing of SS Fidamus (1950)
Fig. 3: General Arrangement of SS Fidamus
On J anuary 31, 1950, the 743 BRT vessel SS Fidamus
capsized in heavy weather bound from Wismar to Ant-
werp close to Langeoog, 54 N, 7 E. The vessel was
loaded with ca. 900t potash (angle of repose ca. 35
Deg.). The vessel was traveling in following seas of ca.
40 m significant wave-length, H
1/3
was ca. 2.0 m and the
vessels speed was ca. 9.5 knots. The vessel suddenly
heeled to more than 30 degrees and remained there with
a steady list of ca. 35 to 40 degrees. Water ingress then
lead to capsizing within 10 minutes. The information
given above was taken from the final report issued on
1950-06-27 by the maritime casualty investigation
board of Bremerhaven (Seeamt Bremerhaven (1950)).
The floating condition prior to the accident could be
reconstructed approximately as follows: The ship had a
total displacement of ca. 1541 tons, resulting in a draft
of 4.69m at the aft perpendicular (a.p.). The trim was
1.12m by stern. Interestingly enough the ship did not
carry any ballast water, although this was strongly rec-
ommended in the stability booklet. The resulting right-
ing levers are shown in Fig. 4. The initial GM in still
water conditions amounts ca. 0.30 m. Based on these
lever arm curves we can conclude that without any
external heeling moment, the vessel would immediately
heel to about 30 degrees if it stays long enough on the
wave crest.
Fig. 5: Simulated time series of SS Fidamus including
entrapped water (blue: starboard side, black: port side).
The red curve shows the roll angle (positive starboard)
During the original investigations it was supposed that
the vessel suffered from insufficient stability. It is of
further importance that in an expertise made on behalf
of Seeamt Bremerhaven, Kempf (1950) concluded that
the vessel was traveling in a 1:1 following sea reso-
nance, where the rolling period of the vessel (for small
angles) was determined as 11.8 s by Kempf. The en-
counter period was determined to 11.1s. It was con-
cluded that the low stability of the vessel, further re-
duced in the crest position, resulted in the large heeling
angle. Another supposition was that entrapped water
between the hatch, forecastle and bulwark had lead to a
heeling moment, which in turn, perhaps together with a
cargo shift resulted in an intermediate equilibrium float-
ing condition at about 35- 40 degrees coinciding with a
local minimum of the static still water righting lever
curve. Water ingress in the forecastle was made respon-
sible that the vessel finally capsized during the original
investigations.

Fig. 6: Polar diagram showing limiting wave heights with
respect to a maximum roll angle of 40 degrees. Significant
wave length is 40 m
Fig. 4: Lever arm curves for SS Fidamus
(cyan: still water, green: wave trough, red: wave crest)
The numerical simulations carried out with our time-
domain code ROLLS (see Kroeger (1987) and Petey
(1988)), show that the vessel is permanently rolling with
a maximum of ca. 45 degrees. As the static angle of
vanishing stability in still-water conditions is beyond 90
degrees, it is theoretically not possible to capsize the
vessel without any additional heeling moment. Our
investigations show that for the dynamically rolling
vessel, additional water ingress is not necessary for the
final capsizing, but that the entrapped water between the
hatchway coaming and the bulwark produces a suffi-
cient heeling moment as the time series in Fig. 5 clearly
demonstrates.. The following conclusions can be drawn
from the time domain simulations:
Fig. 7: General Arrangement of MV Lohengrin (taken
from Die deutsche Handelsflotte (1963))
For nearly all situations the ship is traveling in follow-
ing seas large heeling angles beyond 40 can be ob-
served, as shown in Fig. 6. It can be shown that a rela-
tively small additional heeling moment due to water
between the hatchway coaming and bulwark is suffi-
cient to cause the final capsizing of the vessel. Once a
list of ca. 40 has been reached, the roll motion oscil-
lates around this heeling angle while the amount of
water trapped is reducing very slowly. The vessel cap-
sizes as reported by the surviving crewmembers if the
vessel stays a sufficiently long time in this situation. In
some of the simulated situations, the water-outflow was
sufficiently fast enough to prevent the vessel from cap-
sizing. The vessel then returns to the upright position.
This exact behavior was reported from a voyage before
the accident by surviving crewmembers.
By comparing the results for different intact stability
criteria as presented in Table 1, it can clearly be seen
that all criteria considered the case where the vessel did
actually capsize as dangerous, whereas all criteria con-
sider the 0.50m GM case as determined by the Kastner-
Roden- criterion as clearly safe. Additionally, it can be
stated that a direct ISEI of 0.2 represents a condition,
clearly proven to be unsafe. On the other hand, an ISEI
of 0.0008 represents a condition considered to be safe
by all other criteria.
The Capsizing of MV Lohengrin (1963)
On J anuary 14th, 1950 the 955 BRT vessel MV Lohen-
grin capsized in heavy weather bound from Igge-
sund/Sweden to Kiel. The vessel was loaded with ca.
1195 t cellulose in bales. Half a year before the acci-
dent, the vessel was converted resulting in higher hatch
coamings and an increased VCG of the cargo hold vol-
ume. Based on the information obtained from the
Seeamt Flensburg (1964) the most probable floating
condition at the time of the accident shows a draft of
4.69m at a.p. with a total displacement of ca. 2000 tons
and an initial GM of 0.13 m. The resulting lever arm
curves are shown in Fig. 8. On the day of the accident
the vessel entered the Kiel Fjord at about 14.00 hrs. The
waves encountered the vessel from abaft. Significant
wave period was ca. 6 seconds, the wave height was
reported to be about 2 meters. At about 14.15 hours the
vessel heeled to 40-45 degree starboard side and re-
mained there with a steady list of the same size. The
ship finally capsized and sank ca. 1.5 hours later.
The dynamic analysis by means of numerical simula-
tions clearly shows that the reason for the loss of MV
Lohengrin can be consistently explained by insufficient
stability and a pure loss on the crest situation. The ac-
tual stability at the time of the accident was even below
the recommended Rahola-criteria. As the deckhouse
was weathertight, the vessel could find an intermediate
equilibrium while resting on the superstructures. Thanks
Table 1: Results for different criteria in safe and un-
safe loading condition for SS Fidamus
Criterion GM=0.30 m GM=0.50 m
Kastner/Roden
Capsizing time [s] 487 2386618
Soeding
Capsize Probability 0.25/Roll Cycle 0.1 - 0.7E-6 /year
Blume (Modified)
E_R - 3 S E_R =S =0 1.772 mmRad
ISEI (direct) 0.20078 0.0008
Empirical Criteria
Crest lever <0.05 >0.05
Crest range <16 Deg >16 Deg.
Blume c- factor none fulfilled all fulfilled
Fig. 8: Lever arm curves for MV Lohengrin
(cyan: Stillwater, green: wave trough, red: wave crest)

Fig. 9: Polar diagram showing limiting wave heights with
respect to the Blume-criterion. Significant wave length is
60 m
to this fact, most of the crew could be saved. This in-
termediate equilibrium is possible only if an additional
heeling moment acts on the ship, for example due to
cargo shift. Our investigations show that a shift of TCG
of the cargo by 4cm would be enough to keep the vessel
in the listed position. In general, in the given loading
condition the vessel was theoretically unsafe in all fol-
lowing sea situations.
Again, comparing all criteria, the results are unique. All
criteria consider the case where the vessel did actually
capsize as dangerous. The GM value originally pre-
dicted by the Kastner-Roden-criterion for this case
seems to be too low. For the GM-value of 0.305m,
which was selected from the simulations in artificially
amplified waves, all criteria consider this case as safe.
Additionally, it can be stated that a direct ISEI of 0.189
represents a condition, which has clearly proven to be
unsafe, whereas an ISEI of 0.002 represents a condition,
which is considered to be safe by all criteria.
The Capsizing of SS Irene Oldendorff (1951)
Fig. 10: Side view of SS Irene Oldendorff
On December 31, 1951, the vessel SS Irene Oldendorff
capsized in heavy weather bound from Emden to Ystad
in the Hubert Gat. The vessel was carrying ca. 2750 tons
of coke, of which ca. 440 t has been carried on deck.
The wreck was found later at the position 53 38 27
N and 6 17 10 E. The last known position of the
vessel was close to buoy J /E 1 in the Hubert Gat. Ac-
cording to our findings based on the data of the Seeamt
Bremerhaven (1952) the vessel had a draft of 5.4 m a.p.,
trimmed 0.108 m by head at the beginning of the voy-
age (Emden Lock). This amount equals a displacement
of 4575t.
The simulation-based analysis shows that the reason for
the loss of SS Irene Oldendorff can be consistently
explained as an intact stability accident due to the loss
of stability in extreme weather conditions. The loss of
stability in this particular situation can be clearly related
to water entrapped in the coke deck cargo, which could
not escape fast enough through the freeing ports. The
simulations have shown that even if the superstructures
of the vessel are regarded as weather-tight, the stability
vanishes completely on the crest for waves of 80m in
length and 5m in height. Dynamically, the vessel easily
reaches heeling angles beyond 40 degrees. These steep
waves, however, are quite extreme. Taking all of this
into account the capsizing sequence very likely was as
follows: Due to the low stability in wave crest condi-
tions a large heeling angle must have occurred, shifting
the vessel to an intermediate equilibrium at ca. 45.
Water ingress, cargo shift or both must then have lead to
the final capsize.
The simulations have shown that capsizing is hardly
possible for the stability the ship had when she left Em-
den lock. This fact is clearly stated in the polar-diagram
(Fig. 11, left side) for the Emden Lock situation where
significant wave heights of ca. 7-8m are required to
endanger the vessel. These extreme (significant) wave
heights are hardly possible in the Hubert Gat. Assuming
an amount of 90 tons of water was entrapped in the
coke, the polar-diagram changes significantly (Fig. 11,
right side). Now, the limiting wave height is shifted to a
much lower value of about 4-5 meters for the operating
condition of the vessel. All calculations are based on the
assumption that the deckhouse can be considered as at
least temporarily weather-tight. Otherwise, it seems
impossible that the crew could have manned the rescue
boats.
Table 2: Results for different criteria in safe and un-
safe loading condition for MV Lohengrin
Criterion GM=0.131 m GM=0.305 m
Kastner/Roden
Capsizing time [s] 6024 -
Soeding
Capsize Probability - 0.4E-7/year
Blume (Modified)
E_R - 3 S -3.509 mmRad 51.9 mmRad
ISEI (direct) 0.18911 0.002
Empirical Criteria
Crest lever <0.05 >0.05
Crest range <16 Deg >16 Deg.
Blume c- factor none fulfilled all fulfilled

Fig. 11: Polar Diagram showing limiting wave heights
according to
Table 3 shows that all investigated criteria consider the
case as dangerous, where the vessel did actually capsize.
On the other hand all criteria consider the 0.39m-GM
case as clearly safe. Therefore, the direct ISEI of 0.158
represents a condition, which has clearly proven to be
unsafe. An ISEI of 0.0011 represents a condition, which
is considered to be safe by all criteria. All criteria except
the C-factor and the Blume-criterion do not consider the
situation at Emden lock as sufficiently safe where GM
was ca. 0.20 m. This condition is related to an ISEI of
0.0139. The C-factor concept may suffer from the fact
that the vessel is actually a Shelter Decker, which has
large freeboard and may not be covered by the concept.
All these considerations for the Emden lock condition
must lead to the conclusion that the vessel did not have
enough maximum righting lever. Without the deckhouse
being weather-tight, the maximum righting lever occurs
at 30 degree and amounts ca. 0.2 m.
Although the large freeboard results in a significant
range of positive righting levers, this does not lead to
sufficient stability as in the given situation large heeling
moments acted upon the ship. They can only be covered
by a sufficiently large maximum righting lever and not
by range alone. Taking all findings into account, the SS
Irene Oldendorff -accident can be regarded as a case,
which did occur close to the limit that distinguishes a
ship from being safe or unsafe.
Conclusions
In the recent years a large number of ships was investi-
gated with respect to their dynamic behavior in waves
by means of numerical simulations in the time domain.
Based on this database, a new intact stability concept
was developed called Insufficient Stability Event Index
(ISEI).
The new concept is based on long-term probabilities
taking into account the probability of occurrence for
sea-state, course and ships speed. The actual failure
criterion for the ship in a specific operating condition is
implemented via a safe/unsafe-decision based on
the Blume-criterion and the maximum roll angle ob-
served during the simulation.
Additionally, a number of intact stability accidents, all
leading to the total loss of the vessel, were investigated
to get a more precise picture of the complex failure
modes leading to the capsizing of a ship.
A set of intact stability criteria, presented at the begin-
ning of this paper, was then applied to the respective
loading condition of the ship. Based on these criteria a
safe loading condition was estimated and subse-
quently the newly developed ISEI was calculated for
both conditions. The results show that the values ob-
tained for the ISEI differ significantly between the two
calculated loading conditions. This observation indi-
cates that the new concept is able to give clear advice as
to whether the ship is safe or unsafe in a specific condi-
tion. Based on this work it will be possible in the near
future to define threshold values for the ISEI.
References
Arndt, B. (1960). Ermittlung von Mindestwerten fr
die Stabilitt, Schiffstechnik pp.35.
Blume, P (1987). Development of New Stability Crite-
ria for Modern Dry Cargo Vessels, Proceedings
PRADS87 Conference
Blume, P. and Hattendorf, H. G. (1987a). Ergebnisse
von systematischen Modellversuchen zur Kentersi-
cherheit, J ahrbuch der Schiffbautechnischen Ge-
sellschaft, p 219
Blume, P. and Hattendorff, H. G. (1987b). Stabilitaet
und Kentersicherheit moderner Handelsschiffe,.
HSVA Report S 165/83 and S 184/84
Cramer, H. and Krueger S. (2005). Numerical capsi-
zing simulations and consequences for ship design,
J ahrbuch der Schiffbautechnischen Gesellschaft
2001, Springer, 2005
Die deutsche Handelsflotte (1962). Plne der Fracht-
schiffe ber 300 BRT Seehafen-Verlag
Table 3: Results for different criteria in safe and un-
safe loading condition for SS Irene Oldendorff
Criterion GM=0.14 m GM=0.39 m
Kastner/Roden
Capsizing time [s] 13800 No capsize found
Soeding
Capsize Probability -
0.3E-6 for h>9m
none for h<9m
Blume (Modified)
E_R - 3 S -64.0 mmRad 181.5 mmRad
ISEI (direct) 0.158 0.0011
Empirical Criteria
Crest lever 0.008 <0.05 >0.05
Crest range 10 <16 Deg >16 Deg.
Blume c- factor none fulfilled all fulfilled
Kastner, S. (1962). Kenterversuche mit einem Modell
im natrlichen Seegang, Schiffstechnik 1962 p. 161
Kempf, G (1950). Gutachten ber die Ursachen, die
zum Untergang des Frachtdampfers FIDAMUS ge-
fhrt haben mgen. Staatsarchiv Bremen
Kroeger, P. (1987). Simulation der Rollbewegung von
Schiffen, Bericht, Nr.: 473, Institut fr Schiffbau
der Universitt Hamburg
Krueger, S. (2002). Performance Based Stability,
DCAMM 2002, Lyngby
Krueger, S. and Kluwe, F. (2006). Development of
Dynamic Stability Criteria from Seakeeping Simula-
tion., International Marine Design Conference
(IMDC2006), Ann Arbor, Michigan, USA
Krueger, S. , Hinrichs, R., Kluwe F. and Billerbeck, H.
(2006). Towards the Development of Dynamic Sta-
bility Criteria HANSA Maritime J ournal, 10/2006,
p.204
Germanischer Lloyd (1967). Ergebnisniederschrift
Stabilitt MS TRIER v. 8. 5.1967, Germanischer
Lloyd /SeeBG 1967.
Petey, F (1988). Ermittlung der Kentersicherheit lecker
Schiffe im Seegang aus Bewegungssimulationen,
Bericht Nr.: 487, Institut fr Schiffbau der Universi-
tt Hamburg
Seeamt Bremerhaven (1950). In seeamtlichen Untersu-
chungen ber den Untergang des Frachtdampfers
FIDAMUS. Staatsarchiv Bremen, 1950.
Seeamt Bremerhaven (1952). Dampfer Irene Olden-
dorff aus Luebeck. UNTERGANG. SEEAMT
BREMERHAVEN. 4. J UNI 1952., Staatsarchiv
Bremen
Seeamt Flensburg (1964). Frachtmotorschiff LO-
HENGRIN aus Bremen, KENTERUNFALL,
Seeamt Flensburg, 22. J anuar 1964, aus: Landesar-
chiv Schleswig- Holstein.
Soeding, H. (2001). Global Seaway Statistics, TU
Hamburg-Harburg, Schriftenreihe Schiffbau, Report
Nr. 610
Soeding, H. and Tonguc, E. (1986). Computing capsiz-
ing frequencies in a seaway, Proc. of Stab.,
Gdansk, Vol II. Add 1, p. 52
Wagner, D. (1987). Kentersicherheit intakter Schiffe,
23. Fortbildungskurs, Inst. f. Schiffbau Univ. Ham-
burg
Wendel, K. (1954): Stabilitaetseinbussen im Seegang
und durch Koksdeckslast., HANSA 1954, P. 2009.