See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/341107979

Reliability analysis of GM and KG of ships from the inclining experiment

Article in Proceedings of the Institution of Mechanical Engineers Part M Journal of Engineering for the Maritime Environment · May 2020
DOI: 10.1177/1475090220913709

CITATIONS READS

3 1,911

1 author:

Metin Taylan
Istanbul Technical University
28 PUBLICATIONS 328 CITATIONS

SEE PROFILE

All content following this page was uploaded by Metin Taylan on 15 September 2020.

The user has requested enhancement of the downloaded file.

Original Article

Proc IMechE Part M:

J Engineering for the Maritime Environment
1–7
Reliability analysis of GM and Ó IMechE 2020
Article reuse guidelines:
KG of ships from the inclining sagepub.com/journals-permissions
DOI: 10.1177/1475090220913709

experiment journals.sagepub.com/home/pim

Metin Taylan

Abstract
It is of paramount importance to determine the lightship characteristics of a newly built vessel correctly. Conventionally,
an inclining experiment is performed after the ship is launched in order to determine the lightship GM and in turn KG
values. However, there has been an ongoing debate about the validity of the results since it bears some assumptions dur-
ing the test and calculation procedure. Although the inclining experiment is mandatory by the Intact Stability Code of
International Maritime Organization, there emerged various alternative methods recently claiming to reveal better
results. In this study, an uncertainty analysis has been carried out on GM and KG values of various sample vessels. Based
on the results on the uncertainty analysis, a cumulative margin of error from various sources has been determined within
a specified confidence level. Finally, the outcome for the sample vessels was detailed and evaluated in reference to that
of the actual inclining experiment reports.

Keywords
Inclining experiment, lightship characteristics, uncertainty analysis, GM

Date received: 27 September 2019; accepted: 11 February 2020

Introduction in GM and KG. For this reason, the inclining experi-

ment has drawn some mistrust and subjected to fierce
One of the most important design aspects of ships is discussions from the beginning.
stability and in turn safety in all feasible loading condi- In recent years, some research has been done on
tions. Overall, the stability of a ship is comprised of alternative methods aiming to discard the ambiguities
form and weight components. The stability of weight, of the inclining test. These alternative methods may be
however, is mainly governed by the lightship character- listed as the graphical method proposed by Kanifolskyi
istics. Therefore, determination of the lightship charac- and Konotopets,4 the generalized method proposed by
teristics—namely, lightship weight and center of gravity Dunworth,5 and the polar method proposed by
(especially, the vertical one (KG))—is extremely crucial Karolius and Vassalos.6 As a common ground, none of
since some vessels may have small stability margins.
these alternative methods rely on the fixed metacenter.
The origin of this method dates back to 17th
Almost every experiment contains some sort of error
century.1,2 In order to determine the lightship charac-
stemming from various sources such as calibration of
teristics of ships, a conventional inclining experiment—
the equipment used, readings, and measured input
which is mandated by the International Maritime
data. Uncertainty analysis is often applied to ensure
Organization (IMO)’s 2008 Intact Stability (IS) Code
the validity of the final results.
for every commercial ship above 24 m in length and all
passenger ships—is carried out.3 The main principle of
this experiment relies on shifting of certain weights
across the deck and recording ensuing inclinations. Naval Architecture and Marine Engineering Department, Istanbul
Inherently, this test is based on certain assumptions Technical University, Istanbul, Turkey
such as the emerged and submerged volumes are equal
Corresponding author:
at a heel angle, unchanged metacenter, and wall- Metin Taylan, Naval Architecture and Marine Engineering Department,
sidedness of the ship. Along with these assumptions, a Istanbul Technical University, Istanbul 34469, Turkey.
number of sources of error may contribute to deviations Email: taylan@itu.edu.tr
2 Proc IMechE Part M: J Engineering for the Maritime Environment 00(0)

This study aims at conducting an uncertainty analy- Since the inclining test yields GM of a ship directly,
sis on 10 sample ships whose inclining test reports are vertical position of center of gravity can be calculated
readily available. These uncertainties are investigated from the difference, as shown in equation (1)
on pertinent parameters (draft, volume, displacement,
tangent line, etc.) which have direct impact on GM and KG = KM  GM ð1Þ
KG. In the above equation, metacentric height (KM) is
the representative of form and is obtained from the
hydrostatics. When a test weight is moved to a distance
The inclining experiment d transversely, the vertical center of gravity shifts from
The inclining experiment is mandatory requirement of G to G1 (Figure 2). GM may be determined graphically
IMO by the 2008 IS Code for every commercial ship from the triangle easily
above 24 m in length and all passenger ships. The
GG1
details of the test procedure are outlined in the code. GM = ð2Þ
Ships undergo this test following the launch at 80%– tanu
90% completion ratio. The experiment is usually car- The above equation can also be represented in terms
ried out shifting solid weights across the deck and read- of heeling moment as
ing the deflections by pendulums or inclinometers.
During the test, the ship should float freely under no wd
GM = ð3Þ
external force or moment and ideally with no trim and D  tanu
heel. Favorable environmental conditions, no or little
wind and waves, are preferred at the time of the test. Tangent of the heel angles is determined from the
The weights should be placed symmetrically on the pendulum readings as follows
deck about amidships yielding small inclinations port Pendulum deflection
and starboard. tanu = ð4Þ
Length of plumb line
Eight weight shifts should be conducted as a stan-
dard required by the 2008 IS Code. A sample sequence Repetition of the weight shifts ensures an average
of the shifts is shown in Figure 1. tanu and therefore minimizes the error from pendulum

Figure 1. Weight shift sequence.

Taylan 3

assumptions may be the so-called ‘‘theoretical.’’ It is a

known fact that the waterplane area of ships having
change in knuckles, flares, and sharp chines consider-
ably affect the results.
Besides the above-mentioned issues, environmental
factors such as wind, waves, and currents play an
important role during the test. Free surface effect due
to slack tanks, suspended weights onboard and so on
may also contribute to possible error along the line.
Finally, human factor comes into play during the pro-
cedure while reading the drafts and pendulum
deflections.

Uncertainty analysis
Figure 2. Inclined ship by test weights.
In general terms, uncertainty may be defined as a quan-
tification of the doubt about the measurement results
in engineering experiments. In other words, it defines a
possible value that an error may contain.
Intrinsically, inclining experiment bears certain
uncertainties in the final outcome in both KG and
GM. Draft readings, pendulum deflections, hydrostatic
parameters, density readings, displacement, and vol-
ume are to be considered as the leading contributors to
the final result. Therefore, it is imperative to follow the
test procedure strictly, especially at key steps to avoid
possible errors before proceeding to the uncertainty
analysis. Calibration of measuring devices may also
help to eliminate some of the doubts over the readings
and/or measurements.
In this section, sources of uncertainties such as draft
reading, displacement, heel angle, wind, and wave are
examined. The effect of these uncertainties has been
applied to all sample ships considered in the study.
Uncertainties in inclining experiment have been investi-
gated by various researchers such as Wilczynski et al.,7
Figure 3. Inclination–moment graph. Woodward et al.,8 Sales et al.,9 and Shakshober and
Montgomery.10 They commonly examined the uncer-
tainty in GM and KG which result from the bias and
readings. This situation is depicted as a moment–tanu accuracy deviations related to the measured and calcu-
graph encompassing all weight shifts and a linear line lated variables one way or the other.
passing through these points according to the least In general, uncertainty in a result may be explained
squares method. The slope of the linear line reveals as follows. Consider the function R as
GM (Figure 3)
R = RðX1 , X2 , X3 , :::, Xn Þ ð6Þ
1
GM = ð5Þ The uncertainty in the result UR may be given by a
D  Rslope
root sum square equation which is derived from a
Taylor series expansion
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2ffi
Shortcomings of the inclining experiment 
∂R
2 
∂R
2 
∂R
UR = UX + UX +  + Un
The inclining experiment has been drawn some criticism ∂X1 1 ∂X2 2 ∂Xn
by many researchers and interested parties. The under- ð7Þ
lying reasons behind this criticism are emanating from
certain assumptions made in the procedure. First of all, For presentation simplicity, the relative uncertainties
for small angles of inclination (2°–4°), the metacenter is are given in the shorthanded form
assumed to be unchanged. The vessel is presumed to be    
wall-sided, and the submerged and emerged volumes ^ UR ^ UXi
UR [ and UXi [ ð8Þ
are equal when heeled to a side and vice versa. These R Xi
4 Proc IMechE Part M: J Engineering for the Maritime Environment 00(0)

The notation has been kept the same throughout the where the subscript 1 refers to uncertainty due to con-
article when calculating uncertainties for various para- struction tolerances, and the subscript 2 refers to uncer-
meters. The representation of any uncertainty must tainty due to hull deflection and distortion during
include the confidence level of that specific uncertainty. construction.
These confidence levels come either from the measure- Finally, determining the uncertainty in lightship KG
ment statistics or from experience of past experiments. requires a little more rigorous work as summarized
For or a C% confidence that the real value of R lies below
within the interval 6UR from the calculated or esti- qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
mated value of R as follows U^KGL = U ^2 + U ^2 ð17Þ
DL ML

Rreal = Restimate 6UR ðC%Þ ð9Þ where

It is common to use the 95% confidence level. qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
However, it does not change the final result since UR ^ML =
U U^2 + U ^2 + U ^2 + U ^2 + U ^2 ð18Þ
MI Msa Msr Mla Mlr
can be converted from one confidence level to another
level. and
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
^D =
U U^2 + U ^2 + U ^2 ð19Þ
L DI ws wl
Methodology and calculation procedure
In order to calculate the uncertainty in GM, first, the where
uncertainty in the slope of the best fit linear line needs
to be determined. The uncertainty in GM (UGM) is then ML = MI + Msa  Msr + Mla + Mlr ð20Þ
calculated using the expression below found using the The definition of the terms appears in the above
uncertainty of displacement (UD) and slope (Uslope)11 equations
  s ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2  2ffi
UGM Uslope UD ML = KGL  DL
= + ð10Þ
GM Slope D MI = KGI  DI
X
Or for simplicity, the relative uncertainty of GM is Msa = wsa  kgsa ðsolid weights to be addedÞ
X
given by Msr = wsr  kgsr ðsolid weights to be removedÞ
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi X
U^GM = U ^2 + U ^2 ð11Þ Mla = wla  kgla ðliquid weights to be addedÞ
slope D X
Mlr = wlr  kglr ðliquid weights to be removedÞ
Similarly, the relative uncertainty in inclined KG
may be calculated as below since KG is composed of ð21Þ
various hydrostatic parameters The subscript L refers to ‘‘lightship,’’ whereas the sub-
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi script I refers to ‘‘as inclined’’ conditions.
^KG = U
U ^2 + U ^2 + U ^2 ð12Þ The list of these parameters utilized in the calcula-
KB BM GM
tion procedure is given below. Numerical results of the
The uncertainty of BM is uncertainty procedure applied to all 10 sample ships are
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi supplied in the following sections. Sources of uncertain-
^BM = U
U ^2 + U ^2 ð13Þ ties are follows: inclining weights, distance of weight
I r
shift, moment of AWP, deflection, draft, displacement
The uncertainty of KB is volume, density, heel angle (tanu), added/removed
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi weights, KB, BM, GM, and KG.
^KB = U
U ^2 + U ^2 + U ^2 ð14Þ
A D r

where UA is the uncertainty in the waterplane area as Sample calculation for lightship KG
follows of asphalt tanker 1
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
^A = U ^2 + U ^2
In this section, a sample calculation of the relative
U B L ð15Þ uncertainty of lightship KG for asphalt tanker 1 is sup-
plied as an example. The rest of the calculations have
The relative uncertainties of length, beam, and depth
been carried out in the same manner for all parameters,
related to construction tolerances in addition to deflec-
and all ship types according to the procedure are out-
tion and distortion may be given as
lined in section ‘‘Methodology and calculation proce-
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
^L = U ^2 + U ^2 , U ^B = U ^2 + U ^2 , dure.’’ Unfortunately, it is not possible to supply all
U L1 L2 B1 B2 details of the entire calculation scheme due to space
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
^D = U ^2 + U ^2 limitations. The readers are referred to Tables 1 and 2
and U D1 D2 ð16Þ for specific results.
Taylan 5

Table 1. Parameters used ML calculations. Table 2. Calculation of relative uncertainty for lightship KG.

Parameters Values Units Parameters Values Parameters Values Units

wsa 50.657 ton Ûws 0.0400

wsr 38.5876 ton Msa 499.934 Ûkgsa 0.0309
kgsa 9.869 m Msr 353.848 Ûkgsr 0.0332
kgsr 9.17 m Mlr 5979.285 Ûwl 0.0200
wlr 1093.305 ton Mı 42,401.55 Ûkgl 0.0223
kglr 5.469 m Ûmsa 0.0505
Ûmsr 0.0520
Ûlr 0.0299
Umsa 25.2645 ton m
Sample ships Umsr 18.4029 ton m
Ulr 179.0768 ton m
There are 10 sample ships of various types and sizes UDı 112.6933
whose inclining test results are already available are uti- Umı 891.9035 ton m
lized in this study. The spectrum of vessels includes dif- UML 910.2402 ton m
ML 36,569.2934 ton m
ferent types and forms—from fuller to slender, from UDL 114.8506 ton
big to small, from round to chine, and knuckle ones ÛDL 0.0266
for a meaningful comparison of different parameters. ÛML 0.0249
The characteristics of the sample ships are supplied in UKGL 0.3083
Table 3.

Uncertainties in the sample ships GM and KG are also depicted graphically in Figures 4
and 5, respectively.
To this end, the aforementioned procedure to calculate
the uncertainties numerically has been applied to the
sample ships whose characteristics are given in Table 3.
Discussion
The results from the uncertainty analyses are tabulated
in Tables 4 and 5. Uncertainty analysis reveals a possible margin of error
The summary of the uncertainty analysis for the from the inclining experiment. Table 6 depicts what is
sample ships is tabulated in Table 6. Uncertainties in so-called the margin of error in GM and KG for the

Table 3. Sample ships.12

Sample ships LBP (m) B (m) T (m) D (m) D (tons) CB AWP (m2) KG (m)

Chemical tanker 1 102.18 18 6.8 9.2 10,021 0.774 1458 7.449

Chemical tanker 2 123.99 18.9 7.98 10.2 14,834 0.770 1709.7 8.466
Asphalt tanker 1 116.8 19.4 7.35 10.8 14,241 0.740 1649.8 1.663
Asphalt tanker 2 147.9 25 9.25 14.6 27,960 0.800 2915.7 8.501
Service boat 14.225 4.8 1.275 2.31 51.09 0.517 58.375 8.153
Tugboat 17.55 8.4 3.1 3.25 235.95 0.520 88.45 3.425
Roll-on/roll-off ship 120 19.4 4.88 7.30 6996.6 0.600 1902.2 9.413
Research vessel 56.45 8.8 3.95 5.11 1167.6 0.650 316.8 4.031
Fast boat 13.06 4.04 0.765 2.09 17.355 0.662 43.96 1.294
Container ship 134.47 22.6 3.576 11.3 17,887 0.620 2135.7 9.926

Table 4. Results of measured/calculated values of sample ships.

Chemical Chemical Asphalt Asphalt Service Tugboat Roll-on/ Research Fast Container
tanker 1 tanker 2 tanker 1 tanker 2 boat roll-off ship vessel boat ship

Slope (1/ton m) 5.8 3 1026 16,885 12,431 76,536 55 398 0.855 266 14 16,091
Draft, T (m) 6.800 7.984 7.350 9.250 1.162 3.10 4.879 3.948 0.765 8.574
Displacement volume, V (m3) 4074 5573 5305 10,054 42 194 6826 1144 17 6467
Density, r (ton/m3) 1.017 1.017 1.018 1.019 1.019 1.022 1.017 1.021 1.019 1.020
Displacement, D (tons) 4143 5668 5401 10,245 43.1 199 6942 1168 17 6596
As inclined GM (m) 4.206 2.971 2.300 7.471 1.266 1.998 1.256 0.445 0.828 2.440
As inclined KG (m) 5.871 6.809 7.851 7.602 1.697 2.624 9.026 3.529 1.387 9.129
Lightship KG (m) 7.449 8.153 8.466 8.501 1.663 3.425 9.413 4.031 14.268 9.926
6 Proc IMechE Part M: J Engineering for the Maritime Environment 00(0)

Table 5. Results of uncertainty values of sample ships.

Chemical Chemical Asphalt Asphalt Service Tugboat Roll-on/ Research Fast Container
tanker 1 tanker 2 tanker 1 tanker 2 boat roll-off ship vessel boat ship

Slope (1/ton m) 3.4 3 1028 0.101 0.455 0.388 0.507 0.705 0.003 0.316 0.158 0.793
Draft, T (m) 0.019 0.019 0.019 0.019 0.019 0.019 0.019 0.019 0.019 0.019
Displacement volume, V (m3) 85.83 116.3 110.4 208.2 1.372 4.216 141 23.6 0.882 135.2
Density, r (ton/m3) 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002 0.002
Displacement, ã (tons) 87.507 118.6 112.7 212.7 1.400 4.319 143.8 24.17 0.899 138.3
As inclined GM (m) 0.089 0.062 0.048 0.155 0.043 0.044 0.026 0.009 0.044 0.051
As inclined KG (m) 0.204 0.028 0.024 0.316 0.101 0.102 0.174 0.076 0.111 0.218
Lightship KG (m) 0.471 0.344 0.308 0.311 0.090 0.110 0.288 0.183 0.140 0.349

All uncertainties are at the 95% confidence level.

Table 6. Comparison of uncertainty analyses for sample

ships.

Sample ships GM (m) Error (6m) KG (m) Error (6m)

Chemical tanker 1 4.206 0.09 7.449 0.47

Chemical tanker 2 2.971 0.06 8.153 0.34
Asphalt tanker 1 2.30 0.05 8.466 0.31
Asphalt tanker 2 7.441 0.16 8.501 0.31
Service boat 1.266 0.04 1.663 0.09
Tug boat 1.998 0.04 3.425 0.11
Roll-on/roll-off 1.256 0.03 9.413 0.29
Research vessel 0.445 0.01 4.031 0.18
Fast boat 0.828 0.04 1.294 0.14
Container ship 2.44 0.05 9.926 0.35

Figure 5. Uncertainties in KG.

The uncertainties depicted in Table 4 are in line with

some of the similar studies in the literature. Wilczynski
et al.7 gave a sample calculation on the uncertainty of
GM of an offshore supply vessel (OSV) having length
44.5 m and GM = 6.593 m from the inclining experiment.
They calculated 60.122 m uncertainty in GM. Woodward
et al.8 analyzed the uncertainties of five ships (with a fixed
reference GM = 0.15 m): buoy tender 0.15 6 0.15 m,
superyacht 0.15 6 0.033 m, supply ship 0.15 6 0.047 m,
containership 0.15 6 0.029 m, and roll-on/roll-off passen-
ger (ROPAX) 0.15 6 0.077 m. Shakshober and
Montgomery10 found 60.1024 m uncertainty in lightship
KG of large naval vessel.
Figure 4. Uncertainties in GM.

Conclusion
sample vessels. In terms of lightship GM, none of the The inclining experiment is the only regulatory mean at
sample vessels have a critical value that may endanger the moment to determine lightship characteristics of
its safety. However, as far as KG values are concerned, ships. However, ongoing criticism about the procedure
there may be certain loading conditions that are critical raises some doubts among the related parties such as
to satisfy the related IMO criteria. Both chemical tan- ship designers, builders, and operators. Therefore,
kers, both asphalt tankers, container ship, and roll-on/ utmost care must be exercised during the test to mini-
roll-off (ro-ro) vessel have higher uncertainty margins mize possible errors. Furthermore, an uncertainty anal-
in KG. This may be important for certain loading con- ysis may also be helpful in order to estimate the margin
ditions when the margin of stability is critical. of safety which depends on GM and KG.
 Taylan 7

In this work, uncertainty analyses of 10 sample ves- www.navcen.uscg.gov/pdf/marcomms/imo/msc_resolu-

sels whose inclining test reports are available have been tions/MSC267.pdf
performed. Most of the vessels have different forms 4. Kanifolskyi OO and Konotopets MM. The graphical
and parameters. Although it is not easy to generalize method for analysis of the inclining test. Mod Inf Technol
the results, slender vessels seem to give lower uncertain- 2016; 3(27): 37–39.
5. Dunworth R. Up against the wall. In: Proceedings of the
ties when compared to the fuller ones both in GM and
Pacific international maritime conference 2013, Sydney,
KG at the 95% confidence level. Asphalt tanker 2 has
NSW, Australia, 7–9 October 2013. Geelong, VIC, Aus-
the highest uncertainty in GM as 60.16 m, whereas tralia: Maritime Australia Limited.
chemical tanker 1 gives the highest uncertainty in KG 6. Karolius K and Vassalos D. Tearing down the wall–the
as 60.47 m. From this elaborate uncertainty analysis, inclining experiment. Ocean Eng 2018; 148: 442–475.
the effect of all the constituent parameters other than 7. Wilczynski V, Knowles PH and Diehl WJ. Uncertainty
GM and KG is easily seen. analysis of a vessel’s metacentric height. In: Proceedings
of the probabilistic safety assessment and management
conference, New York, 1998, https://www.researchgate.-
Declaration of conflicting interests
net/publication/323782027_Uncertainty_Analysis_of_a_
The author(s) declared no potential conflicts of interest Vessel’s_Metacentric_Height
with respect to the research, authorship, and/or publi- 8. Woodward MD, Rijsbergen MV, Hutchinson KW, et al.
cation of this article. Uncertainty analysis procedure for the ship inclining
experiment. Ocean Eng 2016; 114: 79–86.
Funding 9. Sales JS Jr, Esperanc xa PTT, Sphaier SH, et al. Uncer-
tainty analysis for inclining tests. In: Proceedings of the
The author(s) received no financial support for the ASME 2014 33rd international conference on ocean, off-
research, authorship, and/or publication of this article. shore and arctic engineering, San Francisco, CA, 8–13
June 2014. New York: ASME.
ORCID iD 10. Shakshober MC and Montgomery JB. Analysis of the
inclining experiment. In: Proceedings of the meeting of the
Metin Taylan https://orcid.org/0000-0003-1701-8699 Hampton Roads section of the society of naval architects
and marine engineers, February 1967. Alexandria, VA:
References SNAME.
1. Hoste P. The´orie de la construction des vaisseaux (Theory 11. U.S. Department of Transportation. MSC technical
of the construction of vessels). Lyon: Arisson & Posule, review and analyses of the SS EL FARO, Washington,
1697. DC, 2016, https://media.defense.gov/2017/Dec/28/200186
2. Bouguer P. Traite´ du navire, de sa construction et de ses 1774/-1/-1/0/MSC%20TECHNICAL%20REPORT%20
mouvemens (Treatise of the ship, its construction and its ON%20SS%20EL%20FARO%20STRUCTURES%20
movements). Paris: Jombert, 1746. AND%20STABILITY.PDF
3. International Maritime Organization. Part B Annex I of 12. Ozsayan S and Taylan M. Another blow on the torn
the international code on intact stability 2008, as adopted down wall–the inclining experiment. Brodogradnja 2019;
in IMO Res. MSC.267(85), pp.68–86, 2008, https:// 70(2): 136–153.

View publication stats