©Marine Technology, Vol. 39, No. 3, July 2002, pp.

137-158
Marine
Technology

Stockholm Agreement-Past, Present, Future (Part 1)

Dracos Vassalos 1 and Apostolos Papanikolaou 2

April 1, 2001 marked the fourth anniversary of the Stockholm Agreement (SA}, a period during which
almost 80% of the roll-on/roll-off (Ro-Ro) fleet in North West Europe have been subjected to calculations,
model testing and numerical simulations in the struggle to meet these demanding new requirements. The
experience gained has been invaluable in understanding better the problem at hand and is being utilized
to shape new developments likely to lead to more meaningful requirements. The North-South divide,
however, continues to cause unrest, particularly at the European level. Efforts to assess the status quo in
North West Europe, and to use the information amassed so far as a means to predict the potential impact
of introducing the SA in the South, led to a dedicated call by the Commission and to a contract being
awarded to two closely collaborating teams, one at the Ship Stability Research Centre of the University of
Strathclyde under the leadership of Professor Vassalos and one at the Ship Design Laboratory of the
National Technical University of Athens, under the leadership of Professor Papanikolaou, representing the
North and South of Europe, respectively. This background provided the incentive for an introspective look
at the SA, with a view to ascertaining its status before embarking into future projections. This forms Part
1 of the SA related research with Part 2 aiming to cover the results of the Commission study itself.

1. Introduction However, since the great majority of Ro-Ro passenger fer-

THE Ro-Ro CONCEPT provides the capability to carry a wide
variety of cargoes in the same ship, thus being able to offer a
competitive frequency with minimum port infrastructure or
special shore-based equipment. Short sea routes are domi-
nated by Ro-Ro ships with lorries, trailers, train wagons,
containers, trade cars and passengers being transferred from
the "outer" regions (U.K., Ireland, Scandinavia and Finland)
to the "main" land (continental Europe). Also in the Southern
Europe corridors, Ro-Ro freight service is progressively in-
creasing in volume. The case for a long-distance Ro-Ro ser-
vice to provide a European maritime highway has also been
made several times before. This is particularly relevant and
important in respect to fast sea transportation where again
Ro-Ro ferries play a prominent role. The main concern with
the Ro-Ro ship design, whether justified or nor, relates to
safety and with safety becoming of paramount importance, it
is vital that a rational approach to safety is demonstrated,
validated and adopted. This is the right way to ensure both
the survival and a meaningful evolution of Ro-Ro ships in the
future. Along these lines, the maritime industry is acutely
aware of recent shipping casualties involving Ro-Ro ferries,
which have resulted in severe loss oflife. Standards for Ro-Ro
ship configuration, construction and operation have come un-
der close scrutiny and new legislation has been put into place
aimed at improving the safety of these vessels, notably
SOLAS '90 [1] as the new global standard for all existing
ferries with dates of compliance ranging from October 1, 1998
to October 1, 2010, depending on a combination of the vessel's
AJ~a} value [2], the number of persons carried and age.
1
The Ship Stability Research Centre, Department of Ship and
Marine Technology, Glasgow, U.K.
2
National Technical University of Athens, Ship Design Labora-
tory, Greece.
3
The A/Amax calculation procedure, [2] is a simplified version of
the probabilistic damage stability calculation of ships, [3] and was
ries were designed and built prior to the coming-into-force of
SOLAS '90, it is hardly surprising that few of them comply
with the new requirements. Furthermore, concerted action to
address the water-on-deck problem in the wake of the Esto-
nia tragedy led IMO to set up a Panel of Experts (PoE) to
consider the issues carefully and make suitable recommen-
dations. However, the complexity of the problem and the
need to take swift action to reassure the public that appro-
priate steps are taken to avoid a repeat of the Estonia disas-
ter influenced and shaped to a large extent both the initial
and final proposals. In this pace of developments and follow-
ing considerable deliberations and debate, a new require-
ment for damage stability has been agreed among North
West European nations to account for the risk of accumula-
tion of water on the Ro-Ro deck. This new requirement,
known as the Stockholm Agreement [4], ameliorates the origi-
nal proposals by demanding that a vessel satisfies SOLAS '90
requirements (allowing only for minor relaxation) with, in
addition, water on deck by considering a constant height
rather than a constant amount of water as was originally
intended. The dates of compliance with the provisions of the
agreement range from April 1, 1997 to October 1, 2002. How-
ever, in view of the uncertainties in the current state of
knowledge concerning the ability of a vessel to survive dam-
age in a given sea state, an alternative route has also been
allowed which provides a nonprescriptive way of ensuring
compliance, through the Equivalence route, by performing
model experiments in accordance with the Model Test
Method of SOLAS '95 Resolution 14 [5].
Deriving from systematic research over the past twelve
years, numerical simulation models have been developed ca-
pable of predicting with good engineering accuracy the cap-
sizal resistance of a damaged ship, of any type and compart-
adopted by IMO as a means of trying to compare the survivability of
one vessel against ~mother in order to achieve a hierarchy for phas-
ing-in purposes. It is not a survivability standard.

JULY 2002 0025-3316/02/3903-0137$00.69/0 MARINE TECHNOLOGY 137

mentation, in a realistic environment while accounting for
progressive flooding. A comprehensive calibration/validation
program has allowed for sufficient confidence to be built up,
rendering the developed models a valuable design "tool" [6].
This, in turn, offered the ferry industry the attractive possi-
bility of utilizing such "tools" to assess the damage surviv-
ability of ferry safety by using numerical simulation pro-
grams to effectively plan or, in time, replace the model tests,
the so-called Numerical Equivalence route. Numerical simu-
lation readily allows for a systematic identification of the
most cost-effective and survivability-effective solutions to im-
proving ferry safety and hence offers a means for overcoming
the deficiency of the physical model test route in searching
for optimum solutions and an indispensable "tool" for the
planning and undertaking of such tests.
The close involvement of ferry owners/operators in North-
western Europe with research projects in the wake of the
Herald of Free Enterprise and the Estonia accidents was in-
strumental in nurturing industry to firmly accept the Nu-
merical Equivalence route as a viable alternative for assess-
ing Ro-Ro vessel survivability. This afforded SU-SSRC a
unique opportunity to develop in close collaboration with
NTUA-SDL a rational approach to ferry safety with the ca-
pability of attending to the needs of the shipping industry
cost-effectively and led to the establishment of what is
termed a "Total Stability Assessment" (TSA) procedure. The
procedure comprises assessment of a vessel's survivability
utilizing all the currently available instruments, namely:
A.265 (VIII)+ amendments (probabilistic procedure), SOLAS
'90, Stockholm Agreement (prescriptive criteria) and safety
Equivalence tests by means of physical model experiments
and numerical simulations (performance-based criteria). A
schematic illustration is provided in Fig. 1. The tightening of
Calculate Al Arna
sim lified
~------~
SOLAS'9().
Damage Stability Criter
for 2 compartment stan
*)dates: 1.10. 799B10.201,
**) dates: 1. 10. 799B 70.200
---------------------------
SOLAS 90 +SC
(Stockholm Agreemer
Damage Stability Criter
for 2 compartment stanc
Module
Physical•M~ITestl
oftbe.O~dS.
in the Operatmna
M<:Xlel~;'Jrne1'its i - - - - - -
a Suital:ile 1)gf1anf< ____..____
FINAL DECISION
Fig. 1 "Total stability assessment" procedure
138 JULY 2002
legislation described above is coupled with serious consider-
ations at IMO for regular application of risk assessment
methods, for example, the Formal Safety Assessment. In this
context, considerable attention has been focusing on the ap-
plication of probabilistic procedures of damage stability as-
sessment for the evaluation of Ro-Ro vessels and it appears
more than likely that developments in the foreseeable future
will most certainly adopt a framework of a probabilistic de-
scription. The regulatory regime described in the foregoing
has understandably left the shipping industry in a state of
confusion and uncertainty concerning the available options,
approaches and optimum choice to ensure compliance and to
ascertain the level of safety attained with regard to any such
choice. Stated specifically, a shipowner today is faced with
the following choices concerning safety standards:
(i) Deterministic (SOLAS '90) vs probabilistic (A.265
(VIII))
(ii) Prescriptive (SOLAS '90 + 50) vs performance based
(physical model experiments or numerical simulations)
Standards in each group are assumed to ensure an "equiva-
lent" level of safety, correspondingly, while a serious attempt
to demonstrate such equivalence is totally lacking. Adding to
the confusion is the fact that the dates of compliance with
deterministic/prescriptive standards are decided on the basis
of a simplified probabilistic approach (calculation of A/~ax).
In response to the challenge presented by this state of affairs,
the maritime industry, slowly but steadily, appears to be
favoring the model experiments route, implicitly demonstrat-
ing a preference towards performance-based safety stan-
dards over deterministic static stability standards when ad-
dressing the damage survivability of new concept designs.
Not only is the introduction of performance standards a ma-
jor development in assessing safety but it is also seen as
beneficial from the industry as these readily allow consider-
ation of alternative designs as well as a rapid implementa-
tion of technological innovation.
2. Background
Historical overview
Historically, most changes in international regulations for
ship design and operation have been introduced as a result of
major disasters with a large loss of life. The first notable of
such disasters was the sinking of the Titanic, which led a
year later to the first International Convention for the Safety
of Life at Sea in London. The first damage stability require-
ments were introduced, however, following the 1948 SOLAS
Convention and the first specific criterion on residual stabil-
ity standards at the 1960 SOLAS Convention with the re-
quirement for a minimum residual GM (0.05 m). This repre-
sented an attempt to introduce a margin to compensate for
the upsetting environmental forces. "Additionally, in cases
where the Administration considered the range of stability in
the damaged condition to be doubtful, it could request further
investigation to their satisfaction." Although this was a very
Damage Stability
Software Mbdule vague statement, it was the first attempt to legislate on the
SOLAS '90+ WO range of stability in the damaged condition. It is interesting
to mention that a new regulation on "Watertight Integrity
~l'.lsons Above the Margin Line" was also introduced reflecting the
Duck Tail general desire to do all that was reasonably practical to en-
Side·. Ca&inQS
Cros5f(ooding sure survival after severe collision damage by taking all nec-
Bulkheads essary measures to limit the entry and spread of water above
Buoyancy· Device
Curv~ .Decks the bulkhead deck. The first probabilistic damage stability
Reduce Draft
Other
rules for passenger vessels, deriving from the work of Kurt
Wendel on "Subdivision of Ships," [7] were introduced in the
late sixties as an alternative to the deterministic require-
MARINE TECHNOLOGY
ments of SOLAS '60. Subsequently and at about the same
time as the 197 4 SOLAS Convention was introduced, the
International Maritime Organization (IMO), published Reso-
lution A.265 (VIII). These regulations used a probabilistic
approach to assessing damage location and extent drawing
upon statistical data to derive estimates for the likelihood of
particular damage cases. The method consists of the calcula-
tion of an Attained Index of Subdivision, A, for the ship which
must be greater than or equal to a Required Subdivision
Index, R, which is a function of ship length, passenger/crew
numbers and lifeboat capacity. The Equivalent Regulations
raised new damage stability criteria addressing equilibrium
as well as recommending a minimum GZ of 0.05 m to ensure
sufficient residual stability during intermediate stages of
flooding.
The next major step in the development of stability stan-
dards came in 1992 with the introduction of SOLAS part B-1
(in Chapter II-1), containing a probabilistic standard for
cargo vessels, using the same principles embodied in the
aforementioned regulations. The same principle is also the
basis for the current IMO regulatory development of "Har-
monisation of Damage Stability Provisions in SOLAS based
on the Probabilistic Concept of Survival." The 1980 U.K. Pas-
senger Ship Construction Regulations introduced require-
ments on the range of the residual stability curve as well as
on the stability of the vessel at intermediate stages of flood-
ing. The 1980 U.K. Passenger Ship Construction Regulations
introduced requirements on the range of the residual stabil-
ity curve as well as on the stability of the vessel at interme-
diate stages of flooding. The loss of the Herald of Free Enter-
prise in 1987 drew particular attention to Ro-Ro ferries in
which the absence of watertight subdivision above the bulk-
head deck is a particular feature. The implications of this
feature were highlighted by the Court of Inquiry, which ob-
served that the SOLAS Conventions and U.K. Passenger
Ship Construction rules had been aimed primarily at conven-
tional passenger ships in which there is normally a degree of
subdivision above the bulkhead deck, albeit of unspecified
ability to impede the spread of floodwater. In response to
this, the U.K. Department of Transport issued Consultative
Document No 3 in 1987, which outlined a level of residual
stability that required all existing Ro-Ro ferries to demon-
strate compliance with the 1984 Passenger Ship Construc-
tion Regulations. This standard had previously formed the
basis of a submission by the U.K. and other governments to
IMO, which considered the question of passenger ship stabil-
ity in some detail. This was the forerunner to SOLAS '90.
U.K. Ro-Ro research programme
In the wake of the Herald of Free Enterprise disaster, the
need to evaluate the adequacy of the various standards in
terms of providing sufficient residual stability to allow
enough time for the orderly evacuation of passengers and
crew in realistic sea states has prompted the U.K. Depart-
ment of Transport to set up the Ro-Ro Research Programme
comprising two phases. Phase I addressed the residual sta-
bility of existing vessels and the key reasons behind capsizes.
To this end, theoretical studies were undertaken into the
practical benefits and penalties of introducing a number of
devices [8], for improving the residual stability of existing
Ro-Ra's. In addition, model experiments were carried out by
British Maritime Technology Ltd. [9] and by the Danish
Maritime Institute [10] in order to gain an insight into the
dynamic behavior of a damaged vessel in realistic environ-
mental conditions and of the progression of flood water
through the ship. Phase II was set up with the following
objectives in mind:
JULY 2002
• To confirm the findings of Phase I in respect of the ability
of a damaged vessel to resist capsize in a given sea state.
• To carry out damaged model tests, in which the enhanc-
ing devices assessed in Phase I would be modeled to
determine the improvements in survivability achieved
in realistic sea-going conditions.
• To confirm that damage in the region amidships is likely
to lead to the most onerous situation in respect of the
probability of capsize.
• To undertake theoretical studies into the nature of the
capsize phenomenon, with a view to extrapolating the
model test results to Ro-Ro passenger ships of different
sizes and proportions.
Strathclyde was one of three organizations with the re-
sponsibility of developing and validating a theoretical capsize
model which could predict the minimum stability needed by
a damaged vessel to resist capsizing in a given sea state. This
was subsequently to be used to establish limiting stability
parameters that might form the basis for developing realistic
survival criteria [11].
Joint R&D project
As the UK stood poised to share the findings from the
Ro-Ro Research Programme with the rest of the world, the
Estonia tragedy had once more shaken the foundations of
shipping, forcing the profession to provide answers "immedi-
ately" and, in attempting to do so, to use the right expertise
and experience to provide the right answers. The Nordic
countries reacted quickly in undertaking this responsibility
leading to a wider-based project within a very short period,
taking onboard the fact that, in addressing the probability of
a ship surviving a given damage, the problem of damage
survivability does not end with quantifying the probability of
damage and the consequences of damage. The Estonia disas-
ter was the strongest indicator yet of the urgent need to de-
fine acceptable risks and maximum tolerable consequences
as well as to identifying procedures for managing such con-
sequences and dealing with residual risks. To this end, the
principal aim of the project was to develop a proposal for a
new probabilistic stability framework leading to improved
safety for new vessels with particular reference to the dam-
aged and flooded conditions. A second aim was the develop-
ment and application of safety assessment procedures to pas-
senger/Ro-Ro vessels. Leading experts from Europe
contributed to the technical work as shown in Fig. 2.
Survivability measure-critical height of water
on deck
The new damage stability framework proposed by the Joint
North West European R&D Project is based on the probabi-
listic concept of survival. This means that the standard of
survivability is expressed in terms of the probability that the
vessel will survive, given damage with water ingress has
taken place. The total probability of survival depends on two
factors: the probability that a compartment is being flooded
and the probability that the vessel will survive flooding of
that compartment. The concept itself is simple, but it takes a
great deal of effort to establish correct formulation of these
two factors, particularly when it involves large-scale flooding
of extensive undivided deck spaces such as the vehicle deck in
Ro-Ro ferries. Concerning the latter and taking into account
that there are many effects causing a vessel to capsize, the
probability of survival can also be divided into two different
factors: the probability to survive pure loss of stability, heel-
ing moments, cargo shift and angle of heel and the probabil-
ity to survive water accumulation on deck as a result of wave
action. The calculation of this last factor, referred to as sur-
MARINE TECHNOLOGY 139
 STABILITY
Phase 1

Task 3 Task 4 Task 6

Dynamic Cargo Shift Framework for
Effects /Securing New Damage
in Waves Stability Standard

I
Responsible: Responsible: Responsible: Responsible: Responsible: Responsible: Responsible: Responsible:
Knud E. Hansen Task 2.1: DTU Marintek MariTerm University Kvcerner-Masa Bureau Veritas DNV Technica
Task 2.2: DMI of Strathclyde Delta Marin DNV Technica
_J Germ. Lloyd

Nordic Group Tech. Univ. Lyng by DMI SSPA Task -Leaders Germ. Lloyd Task Leader
SSPA DNV Univ. Strathclyde Subprojects Bureau Veritas Task 6
Univ. Helsinki 2,3,6
Germ. Lloyd
Fig. 2 Nordic Project Organization

vival factor with water on deck, sw, is based on a concept where the righting moment has its maximum, i.e., ernax• cal-
whereby the critical wave height at which the vessel will culated traditionally by using the constant displacement
capsize is found, and sw will simply be the probability that method and allowing for free-flooding of the vehicle deck
this wave height is not exceeded. Strictly speaking, this criti- when the deck edge is submerged. This fact, coupled with
cal significant wave height cannot be determined uniquely observations from physical model experiments and the expe-
because of the random nature of the sea. In connection to rience amassed from studying large numbers of numerical
this, the term "Survivability Boundary" represents a contour tests led to the development of a Static Equivalent Method
within the capsize region ("Capsize Band") with equal prob- (SEM) which allows for the calculation of the critical amount
ability of vessel capsize. Therefore, the main task in estimat- of water on deck from static stability calculations. To this
ing the probability of survival with water on deck has been to end, a flooding scenario is considered in which the ship is
formulate a connection between the critical sea state and damaged only below the vehicle deck but with a certain
parameters which can be readily calculated without resort- amount of water on the (undamaged) deck inside the upper
ing to numerical simulations or physical model experiments. (intact) part of the ship. The critical amount of water on deck
Observations from the latter revealed that the dominant fac- is then determined by the amount causing the ship to assume
tor determining the behavior of the vessel is the amount of an angle of loll (angle of equilibrium) that equals the angle
floodwater accumulating on the vehicle deck, Fig. 3. In case
of large-scale flooding, the vessel motions become subdued
with the mean heel angle increasing slowly until a critical ROLL MOTION
value is reached beyond which heeling increases exponen-
tially and the vessel capsizes very rapidly. In this context, the 70
term "point of no-return" is used as indicative of the fate of 60
the vessel when this critical heel angle is attained. Put dif- g; 50
40
ferently, the floodwater on the vehicle deck increases slowly, ~
w 30
depending on the vessel and environmental conditions, until ...J
(!) 20
z
the amount accumulated reaches a critical level that cannot ct 10
be supported by the vessel/environment and the vessel cap-
-10
sizes quickly as a result. In relation to this, two points de-
serve emphasis. This amount is substantially less than the TIME (sec)
amount of water just before the vessel actually capsizes but is
WATER ON VEHICLE DECK
in excess of the amount required to statically capsize the
ship. In this respect, the energy input on account of the waves
7000
helps the vessel sustain a larger amount of water than what
here static restoring characteristics appear to dictate. Be-
6000
I -f···
cause of the nature of the capsize mode describe above, it is
~
en
(I)
c
c
5000
4000
3000
+ :
-f---·

r
r
-

not difficult to estimate the critical amount of water on deck ~

w I
···-

2000
at the point of no-return from experimental or numerical I- _,Ii,,_
simulation records considering either the floodwater on the ~ 1000
0 -
vehicle deck or the roll motion of the vessel as indicated by -1000 - n,,n
-
nAn •n
•v
">n - ·- ~ln --
the arrow in Fig. 3. TIME (sec)
A key observation from model experiments and numerical
simulations was that vessel capsize occurs close to the angle Fig. 3 Typical capsize mode with water on deck

140 JULY 2002 MARINE TECHNOLOGY

8max· Based on this, the volume of water on deck causing the Table 1 Distribution of test results among categories of vessel behavior
vessel to assume an angle of loll that equals emax> was com- (the 49 test cases considered by DMI in the aforementioned test
pared with the critical volume of water at the instant of cap- program [11,12])
size and a good correlation was found. The scenario described
Number Survival Survival
above and depicted in Fig. 4, is believed to represent closely
No. of of Heel to Heel to
observations of the flooding process near the capsize bound- Arrangement Tests Capsizes Intact Damage
ary or when a stationary (steady) state is reached with the
water on deck elevated at an average height, h, above the Open deck 44 21 19 4
mean water plane, as a result of the wave action and vessel Open deck with
motions. It was subsequently shown that this height is a bulkheads 7 1 3 3
unique measure of ship survival in damaged condition-the Center casing 12 7 0 5
higher the water elevation the higher the sea state needed to Center casing 18 0 0 18
with bulkhead 14 13 0 1
elevate the water to this level and the higher the capsizal Side casings 14 5 1 8
resistance of the ship-that could be applied universally to Side casings 6 0 0 6
all the arrangements studied, involving ship size and shape, with bulkhead 5 1 0 4
subdivision arrangements and loading conditions. It follows
that the relationship between h and H 8 will also be unique
for a given ship, thus allowing the survivability of the vessel
to be expressed as a function of the critical significant wave basis for taking into account water on deck in assessing the
height as denoted below: damage survivability of existing ships. Figure 5 summarizes
how to calculate the height of water on deck according to the
hcrit = f(Hs) = 0.085 (H8 cnt. )1. 3 Agreement, depending on the vessel residual freeboard and
the operational sea state, characterized by the significant
where hcrit is the difference between the inner and outer wa-
wave height, H 8 .
terline at the instant of capsize and H 8 crit the critical signifi-
cant wave height. Model test method
This finding was the precursor to the subsequent develop-
ments that led to what subsequently became known as the As indicated earlier, in view of the obvious lack of in-depth
Stockholm Agreement. understanding of the pertinent phenomena governing dam-
age vessel behavior in extreme sea-going conditions during
large-scale progressive flooding and uncertainties in the
3. Regulatory developments
state of knowledge concerning damage survivability, the PoE
PoE, Danish Proposals and Stockholm Agreement recommended after some debate an alternative method of
ensuring compliance through the Equivalence route, the
While the developments described above were taking Model Test Method of SOLAS '95 Resolution 14 [5]. The ex-
place, the prevailing climate following the Estonia disaster perimental set-up is depicted in Fig. 6. The method itself has
could not wait for the concept pertaining to SEM to be vali- been subjected over the past two years to rigorous scientific
dated and nurtured to maturity. Instead, using the idea of scrutiny by a group of technical experts, known as the
the critical height of water on deck as the basis, the Joint Gothenburg Technical Group, and a number of suggestions
Nordic Project commissioned a series of experiments at the for improvement were recommended to the Gothenburg
Danish Maritime Institute ([12,13] and Table 1) aiming to Group (the original signatories of the Stockholm Agreement)
produce evidence in support of the proposal of the PoE re- for a modified Model Test Method for assessing the damage
quiring a vessel to meet SOLAS '90 requirements with in survivability of new passenger/Ro-Ro vessels that could be
addition of up to 0.5 m 3 /m 2 water on deck. Results from these used for new vessels following agreement initially among the
experiments were subsequently used by the Danish to for- member countries and subsequently within IMO. The adop-
mulate a proposal to the first Stockholm Conference on Janu- tion of performance-based standards and of first principles
ary 27/28, 1996, relating the amount of water on deck to a approaches to assessing ship safety is undoubtedly of para-
constant height rather than a constant volume as was the mount importance. In this respect, the Model Test Method
case with the PoE proposal. This was again to be applied in will prove invaluable in paving the right way forward.
a static and deterministic sense and was eventually accepted
in the second Stockholm Conference one month later as the A critique of the Stockholm Agreement requirements
The introduction of the Stockholm Agreement is closely
associated with three unprecedented steps in the history of
damage stability/survivability assessment:

Height of 0.5
Water on
Deck
WL (Metres)
=+=====i===::::::::=
WL 0

0.0 Hs=l.5 m (basis)

2:0
Residual Freeboard {m)
Fig. 4 Stability of a damaged ship with water accumulated on deck (Static
Equivalent Method-SEM) Fig. 5 Stockholm Agreement (height of water on deck)

JULY 2002 MARINE TECHNOLOGY 141


Shallow End
....___ _ _ Carriage Rail
Model
Carriage
Fixed
Wave Probe
Wave Maker
Fig. 6 Model test method-experimental setup
• Water on deck was explicitly taken into account for the
first time. This is remarkable in view of the knowledge
that 85% of all deaths with ferry accidents relate to car
deck flooding.
• The effect of waves, and this is even more remarkable,
was explicitly taken into account also for the first time.
• It paved the way to the introduction of performance-
based standards for assessing the damage survivability
of ships (Model Test Method of SOLAS '95 Resolution
14).
All three steps represent gigantic improvements in the ap-
proach to addressing ferry safety but any potential benefits
will have to be balanced against any likely costs that might
be incurred through the introduction of inappropriate stan-
dards. There are certainly some obvious weaknesses in the
requirements of the Agreement and this must be borne in
mind when assessing Ro-Ro safety. Key among these include
the following:
• The Stockholm Agreement was built on the presumption
that a vessel designed (or modified) to SOLAS '90 stan-
dards ensures survival at sea states with H 8 of only 1.5
m. This was suggested in the face of uncertainty and
lack of understanding of the phenomena involved. The
evidence amassed so far and presented in the following
suggests that this was a considerable underestimate.
• The maximum penalty of 0.5 m height of water on deck
is ill based. It is to be noted from Table 1 that the 49
tests used to measure water accumulation on the car
deck comprised only 4 open-decked ships, the others
having car decks with: 3 transverse bulkheads, 5 central
casing, 19 central casings with transverse bulkheads, 8
side casings and 10 side casings with transverse bulk-
heads. It is straightforward to prove that the height of
142 JULY 2002
water accumulated on a subdivided deck is considerably
larger than the height of water accumulated on open
decks. More importantly, requirements based on subdi-
vided decks are likely to promote designs with similar
arrangements, which is contrary to the Ro-Ro concept
itself.
• Furthermore, all the tests performed at DMI referred to
midship damage, and the Stockholm Agreement was
thus calibrated on the basis of this damage. As a result,
and as the evidence accumulated so far and presented in
the following clearly shows, the maximum disagreement
between the Stockholm Agreement and performance-
based standards occurs when comparisons are made on
damages outside ±0.1 L from amidships, which are nor-
mally the worst damages, particularly when car decks
are subdivided.
• Finally, the effect of water on deck is taken into account
by a calculation method that does not preserve the phys-
ics of the problem, and being based on static and deter-
ministic approaches, it tends to negate the potential for
adopting rational approaches to safety through the in-
troduction of operational sea states and performance-
based standards.
4. Safety equivalence of the
Stockholm Agreement
Test matrix
A sample of 42 Ro-Ro vessels is utilized to form a suitable
matrix that allows for meaningful comparisons between the
various methods of assessing damage survivability as well as
between the routes to ensuring compliance with current sur-
vivability standards. The main aim of this study is to present
a critical evaluation of the emerging trends concerning the
level of safety provided by the current damaged survivability
standards for Ro-Ro vessels with the focus on the Stockholm
Agreement requirements. More specifically, the study ad-
dresses the following methods and associated parameters:
• SOLAS '90 [GZmax• Positive GZ Range, Area under GZ
Curve]
• Stockholm Agreement Calculations [Limiting H 8 ]
• Numerical Simulations [Limiting H 8 ]
• Model Experiments [Limiting H 8 ]
In addition to the above, the residual freeboard and dam-
aged GMr are also considered. The sample of ships consid-
ered includes ships ranging in length from 85 m to 205 m and
in damage stability standards from SOLAS '74 to SOLAS '90.
The relevant details concerning the vessels in question to-
gether with the results pertaining to the damage survivabil-
ity assessment according to the aforementioned methods are
presented in Appendix A (available in "More MT Online" on
the Web) with the operational sea states given in Table 2.
Analysis of these results allowed for illustration of trends,
consistency and relative significance between the methods
used. A typical sample is shown in Figs. 7 to 10.
Wave environment
The wave environment used in the numerical simulations
and physical model tests is representative of the North Sea
and is modeled by using a JONSWAP spectrum as shown in
Table 2.
Comparative assessment of Ro-Ro
damage survivability
Limiting H 8 in the derived results represents the maxi-
mum sea state the damaged vessel can survive repeatedly.
MARINE TECHNOLOGY
 Table 2 Sea states (JONSWAP spectrum with 'Y =3.3) The norm that has been adopted in presenting the results of
numerical simulations is to provide a capsize region rather
Signif. Wave Height Peak Period Zero-crossing Period than a capsize boundary to correctly reflect the fact that,
H 8 (m) TP (s) T0 (s)
because of the random nature of all the parameters deter-
1.0 4.00 3.13 mining a capsize event, a single boundary curve does not
1.5 4.90 3.83 exist. A limiting H 8 in the SA calculations is the maximum
2.0 5.66 4.42 value of the significant wave height (and hence height of
2.5 6.33 4.95 water on deck) in which the vessel fails any one of the rel-
3.0 6.93 5.42 evant criteria. A close observation of Figs. 7 to 10 combined
4.0 8.00 6.25 with a careful study of Tables Al to A3 of Appendix A leads
5.0 8.95 7.00 to the following noteworthy points:
• The agreement between physical model tests and nu-
merical tests is very impressive. With larger ships, in
particular, the results between the two are identical.

I(/) 3
::i:

1 2 3 4 5 6 7 8 9101112131415161718192021222324252627282930313233343536373839404142
SHIPS
Fig. 7 Comparative assessment of Ro-Ro damage survival (worse SOLAS damage)

5 i
i:
:
i :
i

,. ! tl
4 . ,. ' '· '''"' '·· . '"'''' •·· ..... ..., ...•.

·;,' I
L

i i i
ii
ii
ii n
I(/) 3 >·· i• • ~ ..
•
•••..•..J ,. ,, •... : .. ; .. ... ,..... ;,,, . •·· L ..

::i: I
: ; l

:...
i
,.. ,.i
..
2 . '···
i
:
'··
l' ;,.,.,, ,,,, ,,.

i i
i ! i
i
i .• ~.

[·1" .. 'i ••. : .• : ....

:
r
0
j
:'
i

I I 11111
1 2 3 4 5 6 7 8 9 101112131415161718192021222324252627282930313233343536373839404142
SHIPS
Fig. 8 Comparative assessment of Ro-Ro damage survival (midship damage)

JULY 2002 MARINE TECHNOLOGY 143

 6

•• *• •• •
5
•• t
•• • • •
4 •
l
.i+;d
. • ••1m" ,1'¢

• •• ..•
,.....,
E
..__.
3 ;
.:·• •• •• • • •
• •
.... jjlJi

....
(/)
:c t • ... """
·<llllJ

2 •
• HSPB
HSSTO
-Linear (HSPB)
0 ~Linear (HSSTO:
0.05 0. 1 0. 1 5 0.2 0.25 0.3

A GZmax [m]
5.5 ..

5 '

4.5

4
,.....,
E 3.5
..__.

•
(/)
:c 3

2.5

2, • + HSPB
HSSTO
1.5
-Linear (HSPB)
1 Linear (HSSTo:;
0.05 0. 15 0.25 0.35 0.45 0.55 0.65

B GZmax [m]
Fig. 9 (a) Worst SOLAS damage; (b) Midship damage

With the smaller size vessels, floodwater sloshing is
more pronounced and so should be the damping effect on
roll motion. Research to quantify the latter is currently
under way.
• In general, ships that satisfy SOLAS '90 criteria "pass"
the numerical/physical model tests and by implication
will be deemed to be safe according to the Equivalence
route. This is true for all the ships considered in this
sample. There are exceptions, of course, and it has to be
appreciated and understood that prescriptive criteria
could not possibly represent reality meaningfully in all
cases. This result is very encouraging, considering that
SOLAS '90 has been adopted as the new global standard
for all existing ferries. It is also somewhat surprising to
see that the previously adopted conjecture that vessels
constructed to meet SOLAS '90 standards were capable
of avoiding rapid capsize after damage in moderate sea
states with a significant wave height of only 1.5 m was a
drastic underestimate. Results of ships meeting SOLAS
'90 standards appear to be capable of surviving, on the
average, sea states above 3 m H 8 . In this respect, SOLAS
'90 provides the right platform for future developments.
• The critical parameter in achieving compliance with
SOLAS '90 is usually GZmax·
• The Stockholm Agreement standard is in general more
difficult to satisfy than the numerical/physical model
tests specified by the Equivalence route. The reason for
this derives directly from the fact that the height of wa-
ter on the vehicle deck postulated by this standard is
unrealistic. Its derivation was influenced largely by re-
sults from vessels with vehicle deck configurations that
were conducive to increased heights of water on deck,
namely, ships with side casings and transverse bulk-
heads [12,13]. In only very few occasions, the limiting H 8
calculated on the basis of the Stockholm Agreement ex-
ceeds that corresponding to the operational sea state.
This is shown clearly in Figs. 7 and 8, particularly so for
the worst SOLAS damage.
• Examining Figure 9, it would appear that the trends
between physical/numerical tests and Stockholm Agree-

144 JULY 2002 MARINE TECHNOLOGY

 4.5 - - - -
Hs[m]
4-
3.5
2.5
Fig. 1O Comparison between Stockholm Agreement and performance-based standards
ment are in general similar with the best correlation
resulting when considering GZmax· However, a system-
atic study is required before any concrete conclusions
can be derived concerning generalizations of such corre-
lations. In addition, the Stockholm Agreement underes-
timates limiting H 8 on the average by 2 m for the worst
SOLAS damage and by 1 m for the midship damage.
• Figure 10 clearly shows that the critical damage for com-
pliance is the worst SOLAS damage. This result con-
trasts with findings from earlier research where it was
shown that the midship damage is the most onerous
from a survivability point of view-a discrepancy that
can be attributed to the positioning of bulkheads on the
car deck.
• On several occasions what is defined to be worst damage
according to SOLAS '90 calculations is not the worst
damage from a survivability point of view. This is shown
in Tables Al to A3 of Appendix A where discrepancies up
to 1 m H 8 can be noted.
The investigation presented in the foregoing continues
with several other ships, thus offering a unique opportunity
to provide more convincing evidence of the correlation be-
tween safety standards as postulated by the current rules
and survivability standards as dictated by the vessel's opera-
tional environment. Based on this investigation, the follow-
ing observations can be made:
• The agreement between numerical and physical model
tests has been impressive enough so far to warrant care-
ful consideration for adopting the numerical tests as an
alternative to physical model testing, for compliance.
• The results derived from this study show worrying in-
consistencies between SOLAS '90 and Stockholm Agree-
ment standards, which are not in favor of the shipown-
ers/operators. At the current stage of research, the
following findings must be noted:
SOLAS '90 is a "good" standard reflecting meaning-
fully the safety of Ro-Ro vessels at a level of safety
JULY 2002
Worst SOLAS Damage
that is generally in agreement with that determined
through performance-based methods. However, it
must not be overlooked that it is a deterministic stan-
dard and hence it could misrepresent the true level of
a vessel's safety.
SA appears to be unrealistically stringent, in general,
demanding levels of safety well beyond those deter-
mined through performance-based methods and, at
times, simply not attainable.
5. Practical impact
In addition to upgrading concerning damage stability
safety standards, operational safety is as big a challenge for
designers and operators in the adversity of increasing com-
petition where lower costs, increased earning capacity and
enhanced safety standards, all conflicting criteria, are be-
coming key factors to success. Deriving from this, the prac-
tical impact on the design and operation of existing Ro-Ro
passenger ships deriving from the formal application of the
provisions of the Stockholm agreement is in general severe,
depending on the ship and area of operation. Relevant costs
include running costs (e.g., effect of increased resistance on
fuel costs following external modifications of the ship geom-
etry), operational costs (e.g., effect on line length and on turn-
around times following modification of internal ship layout)
and other less straightforward to quantify effects (e.g., com-
fort related implications because of the normally increased
vessel restoring stiffness as a result of the upgrading pro-
cess). The governing factors that determine the ensuring
costs comprise the following:
1. Current Stability Standard (of the ship in question)-In
general ships complying with SOLAS '90 two compartment
standard, or the equivalent A.265 probabilistic standard,
might be required to undergo very little or no modification,
when exploiting the optimization route, outlined in Fig. 1
(TSA). SOLAS '74 and especially SOLAS '60 ships, on the
MARINE TECHNOLOGY 145
 Table 3 Technical solutions adopted by the sample of retrofitted
Ro-Ro/passenger ships
Item Technical Information
Transverse doors on car deck major modification as it effects
overall cost, survivability and
operation significantly
Ducktail major modification as it effects
overall cost, survivability and
operation significantly
Ducktail sponsons major modification as it effects
overall cost, survivability and
operation significantly
Side sponsons major modification as it effects
overall cost, and survivability
operation significantly
Side casings it could be major or minor
conversion depending on cost
and effect on cargo capacity
Making existing rooms
watertight on car deck minor
Internal Tank-Rearrangement minor
Buoyancy tanks minor
Additional subdivisions minor or major, depending on
location and size of conversion
Making existing rooms
watertight below car deck minor
B/5 longitudinal bulkheads minor or major depending on
location and size of conversion
Cross-flooding arrangement minor
Heeling tanks minor
FW tanks minor
Ballast tanks minor
Stabilizing tanks minor
Scupper arrangements minor
Heeling tanks minor
Additional center casing on car
deck minor
Stern boxes minor
In flooding valves minor
New bulbous bow major
Foam filling s in void tanks major
other hand, are expected to undergo more severe modifica-
tions, which might give rise to technically and/or economi-
cally unviable solutions.
2. Ship characteristics-Independent of the currently valid
stability standard, the practical impact on existing ships will
depend to a large extent on the actual intact and damage
stability characteristics, especially the type of compartmen-
tation and compartment standard, intact and damage free-
board and intact and damage GM values. Ships with B/5
longitudinal BHDs under the Ro-Ro deck will in general be
less affected, as they comply with the design philosophy re-
sulting from SOLAS '90 provisions.
3. Area of operation-Considering the entire European
route network without restrictions, the most severe impact
concerns ships operating in the Northern North Sea and
Irish Sea, characterized by relevant significant wave heights
of 4.0 m for the purposes of compliance with the provisions of
SA. On the same basis, severe impact is also expected for
ships on the routes to Madeira, Azores (serviced from Portu-
gal) and Canary Islands (serviced from Spain), if the area of
application of the Stockholm Agreement is extended to South
Europe.
4. Technical solutions applied-There is a great variety of
possible technical solutions leading to compliance with the
provisions of the Stockholm Agreement, depending greatly
on the extent of the required modifications (factors 1 and 2
above) and the expertise of the technical consultant. Ap-
proaches are normally classified according to the choice of the
survivability-enhancing device (e.g., structural modifications
referring to changes in the internal and/or external ship ge-
ometry), operational measures (reducing draft/payload, in-
creasing displacement/payload via external modifications,
lowering operational KG, changing route to one with less
severe operational sea states or a combination of these) and
finally to the mode of achieving compliance (model test
method or calculation method). Concerning the latter, formal
application of the requirements of the SA, without optimiza-
tion, is likely to lead to ineffective modifications and economi-
cally unviable solutions. An indicative list of possible techni-
cal solutions is given in Table 3.
Some indicative cost values regarding compliance with the
provisions of SOLAS '90 and the SA and addressing initial
and operational cost are also shown, based on data of ships
operating in NW Europe. Table 4 refers to ships already con-

Table 4 Indicative costs of compliance with provisions of Stockholm Agreement (based on data of NW European operators [14])

Bow/Stern Cross-Flooding
Door Transverse Longitudinal Sponsons/ Drainage/
Reinforcement Doors/BHDs BHDs Ducktails Miscellaneous
Design
procurement mio ($US) mio ($US) mio ($US) mio ($US) mio ($US)
installation 0.25-0.60 0.60-1.80 0.60-1.80 0.60-3.60 0.12-1.20
Loss of payload tonnes tonnes tonnes tonnes tonnes
(lane length) 2-10 30-80 30-80 50-400 0-20
(5-10% on lane length)
mio ($US) mio ($US)
Loss of revenue not checked not checked not checked
1.2/yr 2.4/yr
Loss of speed 0 0 0 1-2 knots 0
mio ($US) mio ($US) mio ($US)
Manning 0 0
0.12/yr 0.12/yr 0.06/yr
mio ($US) mio ($US) mio ($US) mio ($US)
Maintenance 0
0.06/yr 0.06/yr 0.024/yr 0.024/yr

Notes:
1. Indicative costs are likely to change significantly from ship to ship.
2. Indicative costs refer to ships that underwent conversion for compliance with provisions of SA without adopting any design optimi-
zation.

146 JULY 2002 MARINE TECHNOLOGY

 Table 5 Indicative costs of compliance with provisions of SOLAS '90 Table 6 Continued
(based on data 15 NW European ships according to a U.K. study [16])
Previous
Increased Ship Stability Modifications to Stockholm
Initial Running No. Standard Agreement Requirements
Cost Mio costs/yr
LOA/yr of built Modification [$US] Mio[$ US] Ship 7 SOLAS '74 two transverse bulkheads on car deck
minor modifications below car deck
158.43 m/1974 doors on car deck 3.410 1.838 large sponsons
169.50 m/1987 doors on car deck 4.203 0 two transverse bulkheads on car deck
131.70 m/1976 sponsons* 3.122 0.050 Ship 8 SOLAS '74 modification to internal tank arrangement
80.40 m/1988 sponsons 2.024 0 Ship 9 SOLAS '74 large sponsons ("" 100 m)
137.01 m/1978 raise main deck 3.188 0.464 Ship 10 SOLAS '74 two transverse bulkheads on car deck
170.59 m/1977 buoyant wing tanks 1.797 *** longitudinal bulkheads on car deck along CL
126.50 m/1967 cross-flooding 0.166 0 Ship 11 SOLAS '74 optimized subdivision of existing side casings
129.85 m/1968 nil 0.017 0 Ship 12 SOLAS '74 one transverse bulkhead on car deck
120.71 m/1979 buoyant wing tanks 0.961 0.712 extended side casings
131.02 m/1980 doors 0.745 0.414 Ship 13 SOLAS '74 two transverse bulkheads on car deck
119.51 m/1975 sponsons** 3.125 0.041 sponsons
119.87 m/1976 sponsons 4.322 0.133 Ship 14 SOLAS '74 relocation of existing car deck barrier
107.60 m/1975 weathertight deck or ... 0.133 0 Ship 15 SOLAS '74 two transverse bulkheads on car deck
107.60 m/1975 sponsons 4.090 4.3 sponsons ("" 60 m)
161.50 m/1987 weathertight deck or ... 0.166 0 Ship 16 SOLAS '74 relocation of existing car deck barrier
161.60 m/1987 sponsons 3.585 8.280 sponsons with optimized subdivision
116.13 m/1974 sponsons 3.043 1.515 Ship 17 SOLAS '74 large sponsons
Ship 18 SOLAS '90 two transverse bulkheads on car deck
Notes: extended side casing
1. Indicated costs are based on data for 15 U.K. flag ships op- Ship 19 SOLAS '74 large sponsons
erating in NW Europe (North Sea, Irish Sea and Channel). Basis extended side casings
ships are considered to comply with SOLAS '74 stability standard. Ship 20 SOLAS '74 one transverse bulkhead on car deck
2. Proposed modification solutions are not optimal and in some modification to intenal tank arrangement
cases so severe, that removal from service and replacement appears Ship 21 SOLAS '74 four transverse bulkheads on car deck
necessary. Ship 22 SOLAS '74 two transverse bulkheads on car deck
3. * 0.083 mio [$US] to be added for berth modifications. modification to internal tank arrangement
** 0.828 mio [$ US] to be added for berth modifications. Ship 23 SOLAS '74 two transverse bulkheads on car deck
*** 0.265 mio [$ US]/yr savings due to increased payload. anti-heeling tanks
extended side casings
Ship 24 SOLAS '74 one transverse bulkhead on car deck
Ship 25 SOLAS '74 modification to internal tank arrangement
Ship 26 SOLAS '74 ducktail
verted for compliance with the prov1s1ons of Stockholm Ship 27 SOLAS '90 none
Agreement, without exploiting the possibility of optimization Ship 28 SOLAS '90 none
within the "Total Stability Assessment Procedure." Table 5 Ship 29 SOLAS '74 two transverse bulkheads on car deck
refers to data of ships operating in U.K. waters, deriving web side casings
from an investigation on possible compliance with the provi- modification to internal tank arrangement
sions of SOLAS '90. It is of interest to note that the practical Ship 30 SOLAS '74 large midship side casings
one transverse bulkhead on car deck
impact concerning compliance of Ro-Ro/passenger vessels modification to internal tank arrangement
with the requirements of SOLAS '90 are at approximately the Ship 31 SOLAS '74 two transverse bulkheads on car deck
same level as those required for compliance with the SA pro- web side casings
visions, much in agreement with the conclusions derived modification to internal tank arrangement
from performance-based criteria concerning the level of sur- Ship 32 SOLAS '74 stern boxes
vivability attained by SOLAS '90 ships. Finally, Table 6 re- one transverse bulkhead on car deck
fers to a sample of 42 ships operating in NW Europe that modification to internal tank arrangement
have been investigated at SSRC for possible compliance with Ship 33 SOLAS '90 none
the provisions of SA Typical examples shown different levels Ship 34 SOLAS '74 two transverse bulkheads on car deck
large stern side casings
of modification are shown in Figs. 11 to 13. modification to internal tank arrangement
anti-trimming tanks
Ship 35 SOLAS '74 one transverse bulkhead on car deck
modification to internal tank arrangement
Table 6 Sample of ships tested by SSRC for compliance with Stockholm ducktail
Agreement and Proposed Modifications following TSA Procedure Ship 36 SOLAS '74 two transverse bulkheads on car deck
modification to internal tank arrangement
Previous Ship 37 SOLAS '74 one transverse bulkhead on car deck
Ship Stability Modifications to Stockholm ducktail/sponson hybrid
No. Standard Agreement Requirements modification to internal tank arrangement
modification to forward sections of hull
Ship 1 SOLAS '90 none Ship 38 SOLAS '90 none
Ship 2 SOLAS '90 none Ship 39 SOLAS '74 large sponsons (""100 m)
Ship 3 SOLAS '90 none Ship 40 SOLAS '90 none
Ship 4 SOLAS '74 small buoyancy boxes on car deck Ship 41 SOLAS '74 one transverse bulkhead on car deck
Ship 5 SOLAS '90 none anti-trimming tanks
Ship 6 SOLAS '74 two transverse bulkheads on deck Ship 42 SOLAS '74 two transverse bulkheads on car deck
modification to internal tank arrangement modification to internal tank arrangement
extended side casings extended side casings

JULY 2002 MARINE TECHNOLOGY 147


Fig. 11 Complex modification-Ship 34
6. Concluding remarks
Based on the research work described in the foregoing and
in the light of past developments, the experience being
gained and of future expectations, the following conclusions
may be drawn:
• In the wake of the Herald of Free Enterprise and the
Estonia disasters, the regulatory authorities have intro-
duced demanding safety standards for passenger Ro-Ro
vessels, notably SOLAS '90 as the new global standard
and the Stockholm Agreement, a regional agreement
among North West European nations that require these
vessels to meet SOLAS '90 standards with up to 0.5 m of
water on the car deck.
• The Stockholm Agreement represents a major milestone
in the history of rule development for assessing ship
damage stability by taking explicitly into account the
effect of water on the Ro-Ro deck, by linking damage
survivability explicitly to operational sea states and by
paving the way to performance-based standards and to
first-principles approaches to ship safety.
• Evidence amassed in the route to compliance allowed for
a comparative assessment between the available regu-
latory instruments showing clearly that while SOLAS
'90 represents meaningfully a level of safety, which is
generally in agreement with that determined through
performance-based standards, the Stockholm Agree-
ment appears to be unrealistically stringent.
• The introduction of the Stockholm Agreement forced at-
tention on the safety of Ro-Ro passenger ships and in so
148 JULY 2002
Fig. 12 Medium complexity modification-Ship 36
doing it helped promote a safety culture in shipping,
pushing safety firmly at the center of the ship design
process and establishing it firmly in the minds of ship
designers and operators as a through lifecycle impera-
tive. The influence of this shift of attention to safety
coupled with technological developments and the need
to adapt to the rapidly changing drivers of shipping are
likely to have profound effects on Ro-Ro ship design,
construction and operation.
• The impact of the Stockholm Agreement on the existing
fleet of North West Europe has been much more positive
than most people feared. Shippers have either found a
cost-effective way to compliance through performance-
based approaches (numerical simulations and model
experiments), raising the safety of their fleet to its
rightful level, or cut their losses and opted for new, mod-
ern, safer, more efficient ship designs. Either way ship-
ping is undergoing a "face lift" and is looking much bet-
ter for it.
• The North-South divide in Europe concerning safety of
Ro-Ro passenger ships continues to trouble shippers and
regulators alike and a way forward is actively being
sought. Serving this need, an SSRC-NTUA partnership
has undertaken on behalf of the European Commission
a study to assess the impact of the Stockholm Agree-
ment on the areas covered by it with the view to evalu-
ating the likely impact of introducing the Agreement to
areas it does not cover. This introspection on the Stock-
holm Agreement will prove invaluable in paving the way
forward. The results of the Commission study itself will
form Part 2 of this paper.
MARINE TECHNOLOGY

Fig. 13 No modification-Ship 33
References
1. IMO Resolution MSC.12 (56) (Annex), "Amendments to the Inter-
national Convention for the Safety of Life at Sea, 1974: Chapter II-1-
Regulation 8,'' adopted on October 28, 1988.
2. MSC/Circ. 574, "The Calculation Procedure to Assess the Surviv-
ability Characteristics of Existing Ro-Ro Passenger Ships when Using a
Simplified Method Based upon Resolution A.265 (VIII)," June 3, 1991.
(Appendix tables Al-A3 follow, pages 150-158)
JULY 2002
3. International Maritime Organization (IMO), "Regulation on Sub-
division and Stability of Passenger Ships (as an Equivalent to Part B of
Chapter II of the 1974 SOLAS Convention),'' IMO, London, 1974. This
publication contains IMO Resolutions A.265 (VIII), A.266 (VIII), and ex-
planatory notes.
4. IMO Resolution 14, "Regional Agreements on Specific Stability Re-
quirements for Ro-Ro Passenger Ships"-(Annex: Stability Requirements
Pertaining to the Agreement), adopted on November 29, 1995.
5. IMO Resolution 14. "Regional Agreements on Specific Stability Re-
quirements for Ro-Ro Passenger Ships"-(Appendix: Model test method),
adopted on November 29, 1995.
6. Vassalos, D., Pawlowski, M., and Turan, 0., "A Theoretical Inves-
tigation on the Capsizal Resistance of Passenger/Ro-Ro Vessels and Pro-
posal of Survival Criteria,'' Final Report, The Joint Northwest European
Project, University of Strathclyde, Department of Ship and Marine Tech-
nology, March 1996.
7. Wendel, K. "Subdivision of Ships," Diamond Jubilee International
Meeting, New York, June 1968, pp. 12-1-12-21.
8. "Research into Enhancing the Stability and Survivability Stan-
dards of Ro-Ro Passenger Ferries: Overview Study." BMT Ltd., Report to
the Department of Transport, March 1990.
9. Dand, I. W., "Experiments with a Floodable Model of a Ro-Ro Pas-
senger Ferry,'' BMT Project Report to the Department of Transport, BMT
Fluid Mechanics Ltd., February 1990.
10. Ro-Ro Passenger Ferry Studies, Model Tests for FlO,'' Danish
Maritime Institute, Final Report of Phase I to the Department of Trans-
port, DMI 88116, February 1990.
11. Vassalos, D. and Turan, 0., "Development of Survival Criteria for
Ro-Ro Passenger Ships-A Theoretical Approach,'' Final Report on the
Ro-Ro Damage Stability Program, Phase II, Marine Technology Centre,
University of Strathclyde, December 1992.
12. Damsgaard, A. and Schindler, M., "Model Tests for Determining
Water Ingress and Accumulation,'' RINA Int. Seminar on the Safety of
Passenger Ro-Ro Vessels, June 7, 1996, IMO HQ, London.
13. Damsgaard, A., "Tests with a Model of a Damaged Ferry-Report
No. 2-Addendum," Joint Nordic Project Report 95016, DMI, February 5,
1996.
14. Kjaer, R., "Impact of Recent Stability Rules on Operation and Man-
agement,'' Proceedings, 3rd International Workshop on Theoretical Ad-
vances in Ship Stability and Practical Impact, Hersonissos, Crete, Octo-
ber 1967.
15. Kanerva, M., "Impact of Recent Stability Rules on the Design of
Ro-Ro Ships," Proceedings, 3rd International Workshop on Theoretical
Advances in Ship Stability and Practical Impact, Herssonissos, Crete,
October 1997.
16. Allan, T., "The Practical Implication of SOLAS '90 on Existing
Ro-Ro Passenger Ships," Proceedings, Ro-Ro 94 Conference, Gothenburg,
1994.
17. Papanikolaou, A., Bartzis, P., Bouolougouris, E., and Spanos, D.,
"Study on the Practical Implications of the Proposed New SOLAS Regu-
lations on Existing Greek Ro-Ro Passenger Ships and Critical Review of
the Proposed New Regulations,'' Vol. I-III, Final Report, NTUA-SDL,
Athens, September 1995.
MARINE TECHNOLOGY 149
 .....
01
0

c..
c
r
-<
I\)
0
0
I\)
Appendix A

Damage survivability standards for typical passenger Ro-Ro vessels from North West Europe
Table A I: Principal Particulars

LOA (m) 84.9 85.15 90.03 94 I01.25 161 166.3 175.48 163.4 176.8 175.3 129.78 152.()3 168 168.5 205.25

LBJ' (m) 78 78 80.4 85.2 91.2 144 150 154.2 157.6 159.6 164.6 119.47 136 153.4 154.5 181.6

B (m) 15.8 15.8 15.8 15.8 18 29 28.4 30.30 25.7 28.4 22 19.75 24.2 29.2 25.6 26.6

T (m) 3.176 3.1 3.18 3.15 4.17 6.4 6.7 6.65 5.5 6.7 5.85 5 5.8 6.5 5.71 6.1

ll (tonnes) 2342 2371 2648 3112 3814 17974 16648 21270 13293 18274 12778 7831 12508 17671 16678 I 19329
KG
Operational (m) I 6.5 I 6.769 I 7.256 I 7.568 I 8.89 I 13 I 12.76 I 13.842 I 11.4 I 13.3 I 10.57 I 9.5 I I 1.6 I 13.45 I l l.67 I 12.43
GMT
Intact Cm) I 1.937 I 2.108 I 1.909 I l.65 I 1.693 3.221 I 2.072 I 4.295 I 3.428 I 2.098 I 1.414 I 1.383 I 2.866 2.435 I 2.152 I I .498

s:
)>
::D
z
m
-i
m
(')
:::c:
z
0
r
0
G>
-<
 c...
c
r-
-<
I\)
0
0
I\)

Note: SC: Side Casings; CC: Centre Casing; FP: Freeing Ports; OD: Open Deck; SD: Subdivided Deck
Table A 1: Principal Particulars (Continued)

Lo" (m) 184.55 161 156.4 151 179.2 184.6 169.4 169.4 178.69 110.5 181.6 84.63 161.85 132.5 149.02 I 166.52

LHI' (m) 163 144 135.8 137 169.5 170.6 149.8 149.8 170.34 96.5 172.3 75 146.4 126.1 131.00 I 150.00

B (m) 26.4 27.6 25.9 23.5 25.08 23.5 27.6 27.6 24.49 17.2 23.4 15.8 27.6 22.7 26.0 I 28.4

T (m) 6.44 6.217 6.08 6.5 6.1 6.35 6.35 6.35 5.5 4.7 5.75 3.1 6 .. 21 5.5 6.12 I 6.8

6 (tonnes) 15902 16574 12359 13095 19101 17306 16810 I 16811 I 15310 I 4618 I 15437 I 2386 I 15922 I 8562.2 I 12536 I 16939
KG
Operational (m) II 10.45 12.81 10.35 I I Cl.3 I 11.989 I 10.2 I 12.95 13.077 I 10.18 I 7.27 I 10.889 7.13 I 13.100 10.370 12.32 I 13.41
GMT
Intact (m) I 3.98 I 2.167 I 3.6 I l.84 I 1.548 I 1.85 I 1.844 I 2.8 I 1.29 I 1.994 I 1.699 I 2.159 I 2.100 I 1.50 I 1.732 I 2.153

s:
)>
:xi
z
m
-I
~ Note: SC: Side Casings; CC: Centre Casing; FP: Freeing Ports; OD: Open Deck; SD: Subdivided Deck
z
0
r-
eG)
-<
.....
~
 ......
(11
I\)

c..
c
I
-<
I\)
0
0
I\)

Table A I: Principal Particulars (Continued)

. .· < ·.>.::...
i{Q~;~}iil . Sbi}J4~
}

Ship33 Ship.34 •. Ship3S Ship36

·.
Ship37
·.·.
SJltpJ8 i.· · . • ' ".< .
· Ship4t: .~~~~····~~· !.?S.fl,~4~·· 1 Sll;jp44;;J. Ship45. j Ship.46 l Ship47 t.~hip48

LOA (m) 179.0 190.0 132.65 181.6 163.9 133.8 163.53 133.0 177.0 137

LHI' (m) 170.0 169.0 120.7 171.99 150.0 126.5 150.0 122.3 159.6 126.0

B (m) 27.8 29.0 21.0 21.60 28.4 22.5 27.0 22.5 28.4 24.0

T (m) 6.25 6.74 5.0 6.1 6.8 5.0 6.505 5.0 6.7 5.64

/:-, (tonnes) 17300 21550 7889.7 14256 16911 8681 15342 9072.2 18375 11368
KG
Operational (m) II 12.89 14.10 I 9.95 9.43 I 13.61 I 10.6 12.442 I 10.266 I 13.71 11.4
GMT
Intact <m) II 2.63 I 1.50 I 1.674 I 0.966 I 3.32 I 1.97 I 2.505 I 2.413 I 1.616 1.591

~
:0
Note: SC: Side Casinbas; CC: Centre Casing;
-- .....
FP: Freeing..... Ports; OD: Open Deck; SD: Subdivided Deck
z
m
-I
m
(")
:c
z
0
I
0
G>
-<
 c...
c:
....
-<
I\)
0
0
I\)

Table A2: Limiting Parameters (Worst SOLAS Damage)

0.124 0.134 0.131 0.202 0.147 0.178 0. 187 0. I 58 0.165 0. I 13 0.1 I I 0. 123 0.201 0.156 0.128 0.108
GZmax (m) II (0.12) (0.13) (0.13) (0.2) (0.14) (0.14) (0.13) (0.15) (0.16) (0. I I) (0.1 I) (0.1 I) (0.!l) (0.14) (0.12) I (0.1)

Range (degrees) I 23.35 22.05 17.35 33.22 20.6 13.93 17.29 l l .63 17.88 I 8.41 16.844 24.425 18.526 21.837 16.92 I 17.595
Arca up to
GZmax (m.rad) I 0.0163 I 0.0131 I 0.0126 I 0.0339 I 0.0153 I 0.0106 I 0.0184 I 0.0101 I o.o 122 I 0.0097 I 0.021 I 0.02 I 0.016 0.018 I 0.012 0.01
Arca up to cp,
(m.rad) I o.0353 I o.m 12 I cHm5 I 0.0108 I o.o3n I 0.0265 I 0.0321 I 0.0206 I 0.0294 I 0.0283 I 0.021 I 0.034 I 0.039 I 0.044 I 0.()23 I Cl.02 I

Limiting Hs -
Worst SOLAS 4.25 - 4.00 - 3.50- 4.25 - 4.25 - 4.75- 4.50- 3.75 - 3.00- >5.00 >5.00 3.75- 3.25 - 4.75 - 3.25 - 3.25
~ II 4.50 I 4.25 I 3.75 I 4.50 I 4.50 I 5.00 I 4.75 I 4.00 I 3.25 I I I 4.00 I 3.50 I 5.00 I 3.50 I 3.50
Limiting Hs -
Actual 3.75 - 4.25 4.00- 4.25 -
(m) II I I I I 4.00 I 4.50 I I I I 4.25 I 4.50

:s:
)>
::0
z
m
-I
m
(")
:c Limiting Hs (m) >4.00 >4.00 >4.00 >4.40 >4.00 >4.50 4.70 3.75 - 3.00- >4.50 >5.00
z 4.80 4.00 3.25
0
....
0
G)
-<
......
CJl
(,.)
 ....
<J1
.;..

c...
c:
!:(
I\)
0
~

Table A2: Limiting Parameters (Worst SOLAS Damage), Continued

0.128 0.119 0.104 0.148 0.113 0.133 0.131 0.117 I 0.131

GZmax (m) 0.183 (0.109) (0.107) (0.1) (0.137) (0.1) (0.1) (0.1) (0.117) (0.125)

Range (degrees) 14.56 13.508 17.81 23.86 13.229 14.86 17.0 14.2 14.08 I 15.5
Arca up to
GZmax (m.rad) II 0.015 I 0.008 I 0.009 I 0.()1 I 0.01 I (l.009 I o.011s I 0.01 I O.()] 0.0118
Arca up to cp,
(m.rad) II O.o31 I 0017 I 0.0228 0.025 0.021 I o.o 174 I 0.0255 0.0192 I 0.017 I 0.0222

Limiting Hs -
Worst SOLAS 3.75- 3.75 - 3.5 - 3.5 - 3.0- 3.25 - 3.25 - 3.0- 3.5 - 2.75 -
(m) I 4.0 I 4.0 I 3.75 I 3.75 3.25 I 3.5 I 3.5 I 3.25 I 3.75 I 3.0

:s:
~
::0
z
m
-I
m
0
::r:
z
0
I""'
0
G)
-<
 c...
c:
r-
-<
I\)
Table A2: Limiting Parameters (Worst SOLAS Damage), Continued
0
0
I\)

0.26 0.101 0.239 0.103 0.101 0.15 I 0.128 0.159 0.119 0.112 0.125 0.152 0.128 0.114 0.175 0.156
GZmax (m) I (0.1) (0.1) (0.1) (0.1) (0.1) (0.1) (0.122) (0.1) (0.1) (0.11) (0.12) (0.121) (0.1) (0.14) (0.142)

Range (degrees) II 11.704 15.57 16.807 15.63 21.53 23.008 13.5 11.98 13.96 16.70 15.918 30 18.57 15.193 22.08 13.2
Arca up to
GZmax (m.rad) II 0.0279 0.0061 0.018 I o.oo8s 0.0086 0.019 0.0094 0.()()9 I 0.0091 I o.c)093 I o.c)11s I 0.0388 I 0.0087 0.008 I 0.012 I 0.013
Arca up to <j\
(m.rad)
I 0.04 0.0169 I 0.046 o.o 183 I o.cl237 1 0.039 I 0.0184 I Cl.0212 I 0.0174 I O.Cl2 l 3 I O.Cl225 I 0.0642 I 0.0267 0.018 0.040 0.0221

STOCKHOLM AGREEtv[ENT

Limiting Hs (m) II 2.65 l.5 2.75 I l.65 I 2.00 3.70 I 1.6 1.50 I 1.5 I 1.7 I 2.5 I >4 I 2.3 1.75 >4.0

-----~-~-------------------.----~~----; __ ;~--~-------------------··-------------------·---------·------···--··:.:............--~----!..~!l£.1:.!~..<!..1.!£~:.IJ.'!..~~efc_£t~~~!if!.~-----~---L..i_,.;;~;.;;:.~~-;...~~----·.. ~---~--:........~c••~L..................................................................

NUMERICAL SlMULATlQN
Limiting Hs -
Worst SOLAS 4.00 >4.50 4.25 3.75 >4.50 4.00- 4.00- 3.75 - 4.00- 2.50 4.50- >4.75 3.25 2.75 - 3.5 3.25 -
(m) 4.25 4.25 4.00 4.25 2.75 4.75 3.5 3.0 3.75 3.5
Limiting Hs -
Actual 3.0-
(m) 3.25

s
I Limiting I ls (m) I 3.8
I
>4.75
I
4.00
I
3.5
I
4.75
I
>4.00
I
4.45
I
3.75
I
>4.00
I
2.75
I
>5.00
I
>4.00 I 3.25-3.5 I 3.0 -
3.25 I
3.5-
3.75 I
3.75 -
4.0
:t:>
::0 Limiting Hs -
z
m Actual 3.25 3.25 - 3.0 - 3.0- 3.0- 2.75 -
-I
m
('")
:c
I (m) II I 3.5 I 3.5 I 3.25 I I 3.25 I 3.25 I 3.0

z
0
r-
0 PHYSICAL MODEL TESTS
C>
-< 3.75 3.0- 3.5 - 3.5 -
Limiting Hs (m) >4.0 4.0 >4.0 >4.0 :us 3.75 >4.0 3.75 I >4.0
.....
(11
(11
......
CJ1
a>

c..
c
I
-<
N
0
0
N Table A3: Limiting Parameters (Midship Damage)

Ship 16

GZmax ~ 0.192 N/A 0.197 ().255 0.161 0.287 0.195 0.274 0.247 0.179 0.208 0.156 NIA 0.165 0.172 I 0.149

Range (degrees) 25.48 22.14 34.553 19.29 17.38 17.03 22.35 20.36 20.05 16.931 27.533 I I 22.911 I 16.31 I 18.784
Arca up to
GZmax (m.rad) 0.0192 I 1 o.cl262 (l.0415 0.0174 0.0603 I 0.0196 0.0227 0.0236 I 0.0463 I ().()30 I 0.024 I I 0.046 I 0.015 I 0.015
J to cp, I I 0.0486 I 0.0927 I 0.()35 I 0.0603 I 0.0334 I 0.067 I o.0546 I 0.0463 I 0.()30 I 0.049 I I 0.046 I 0.028 I 0.029

3.97 3.16 3.45 >4 >4

Limiting Hs -
Worst SOLAS 4.75 3.75 - 4.25 - 4.25 >5.00 4.50- >5.00 4.00 >5.00 4.50- 3.75 - >5.00 >3.50 3.25 -
(m) I 5.00 I I 4.00 I 4.50 I 4.50 I I 4.75 I I 4.25 I I 4.75 I 4.00 I I I I 3.50
Limiting Hs -
Actual 3.75 - 4.50- 3.75 4.25 -
(Ill) II I I I I 4.00 I I 4.75 I 4.00 I I 4.50

[32
:s:
:t>
::0
z
m
I
Umih<>g H, (m) I >4.00
l I
>4.00
l
>4.40
l
>4.00
I
>4.35
I
4.50-
4.75 I
>5.00
I
4.00 -
4.25
I >4.75 I >5.00

-I
m
(")
:::c
z
0
I
0
G)
-<
 c...
c:
r-
-<
I\)
c
c
I\)

Table A3: Limiting Parameters (Midship Damage), Continued

Ship 32

3.4

0.1 0.176
GZmax (m) NIA 0.582 NIA 0.209 0.108 0.164 0.151 NIA 0.153 0.149 0.128 0.174 NIA (0.1) NIA (0.142)

Range (degrees) 41.38 24.23 18.65 25.181 16.46 16.13 14.84 17.56 27.95 27.82 15.01
Arca up to
GZrnax (m.rad) I I 0.1389 I I 0.0264 I 0.0115 I 0.02 I 0.0107 I 0.0143 I 0.0111 0.015 I I 0.0154 I ().()36 I 0.01
Arca up to tj>,
0.2302 I I ().()578 0.022 I 0.046 I 0.0245 I I 0.0245 O.Cl238 I 0.0239 I 0.0516 I I (l.058 I I 0.028

Limiting Hs -
Worst SOLAS I I >4.75 I 3.75 I 3.75 _ I 4.00 _ I 4.00 I 4.00 _ I 2;0~) _ I >5.00 I >4.75 I I 3.75 _ I I 3.75 -
(m) I 4.00 4.25 I 4.25 -.~5 4.0 4.0
Limiting Hs -
Actual 3.0
-----
(m) II I I I I I I I I I I I I I I I 3.25
:s:
)>
::0
z
m
-I
m I Limiting Hs (m) I 4.75 3.75 4.00 I >4.00 I 4.4
I
4 00 2
I :2~5- I
4.50 I >4.00 I 4.0 -
I
3.75
(')
:::c
z
I I I I > · I 4.25 I -4.0
0
r-
0
G>
-<
.....
U1

""'
....
Ul
co

c:...
c
I
-<
l'IJ
0
0
l'IJ Table A3: Limiting Parameters (Midship Damage), Continued

Sbip33 Ship34 Ship35 Ship 36 I Ship 37 Ship 38 Ship 39 f Ship 40 TShip4L I <s~ip42 Ship 43 Ship44 Ship 45 I .Ship 46 Ship47 Ship48

Operational Hs 2.6 4.0 J.4 J.4 2.6 2.5 J.O J.O J.4 2.6
(rn)

-.......................-........... .... __ ...........-.......-....-...

,.. ,,_ ----~
Deterministic Criteria
..--........._..........-.........................................................................-----------.. ----·-...----...----...--..--- ..-"'-"'_"" ..............._,.. ........................................................ _.,. __ .,..._.,..__ ........................................ .., .....,,,._ .. _.,, __ ,,,, ___ ., .............. ""....... ,.. ................................................ ,..,,. ......................................................................................... .., ............................................................................... -

SOLAS '90 .• :

0.284 0.194 0.196 O. IJ5

GZrnax (rn) NIA (0.107) NIA I NIA NIA I (0.1) NIA NIA (0.125)

Range (degrees) 15.399 22.16 20.J JJ.7

Arca up to
GZrnax (rn.rad) II 0.022 0.024 0.()206 0.0485
Arca up to<)\
(m.rad) II 0.046 O.OJ7 (l.0426 (l.029
..

STOCKHOLM AGREEMENT

Limiting Hs (m) >4.0 J.7 >4.0

····--·"·•••.::••·····•••••••••••·•··-·•··-~----··"·--··-·•·•·••·••···---·-.-----·····-·••••·••·•·•·•••••·••·••••••'"···--·---~...-!..'!.!l'!..~!1:!.'!:.~!'?..'!.~!1'!!..!!..i!...9..~i~'!.!...~'!...--•-~----C•---~~-~---~--~,.:.•.~-----··-·•··-·---··--·---:...c.:-~----····-···--···-·············--·-·-··········---·-·-··•--···.:--...
NUMERICAL SIMULATION
Limiting Hs -
Worst SOLAS 4.25 - >4.0 J.75- >J.O
(m) 4.5 4.0
Limiting Hs -
Actual J.25 - J.O-
(m) J.5 J.25

PRYS/CAL MODEL TESTS

s)> >3.4
::0 Limiting Hs (rn) 4.25 - >4.0
z
m
4.5
-f
m
(")
:c
z
0
I
0
G>
-<
© Marine Technology, Vol. 39, No. 4, October 2002, pp. 199-21 O
Marine
Technology

Stockholm Agreement-Past, Present, Future (Part 2)

Dracos Vassalos 1 and Apostolos Papanikolaou 2

This paper pr?vide~ a. succinct summary of the findings ensuing the undertaking of a dedicated EU-funded
research proiect aiming to address the impact of the Stockholm Agreement (SA) on the EU Ro-Ro
passenger ships. !his is achieved by utilizing the experience gained, the data and knowledge accumulated
t.hrouQh the ado~t1on oft.he St?ckholm Agreement in North West Europe to form the basis for predicting the
likely impact ?f intr?duc1n9 th1~ Agreem~nt to vessels operating in EU waters not covered yet by it. The
background, in~lu?1ng a h1stoncal overview and a detailed introspective look at the SA, together with an
~ssessment ?f its impact on pa~senger Ro-Ro ship safety standards, design and operation are presented
in a companion paper (Part 1) 1n the July 2002 issue of MARINE TECHNOLOGY.

Introduction the ferry industry the attractive possibility of utilizing such

CONCERTED action to address the water-on-deck problem in
the wake of the Estonia tragedy led IMO to set up a panel of
experts to consider the issues carefully and make suitable
recommendations. However, the complexity of the problem
and the need to take swift action to reassure the public that
appropriate steps are taken to avoid a repeat of the Estonia
disaster influenced to a large extent both the initial and final
proposals. Following considerable deliberations and debate
(obviously unresolved), a new requirement for damage sta-
bility has been agreed upon only among the northwestern
European nations to account for the risk of accumulation of
water on the Ro-Ro deck. This new requirement, known as
the Stockholm Agreement [1] demands that a vessel satisfies
SOLAS '90 requirements (allowing only for minor relaxation)
with, in addition, water on deck by considering a constant
height calculated as shown in Fig. 1. The term H, charac-
terizing the operational sea state, pertains to the ;verage of
one-third highest waves (significant wave height, typically of
an undeveloped sea-hence narrow banded), a statistical av-
erage correlating very well with the average of observed wave
heights for a given sea state.
The dates of compliance with the provisions of the agree-
ment range from April 1, 1997 to October 1, 2002. However,
in view of the uncertainties in the current state of knowledge
concerning the ability of a vessel to survive damage in a given
sea state, an alternative route has also been allowed which
provides a non-prescriptive way of ensuring compliance,
through the "Equivalence" route, by performing experiments
in accordance with the SOLAS '95 Resolution 14 [2].
Deriving from the above, numerical simulation models de-
veloped on the basis of systematic research over the past 15
years [3], and capable of predicting with good engineering
accuracy the capsizal resistance of a damaged ship, of any
type and compartmentation, in a realistic environment while
accounting for progressive flooding were also used, offering
"tools" to assess the damage survivability of ferry safety, the
so-called "Numerical Equivalence" route. Numerical simula-
tion readily allows for a systematic identification of the most
cost-effective and survivability-effective solutions to improv-
ing ferry safety and hence offers a means for overcoming the
deficiency of the physical model tests route in searching for
optimum solutions and an indispensable "tool" for the plan-
ning and undertaking of such tests.
October 1, 2001 has therefore marked the beginning of the
final year of the period initially allowed for compliance with
the Stockholm Agreement (SA) requirements, a period dur-
ing which almost 80% of the Ro-Ro fleet in North West Eu-
rope has been subjected to calculations, model testing and
numerical simulations on the way to meeting the new re-
quirements pertinent to the Agreement. The experience
gained has been invaluable in understanding better the prob-
lem at hand and is being utilized to shape new developments
for future Ro-Ro designs. All relevant details are presented in
a critical overview of the Stockholm Agreement from its in-
ception to its implementation and ensuing impact in a com-
panion paper (Part 1) in the June 2002 issue of MARINE TECH-
NOLOGY.
The North-South divide, however, continues to cause un-
rest, particularly at the European level. Efforts to assess the
status quo in North West Europe and use the information
amassed so far as a means to predict the potential impact of
introducing the SA in the South, led to a dedicated call by the
Commission and to a contract being awarded to two closely
collaborating research teams in North and South Europe, one
at the Ship Stability Research Centre of the Universities of
Height
of Water
on Deck

1
The Ship Stability Research Centre, NAME, The Universities of H=l.5 m
Glasgow and Strathclyde, UK.
2
National Technical University of Athens, Ship Design Labora- Residual Freeboard
tory, Greece. Fig. 1 Stockholm Agreement (height of water on deck)

OCTOBER 2002 0025-3316/02/3904-0199$00.49/0 MARINE TECHNOLOGY 199

 IMPACT ASSESSM.ENT OF STOCKHOLM AGREEl\fENT METHODOLOGY

Comparative analysis of National

Leteisl•ttio111, particularly w.r.t. Article
of the SA and one.-compartment
ships.

Approach: DGVll, DGm and National

Maritime :Administrations.

Inventory of Ro-.Ro passenger

categorised by A/Amax and Hs.

Approadt: Publications from UK and

Swedbb· A~"10.istmtions covering North
Sea aad pmtic Sea1 respectively.

Inventory of all Ro-Ro

upgradi.ngs tt> ·comply with SA by
31112/lJt and associated costs*

ARrul~U National Maritime

Admi1aisitra~tio11sand collaborating ferry
opemton.

Establish average compliance Assessment of overall costs f<n·

per vessel as a fHction of Hs for compliance with SA of all ships in B(b).
each A/Amax category.
MW-!:!!!~ Use of average costs together wit
ARw:!l~!J..Databases of SSRC and trends and informati<m based on
u Ai....;.,1111., and use of in..fu:mse kmnv upgradings and studies done so far.
how and expertise,.

Assessment of overall cost, for

compliance with SA by 31112/2002 latest,
of all ships in A(b).
"""1'·"'"''''h' Use of average costs together
known trends and information based
on upgradings and studies done so far.

Fig. 2 Adopted methodology for SA study

Glasgow and Strathclyde and the other at the Ship Design work undertaken, a brief account of which is presented here,
Laboratory of the National Technical University of Athens. following an outline description of the background and aims
This study was finalized in March 2001 and a detailed tech- of the study and of the methodology adopted in completing
nical report produced describing comprehensively all the this work.

200 OCTOBER 2002 MARINE TECHNOLOGY

0
(')
-I
0
OJ
m
JJ
"'
0
0
"'

Others Bahamas/Bermuda
Period 1961-1970 8% 7%

After 1991 6%

Estonia
4%

Period 1971-1980
36%

Finland
10%
10%

Period 1981-1990 Netherlands Lituania Italy

30% 5% 2% 3%
2%

(NEU - 286 sampled vessels) (NEU - 295 sampled vessels)

Tunisia Canary Islands

Croatia 1w.1,. 60;;,
After 1991 Period 1961-1970
22%

Panama
1%

Period 1981-1990
14'%

43%

s:
)>
JJ A (SEU - 302 sampled vessels) B (SEU - 302 sampled vessels)
z
m Fig. 3a Distribution of year of built Fig. 3b Current flag distribution
-I
m
(')
:c
z
0
r-
0
G>
-<

"'
~
I\)
0
I\)

0
(')
-I
0
CJ
m
::0
I\)
0
0
I\)
A265

ENE:u=o~ f\JAil1ax NEU-current NArnax - sEU-NArnax I

1.05

1.00

0.95
(NEU - 128 sampled vessels)
I><
0.90
A. 265 SOLAS 60 c:
5% 4% 8
~ 0.85
<
0.80

0.75
SOLAS74
35% 0.70

0.65 ·!111,11111111.11,11111111111111111111111111111111111111111111111111111111111111111111111111111 .llllLlllllll 111111111 .1111111111111il

B
- ~
-M ~
- ~
N
-
M
~
M
~
~

Fig. 4b
~
~
~
~
-
~
~
~
M
~

Distribution of A/Amax
~
~ ~
~ -
~
~
~
M
- -
0
~
0
--
~
-
-
N N
~

~
M
M

SOLAS90/92
54%

2%
A (SEU - 85 sampled vessels)
Fig. 4a Compliance with SOLAS regulations
::::::
)>
::0
z
m
-I
m
(')
:c
z
0
r-
0
G)
-<
 Fig. 5 Distribution of H5 is NEU
The SSRC-NTUA Commission Study
Background and aims of the study
At the conclusion of the second Stockholm Conference at
which the Agreement was adopted, the Commission services
issued a statement, taking note of the Agreement concluded
and expressing the opinion that the same level of safety
should be ensured for all Ro-Ro passenger ferries operating
in similar conditions. Noting that the Agreement is not ap-
plicable to other parts of the European Union, the Commis-
sion announced its intention to examine the prevailing local
conditions, environmental and operational, under which Ro-
Ro passenger ferries sail in all European waters and that this
examination will include the extent and effect of the appli-
cation of the Agreement in the region covered by it. The state-
ment concluded that in light of this examination the Com-
mission would make a decision with regard to the need for
further initiatives and this statement was confirmed at sub-
sequent meetings of the Council highlighting the need to en-
sure the same level of safety for all Ro-Ro ferries operating in
similar conditions was more precisely defined by referring to
both international and domestic voyages.
Furthermore, in its latest proposal for Community legisla-
tion governing the safety of Ro-Ro passenger ships, the Com-
mission included a draft provision that Ro-Ro ferries shall
fulfill the specific stability requirements adopted at the re-
gional level, when operating in the region governed by such
regional rules. This proposal was endorsed by the Council
with a number of adaptations to clarify that host States shall
check that Ro-Ro ferries "comply with specific stability re-
OCTOBER 2002
quirements adapted at regional level, and transposed into
their national legislation, ... provided those requirements do
not exceed those specified in the Annex of Resolution 14 (Sta-
bility Requirements Pertaining to the Agreement) of the 1995
SOLAS Conference and have been notified to the Secretary-
General of the IMO, in accordance with the procedures speci-
fied in point 3 of that resolution."
Taking fully into account the above elements, the Commis-
sion invited tenders to a study to examine the extent and
effect of the application of the Stockholm Agreement concern-
ing specific stability requirements for Ro-Ro passenger ships,
and the suitability of extending its scope to European waters
not covered by it. The contract to undertake this study was
awarded to the NAME-SSRC/NTUA-SDL partnership. More
specifically, the overall aim of the study was to assess the
impact of the Stockholm Agreement on European Ro-Ro pas-
senger ships by targeting the following two objectives:
A. Impact assessment on the extent and the effect of the
application of the Stockholm Agreement concerning
specific stability requirements for Ro-Ro passenger
ships in the area covered by it.
B. Impact assessment on the extent and the effect of the
application of the Stockholm Agreement concerning
specific stability requirements for Ro-Ro passenger
ships in European waters not covered by it.
Proposed methodology
The methodology adopted in completing this work is shown
in Fig. 2, explaining for each of the two distinct areas A and
MARINE TECHNOLOGY 203
 B the scope and approach to be followed to attain the results
(J)
sought. The study took one year to complete and produced
::<: two comprehensive technical reports addressing each one of
~~
I- 0
Cl the two areas separately [4,5]. The key findings of this work
~
(J)
Cl 0
a <(
w 0 (J)
::<: are summarized here for areas A and B, respectively, follow-
z
U)
~*
..J ....
::::>
~ ing the format of the adopted methodology.
:.::
z CD
!Q

;*
z .....
::::; •
CD

Area covered by the Stockholm Agreement

w
w 00
I (jj
[J U)
U)
CD
>
Key findings
(")
l{)

I As shown in Fig. 2, the study comprises five tasks, each

::J
w forming part of the methodology adopted as explained
~ therein and summarized below:
U)
l'J
n:: c
0
Comparative analysis of National Legislation A(a)-The
<(
l'J '@ National Legislations of the counties being parties to the SA
z 0
s~
;;:::
'6
were elaborated upon, aiming to: ascertain if the application
g ("J> 0 of the SA is extended to Ro-Ro passenger ships entitled to fly
u. E
ch
(J) 0c the flag of States non-parties to the Agreement; assess the
0
n::
0
.E extent to which parties to the SA are bringing one-
illl!
0 compartment Ro-Ro passenger ships in compliance with
c technical requirements of the SA as a matter of priority; iden-
0
:s
.0
tify if Contracting Governments to the SA apply earlier
(J) ·;::
:? 0en * implementation dates than those specified in Annex 2 to the
gog 0
n:: Agreement for ships trading between their ports.
l'J
z .c The main results from task A(a) showed that, in the main,
:;;:: CD
<(
~ ci> all countries affected by the SA have applied it without al-
(J)
t'2 u:: terations. Exceptions to this are the U.K. and Norway. In the
I-
::<:
z z first case, it is interesting to note that the U.K. also applies
w
~ ~
>- '#.
w
l'J the Stockholm Agreement requirements to Ro-Ro passenger
0"' z
z~
<( ~~ ships operating on comparable domestic seagoing routes
>- n:: ~
0
::::>
<( (Class II(A)). In addition, the U.K. decided that every ship to
CD uJ
0 n::
:.::
which the Merchant Shipping (High Speed Craft) Regula-
z tions 1996 apply insofar as it implements Chapter 2 Part B of
~
D III the High-Speed Code shall comply with the requirements of
the Agreement relating to specific stability standards. In the
w second case, the Stockholm Agreement requirements apply to
Cf)
0::: Cf) any Ro-Ro passenger ship to which SOLAS apply. As a result,
~a:::~ every Norwegian Ro-Ro passenger ship should comply with
cnOfo
zO-.:t the requirements of the SA on any voyage whether or not it
~o is within the geographical area of the SA. Furthermore, ad-
I-
• 00
(jj
U)
U)
CD
ditional requirements apply on the design of deck barriers. In
all cases, no pertinent information can be discerned address-
> ing specific issues pertaining to bringing either one- or two-
0)
"<!" compartment vessels in compliance earlier than the compli-
::J
I ance dates specified in Annex 2 to the Agreement. Also, it is
w
~
to be noted that France is the only EU country, which
U)
c
whereas it is not one of the signatories of the Agreement, it is
0
""§ partly affected by the latter, taking into account that a large
~
0 number of French vessels operating in the channel were to be
"O
0 modified to comply with SA requirements.
E
Inventory of passenger Ro-Ro vessels A(b)-General infor-
0
·ro mation on Ro-Ro passenger ships was collated, along with
E
relevant technical data, including information on compliance
0
c with relevant stability standards. The vessels were catego-
0
~ rized by A/Amax and operational H 8 • A comparison between
.0
·;:: the databases corresponding to North EU (NEU) and South
Cf)
en EU (SEU), respectively, led to the following conclusions: the
0
(9
z ca
NEU fleet is generally younger (Fig. 3) and has on the whole
U5 CD
higher stability standards than the SEU (Fig. 4); it is also
<(~ ci>
ufo "cf. u:: shown that the value of A/Amax of the North European fleet
LU N
(")
has risen considerably during the last five years, as a result
0
U5 of about 30% of the relevant vessels having already been
Cl upgraded to SOLAS '90 and to Stockholm Agreement stan-
dards by the beginning of the year 2000; by contrast, opera-
<( tional significant wave height values are generally evenly
distributed throughout the EU fleet and are marginally
higher in NEU than the SEU (Fig. 5, to be contrasted against

204 OCTOBER 2002 MARINE TECHNOLOGY

 6000
5000
~ 4000
=>
w
:.
.....
~ 3000
0
0
~ m ~ ~ ~ ~ m ~
~ ~ N N N N N ~
Ship sample
Fig. 7 Distribution of overall cost of upgrading (NEU-58 sampled vessels)
Fig. 10). As a general comment, it has been noted that the
experimental route to compliance with the Stockholm Agree-
ment is normally preferred (77% experimental route to 23%
calculation route based on 79 upgraded vessels), since opting
for this alternative enables shipowners to obtain a margin on
the attained H 8 for their vessels, without increasing the com-
plexity or cost of the upgrading. This margin is particularly
valuable to shipowners, as it is likely to influence positively
the resale value of their vessels.
Inventory of passenger Ro-Ro vessels upgrading and re-
lated costs A(c)-A comprehensive inventory was undertaken
of the technical modifications and adaptations carried out to
all ferries, which had to comply with the SA by 31/12/99 or
earlier, and of the associated costs. This part of the study
showed that although a large number of the ships affected by
SOLAS '90 and SA need major modifications in order to com-
ply with these stability standards (80% major, 13% minor
and 7% none based on 61 vessels), a good part of these is due
to the specific requirements related to SOLAS '90 standard.
Although this is not necessarily the case if larger values of
operational Hs need to be attained, in practice this indicates
that the main effect of the SA in the NEU has been to accel-
erate the schedule of compliance with SOLAS '90 require-
ments.
Distributions of major and minor modifications are shown
in Fig. 6 while the cost distribution is presented in Fig. 7, this
varying from €60k to €5.5M with an average of €2.lM per
vessel, based on 58 vessels.
Average SA compliance cost for NEU Ro-Ro vessels A(d)-
Studies aiming to establish average compliance cost per ves-
sel as a function of the operational sea state and A/Amax
category for a given vessel were undertaken, based on the
experience accumulated by the application of the SA in the
NEU. Statistical trends in this respect were established,
which provided a useful input for evaluating the extent of
modifications required by the SEU fleet. The results of this
task (Fig. 8), showed that: in the NEU there are compar-
atively more ships belonging to the lower upgrading cost
brackets than to the higher ones; there is good correlation
between average overall cost of upgrading and GDP per
capita (GDP per capita referring to the country in which the
ship was upgraded or-in absence of this information-the
country where the ship operates from); cost of upgrading and
A/Amax values are well correlated and since there is good
correlation between age and A/Amax, it is reasonable to use
the first as an indicator of a ship stability standard, all other
OCTOBER 2002
Average Overall Cost €1335
~ ~ ~ m ~ ~ ~ ~ m ~ ~ ~ ~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
data being unavailable; the variation of cost of modification
with ship size is generally best represented by a logarithmic
law, GT seemingly giving the best fit to the data available; it
is virtually impossible to detect a trend of variation of cost of
modification versus significant wave height. It is to be noted
that since the sample data available for achieving this task
were limited, the regression linking the cost of upgrading to
GT and A/Amax implied an unacceptably large error. For this
reason, this analysis was repeated and verified in greater
detail, as explained next.
Assessment of overall SA compliance cost for NEU Ro-Ro
vessels A(e)-In this task it has been attempted to further
demonstrate and better quantify the link between upgrading
cost and relevant parameters such as A/Amax and GT, con-
tinuing from the results presented in A(d) by using a sample
representing about 70% of the NEU fleet that needs to com-
ply with the SA. On this basis, statistical trends were estab-
lished, representative of the present status of the implemen-
tation of SA in the NEU, to be used for estimating on the
whole the possible effect of introducing the SA to SEU. Fur-
thermore, an estimate of the cost of the modifications still
required to complete the SA upgrading in NEU was provided.
In general, this part of the project corroborates all the find-
ings of part A(d), offering a better regression formula linking
A/Amax and GT to overall cost of upgrading (Fig. 9). More-
over, a detailed analysis of the cost of each type of modifica-
tion has also been attempted, leading to similar results in
terms of overall cost of modifications per ship. On the basis of
this analysis and estimating that about 28 vessels were still
undergoing upgrading in the NEU, the total cost of the out-
standing upgrading was calculated to be approximately
€11. 7M. This raised the total cost of upgrading of the NEU
fleet to about €85M, with 36% of the fleet not requiring any
upgrading and about 69% of the vessels having been up-
graded for less than €1.0M.
EU waters not covered by the
Stockholm Agreement
Key findings
As shown in Fig. 2, the study comprises five tasks, each
forming part of the methodology adopted as explained
therein and summarized below:
Survey of prevailing sea conditions and safety-critical local
MARINE TECHNOLOGY 205
 l/)
M

• 0 0 0
xewvrv
0 0
~
0

0
0
N
0
0
0
OJ
0
CD ..,.0 0
0
M ci ci ci ci ci

•• c
l/)
0

•
N
s
.D
(ii ·;::
en
:c
t:
Cl.)
Ol
Qi
I/)
I/)
:0
x
cu
z=
0 cu Cl)
N
a; > E
en "Cl <(
-=co:= ~
~
en
Cl.)
>
Cl)
ii <
;S < ui E
ca
-0
c
cu
>
=
= ~ x
cu 0
I/)
Ol
c
~ E '11:1' :0
~
<( ~
.9' < ::::>
w
Ol
Q_

-=
00 .c ~ 0
::::J

s co
<
Cl)
Ci ii en0
u:: E 0
$ ca
I/)
't:I

~
+x c.. co
r--- l/)
:c Ci
M
0
ci
II
(/J
u::
~N
er ~
II
>.
0

~ ~ 0
§ l/)

°'
ci ci
l/)
00
ci
00
ci
r--- 0r---
l/)

ci ci
l/)
'O
ci
~
ci 0
0
0
0
0
0
0 0 0 0

~
0 0 0
0 0
..,.
0 0 0 0
XBWV/V CD
"' "'
<mm3itl ISO:) lleJill\Q 0

d08/:)0\f
0 0

"' 0 0
OJ
0
CD ..,.
0
~

~ en
a;
0
~ cc en
en
Cl.)
::::J >
0 -0
u
f
Cl.)

£ Ci
en
a;
z ·~
c
E
cu
en
I/) 0
I/)
<D
>
"O
c
Ql
s I
0
'tj-

~
.D
·;::
"'O
<D
(/)
en .,., ::J
w
a.. "O
:0
en0
c:
c: ~
E c g I-
C'll c"' u t- CJ
I/) u:: -0
(!)
co <lJ ui
l.C> "O .~
>
c: (ii Ol
c
::> "'
-c:
Ql E :0
w .<:;
Qi 0c ~
~ z
-0
Ol
Q_

g c
cu 0
::::J

~
L1. en0 en0
0 0
::<'. cu
:::i co co
()

Ci Ci
0 0 0 0 0 0
u:: u::
0 0 0
0 0 0 0 0 0 0 0 0
..,.
"' ..,.
0
"'"' 0
"'"' 0
"' 0
"'
"' N 0
0
0
0
0
0
0
0
0
0
0
0 g
(o~n3itl ISO:) 0 0
..,.0 0 0 0
CD
"' "' N

<( (o~n3itl 1so:) neJaAo ()

conditions B(a)-The prevailing sea conditions and other values. Results of this study show (Fig. 10) that relevant
safety-critical local conditions in SEU geographical areas not significant wave heights (H,) in the Mediterranean are gen-
covered yet by the SA were investigated. The wave heights erally lower than 3.0 m, with the exception of the region west
were determined following two alternative, yet essentially of the island of Corsica where the obtained Hs was approxi-
complementary, approaches. In the first approach the main mately 3.25 m. However, larger wave heights, even exceeding
ferry routes between ports involving at least one SEU state 4.0 m, were noted in the Atlantic routes to Madeira and the
were analyzed. In the second approach, whole geographical Azores. Concerning other, possibly safety-critical local condi-
regions have been associated with characteristic wave height tions, such as wind, air and sea surface temperatures, vis-

206 OCTOBER 2002 MARINE TECHNOLOGY


4500
4000
3500
3000
2500
2000
1500
1000
500
o~~~
0 50000 100000 150000
GDP Per Capita*GT*(1-0rig. A/Amax)
Fig. 9 Rationalized overall cost of upgrading vs. GDP x GT x (1-A/Amax)
(NEU-84 sampled vessels)
ibility, traffic densities and other similar conditions it can be
concluded, based on the collected data, that the local sea
conditions are less safety-critical, when compared to the cor-
responding conditions in NEU waters, due to the higher av-
erage air and sea surface temperatures and the generally less
significant traffic densities in the pertinent local areas.
Inventory of passenger Ro-Ro vessels in SEU waters
B(b)-A comprehensive inventory was undertaken of Ro-Ro
passenger vessels operating in SEU, along with relevant
technical data, including information on compliance with rel-
evant stability standards. The vessels were categorized by
means of a variety of technical, stability sensitive character-
istics and economic indicators. It is to be noted that since
information on A/Amax values for several registered vessels
was very limited (not available or not reliable), the relevant
analysis was mainly based on the stability standard of com-
pliance, thus providing indirectly an indication of the actual
Fig. 10
OCTOBER 2002
A/Amax values of the vessels under consideration. Results
are shown in Figs. 3 and 4, where they are contrasted against
results from NEU vessels.
Ships to be upgraded to comply with SA B(c)-The scope of
this task was to establish which of the ships that operate in
SEU would need to be upgraded to comply with the provi-
sions of the SA and the possible extent of required modifica-
tions. Based on the inventory of the ships under investigation
(Task B(b)), their current stability standard of compliance,
area of operation (Task B(a)) and corresponding A/Amax val-
ues, the results provided a categorization of the affected
ships according to their current stability standards of com-
pliance, relevant A/Amax values and year of built or major
200000 modification. On this basis, it was established which ships
need to be upgraded in order to comply with the SA, the
extent of the required modifications in relation to relevant
provisions of the SOLAS regulations and the expected dates
of compliance (if formally the requirements of the presently
valid SOLAS '90 two-compartment standard are met inde-
pendently of a possible extension of the provisions of the SA
to EU regions not covered by it), (Fig. 11). Based on the tech-
nical characteristics and the area of operation of the affected
ships it was concluded that the techno-economic effort to up-
grade these ships to SOLAS '90, two-compartment standard,
would not deviate much from the effort required to ensure
compliance with the provisions of the SA.
Assessment of overall cost of SEU vessels for compliance
with SA B(d)-The objective of this task was to assess the
costs associated with the necessary modifications of SEU Ro-
Ro passenger ships, identified an analyzed under B(b) and
B(c), for compliance with stability requirements similar to
those of the SA. Taking into account that SEU Ro-Ro pas-
senger ships are generally operating in waters of comparably
lower Hs and also available scientific evidence indicating that
ships complying with the SOLAS '90 standard will survive
SOLAS damages of at least: 2.5 m H8 , as derived from model
Distribution of Hs in SEU
MARINE TECHNOLOGY 207
I\)
0
())

0
(")
-I
0
CD
m Regulation 8-1 Regulation 8-1 Regulation 8-2
:::0
I\)
0 Compliance of 2-comp std ships Compliance conditionally Compliance conditionally (Reg. 8-1)
0
I\)

Oct 1998 Oct 1998 Oct 2002 Oct 2000

8% 8% Oct 2004
Oct 2005 3%
0%
32%
Oct 2000 38% Oct 2010
24% Oct 2005
30% 34%
0%

Oct 2002 11% Oct 2002 Oct 2004 Oct 2008

19%
A 19% 11% 14%
Fig. 11a Ships with recorded A/Amax values (37 two-compartment standard ships out of 54 with conditional compliance)

Regulation 8-1 (simulated ships) Regulation 8-1 (simulated ships) Regulation 8-2 (simulated ships)
Compliance of 2-comp std ships Compliance conditionally Compliance conditionally (Reg. 8-1)

Oct 2000 Oct 2012

Oct 2005 Oct 2004 Oct 2005
5% 16%
3%
Oct 2000 14%
31%
Oct 2004 Oct 2010
14% 23%
Oct 2006
43%
s)> Oct 2002
:::0 52% 65% 25%
z
m
B
-I Fig. 11b Ships with simulated A/Amax values (101 two-compartment standard ships out of 148 with conditional compliance)
m
(")
::i:::
z
0
r
0
G'>
-<
 I GDP per capita, in kEUROI
source: Eurostat

25

20

15

10

0
Greece Spain France Italy Portugal
A
Fig. 12a GDP per country in SEU

South European Ro-Ro Passenger Fleet

Overnll cost of upgrnding

2000

1800

0 1600
ci::
;::I 1400
>.i;l
..:.::
1200
.5
1000
"'0
u 800

E <lJ
600
;..
0 400

200

B
Fig. 12b Overall cost of upgrading= 0.0195 x (GDP per capita x GT x (1-A/Amax)) + 212.4 ± 610.2

tests according to the "Equivalent Model Test Procedure" of
Resolution 14, SOLAS '95, it has been concluded that the
modification cost of SEU ships for compliance with the pro-
visions of the SA will be approximately the same as the
associated cost for compliance with the requirements of the
SOLAS '90 two-compartment standard. Based on the results
of a detailed cost analysis of modifications for the NEU ships
(task A(e)) and the derived regression formulas therein, the
A!Amax values and GT values of the inventory ships and
the GDP of the flag state, the itemized cost/ship as well as
the overall cost for the SEU ships has been deduced. Based
on this, the total modification for the whole SEU fleet
(264 ships) is estimated to range between a minimum
of €106M and a maximum of €250M. It is to be noted
that these estimates do not consider the possible removal
from service of aged SEU ships, which is to be expected
since it might prove economically more advantageous for
shipowners to replace some of these ships with new build-
ings instead of undertaking onerous extensive modifications.
(See Fig. 12.)
Assessment of modification time for compliance of SEU
vessels with SA B(e)-The objective of this task was to assess
the time required to execute the necessary modifications
for the affected SEU ships, identified and analyzed under
B(b), B(c) and B(d), considering the capacity of European
shipyards, anticipated delivery times and the need to en-
sure continuity of service. Taking into account the fact
that the process of upgrading the affected ships is not a
continuous function of time and that the relevant shipping
companies will rather choose to wait until it is absolutely
necessary to modify ships, it is concluded that the time re-
quired for the modifications will be strictly following the
"phase in" procedure for compliance with the provisions of
Stockholm Agreement, to be decided by the European Coun-
cil. Therefore, the present task has been based on the as-
sumption of an accelerated compliance schedule for the af-
fected SEU ships with the full provisions of SOLAS '90 (Reg.
8-1 and Reg. 8-2) based on the deduction outlined in B(d)
above. The assumed time schedule, ranging from 1 October
2002 for ships with lower values of A!Amax, to 1 October
2005 for those in the highest A!Amax category, appears fea-
sible in all respects, as this compliance schedule does not
deviate from the existing compliance with Regulation 1 of
SOLAS '90 (provisions for one compartment standard com-
pliance). More importantly, this holds true for the large ma-
jority of existing vessels (78.1 %, 235 out of 301 existing
ships), whereas for the remaining ships already complying
with Regulation 1, SOLAS '90 (21.9%, 66 out of 301 ships) the
impact is considered to be less severe and feasible within the
set accelerated time schedule. From the point of view of avail-
ability and capacity of European shipyards in order to accom-
plish the requested modifications and the seamless continu-
ation of service, it can be concluded that, since the time
schedule for compliance with the provisions of SOLAS '90 is
practically unchanged, no additional negative effects would
result from the introduction of Stockholm Agreement in

OCTOBER 2002 MARINE TECHNOLOGY 209

SEU. However, the feasibility of the first compliance date
being 1 October 2002 would need to be critically examined,
considering that 59% of the ships identified for upgrading
would be affected.
Concluding remarks
The North-South divide concerning safety of Ro-Ro passen-
ger ships continues to trouble shippers and regulators alike
and a way forward is actively being sought. Serving this
need, an SSRC-NTUA partnership has undertaken on behalf
of the European Commission a study to assess the impact of
the Stockholm Agreement on the areas covered by it with the
view to evaluating the likely impact of introducing it to areas
not covered by it. This introspection on the Stockholm Agree-
ment led to the following key conclusions:
• SOLAS '90 ships appear to be capable of surviving sea
states of at least 2.5 m Hs and that SOLAS '90 is a "good"
standard reflecting meaningfully the safety of Ro-Ro
vessels.
• The Stockholm Agreement appears to be unrealistically
stringent, in general demanding levels of safety well be-
yond those determined through performance-based
methods and, at times, simply unattainable.
• Considering that a SOLAS '90 ship survives 2.5 m Hs
and accounting for the comparative lower Hs distribu-
tion in South European waters, it would appear that the
upgrading cost to the Stockholm Agreement would not
be dissimilar to the cost and time for upgrading to
SOLAS '90 two-compartment standard.
• Projections based on the upgrading experience and in-
curred costs in North Europe reveal that the estimated
210 OCTOBER 2002
overall cost of upgrading the South European fleet
would be between €106M and €250M.
Acknowledgments
The authors would like to express their gratitude to the
European Commission DG Transport for the financial sup-
port of the research described in this paper under Contract
No. B99-B2702010-SI2.144738. The work was undertaken by
two collaborating teams: the Ship Stability Research Centre
team of the Universities of Glasgow and Strathclyde, com-
prising Prof. D. Vassalos, Dr. 0. Turan, Dr. L. Letizia and Dr.
D. Konovessis; the Ship Design Laboratory team of the Na-
tional Technical University of Athens, comprising Prof. A.
Papanikolaou, Asst. Prof. K. Spyrou, Ms. E. Eliopoulou and
Ms. A. Alissafaki.
References
1. IMO Resolution 14, "Regional Agreements on Specific Stability Re-
quirements for Ro-Ro Passenger Ships"-(Annex: Stability Requirements
Pertaining to the Agreement), adopted on November 29, 1995.
2. IMO Resolution 14, "Regional Agreements on Specific Stability Re-
quirements for Ro-Ro Passenger Ships"-(Appendix: Model test method),
adopted on November 29, 1995.
3. Vassalos, D., Pawlowski, M., and Turan, 0., "A Theoretical Inves-
tigation on the Capsizal Resistance of Passenger/Ro-Ro Vessels and Pro-
posal of Survival Criteria,'' Final Report, The Joint North West European
Project, University of Strathclyde, Department of Ship and Marine Tech-
nology, March 1996.
4. Vassalos, D., Turan, 0., Letizia, L., and Konovessis, D., "Im-
pact Assessment of Stockholm Agreement on EU Ro-Ro Passenger Ves-
sels Covered by It,'' (B99-B2702010-SI2.144738), Final Report Part I,
NAME-SSRC, March 2001.
5. Papanikolaou, A., Spyrou, K., Eliopoulou, E., and Alissafaki, A.,
"Impact Assessment of Stockholm Agreement on EU Ro-Ro Passenger
Vessels Not Covered by It," (B99-B2702010-SI2.144738), Final Report
Part II, NTUA-SDL, March 2001.
MARINE TECHNOLOGY