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

net/publication/237803246

Design aspects of survivability of surface naval and merchant ships

Article · January 1998

DOI: 10.1016/B978-008043652-4/50043-2

CITATIONS READS

7 200

2 authors:

Apostolos Papanikolaou Evangelos Boulougouris

Hamburgische Schiffbau-Versuchsanstalt GmbH University of Strathclyde
387 PUBLICATIONS   2,319 CITATIONS    130 PUBLICATIONS   416 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

HARDER View project

Naval Ship Design View project

All content following this page was uploaded by Apostolos Papanikolaou on 29 January 2018.

The user has requested enhancement of the downloaded file.

 DESIGN ASPECTS OF SURVIVABILITY OF
SURFACE NAVAL AND MERCHANT SHIPS

APOSTOLOS PAPANIKOLAOU1 and EVANGELOS BOULOUGOURIS(2) ,

1
Professor, Head of Ship Design Laboratory, National Technical University of Athens,
Department of Naval Architecture & Marine Engineering, Greece, [email protected]
2
PhD. cand., National Technical University of Athens, Department of Naval Architecture &
Marine Engineering, Ship Design Laboratory, Greece, [email protected]

Abstract. The paper addresses various design aspects of survivability for surface naval and
merchant ships through a common rational probabilistic methodology. It is based on earlier
work of Kurt Wendel on a probabilistic approach to the damage stability of surface ships.
This method was adopted by SOLAS 1974 through Res. A.265 (VIII) as an alternative to the
deterministic SOLAS criteria for passenger ships and finally by SOLAS 1992 for the
evaluation of stability of new cargo ships. The method is currently under review by IMO for
application to all types of merchant ships in the framework of harmonization of all existing
stability rules. Recently the method found access into the design process of modern naval
ships. The declining defence budgets world-wide and the reduced manning requirements
bound the size and thus the payload of modern naval ships, required in addition to operate in
an increased dangerous warfare environment. This stressed the need for the introduction of a
new naval ship design philosophy, namely the enhanced survivability. Most designer
decisions, associated with survivability, as compartmentation and arrangements, are taken at
the preliminary design stage and are very difficult and costly to change, if at all, in latter
stages. Therefore, a proper guidance in the preliminary design stage would greatly help the
designer to achieve his goals. The paper addresses the fundamental aspects of survivability
and introduces this relatively new probabilistic approach for assessing the damage stability
and survivability properties of both naval and merchant ships.

Keywords. Survivability, damage stability, probabilistic stability method, A.265-IMO,

vulnerability, lethality, susceptibility, modelling, simulation.

1. INTRODUCTION

One might wonder what is common between the design of a naval surface combatant and a Ro-
Ro passenger ship. The answer is simultaneously easy and complex: they both have to survive
in case of damage, for obvious reasons, however under quite different constraints, external
damage threats and environmental conditions. For surface combatants, although the probability
of damage is very high and it should be a significant factor in their design, there was never
before a systematic examination and assessment of their survivability. On the other hand, many
tragic events occurred recently in passenger shipping, turning the attention of the public to the
inherent safety of Ro-Ro Passenger Ferries. The public outcry stressed the need for enhancing
the inherent survivability of this type of ships in case of damage by efficient design measures.
In the following, a consideration of possible common methodologies for the survivability
analysis of these two totally different ship types is attempted.

The Naval Ship Dimension

Modern naval warfare is characterized by highly sophisticated weapon systems. Surface

combatants are threatened by air, surface and underwater weapons guided with various
sensors: radar, infrared, electroptic, or laser. In order to accomplish their mission they have to
carry a large arsenal and a complicate suit of advanced (but nevertheless sensitive)
electronics. These have increased their acquisition and operational cost and reduced the size of
the fleets operated by various Navies. Then again the need for a high payload to displacement
ratio has driven the designers to a reduction of the shell plate thickness for keeping the
structural weight as low as possible. This resulted to a shift from enhanced armour to sensor
capability, making naval ship designs were more vulnerable as could be proven in the
Falkland Islands Conflict, in the USS Stark (FFG-31) incident and during the Gulf War Naval
Operations. It becomes obvious that an effective solution to this problem is the adoption of a
rather new design philosophy, namely Design for Enhanced Survivability.

Nature makes its creatures adaptive to their environment for survival. In the same way, ships
should be designed with an inherent ability to survive in the threat environment they have to
operate. For naval ships survivability is the capability to continue to carry out their missions in
the combat lethal environment. This is a function of their ability to prevent the enemy from
detecting, classifying, targeting, attacking or hitting them. The inability to "intercept" any of the
above threats is a measure of their susceptibility, mathematically expressed by the probability
PH. On the other hand the degree of impairment she suffers in case of damage characterizes her
vulnerability, expressed by PK/H respectively. The product of susceptibility and vulnerability
defines the killability PK of the combatant. It is obvious that in order to maximise the naval
ship’s survivability we have to minimise its susceptibility and vulnerability. Mathematically
the above can be expressed by the following global formula:

PK = PH × PK / H (1)

Thus the probability of survival S is expressed by:

S = 1 − PK (2)

The susceptibility of a naval combatant is dependent on its signature characteristics. Signature

reduction measures will decrease the probability of being detected and classified. These
include the suppression of the RCS, IR, noise, magnetic and electro-optical signature.

The vulnerability reduction measures must be addressed in the early design phase in order to
maximise the results. They include arrangements, redundancy, protection, and equipment
hardening as well as damage containment.

If we restrict our analysis only to conventional (high explosive) anti-surface weapons then
there are two main damage effects that threat the survival of a combatant: flooding and fire.
Though both are equally essential we will limit our survivability analysis herein only to the
first one. The reason is first of all that the second aspect (fire) can be effectively performed
only at advanced stages of design. Besides for any ship we have to counter a fire onboard, it is
assumed that it is staying afloat and upright. Mathematically this means that we consider
flooding and fire as independent events:

PK[Hit ∩ (Flooding ∪ Fire)] = PK[(Hit ∩ Flooding) ∪ (Hit ∩ Fire)] =

=PK[Hit ∩ Flooding] + PK[Hit ∩ Fire] - PK[Hit ∩ Flooding] × PK[Hit ∩ Fire]

which also means that in the following we will be assuming that the probability of loss after a
hit due to fire, given the progressive flooding due to the same hit, is zero, i.e. PK[(Hit ∩ Fire) |
(Hit ∩ Flooding)]=0. Therefore for the rest of this paper we will be referring to the
survivability of naval ships by meaning just flooding survivability and by calculating the
PK[Hit ∩ Flooding].

The Passenger Ship Dimension

Since the early seventies (SOLAS 1974) the International Maritime Organisation (IMO) has
adopted probabilistic methodologies for the assessment of survivability of passenger ships. We
refer to the regulation A.265 (VIII) of IMO setting an equivalent to the deterministic stability
criteria, namely part B of Chapter II of the International Convention for the Safety Of Life At
Sea, 1960 (SOLAS 60). The vast number of the required calculations for a full probabilistic
assessment of a ship under consideration was, in those days of limited computer hard- and
software, a serious drawback that led to only limited applications to actual ships. This was one
serious reason for the further development of the deterministic criteria (SOLAS 90, 92 & 95),
whereas the results of the probabilistic approach (or a simplified version thereof) were only
used as an indicator for the implementation of new regulatory schemes to existing ships (phase-
in procedure). However, it is a taken decision of IMO to formulate 1 and approve a
"harmonised" new probabilistic stability framework for all types of ships, possibly by the
SOLAS conference in the year 2000.

According to IMO Res. A.265, and following the fundamental concept of K. Wendel (1960),
there are the following probabilities of events in the framework of damage stability:

1. The probability that a ship compartment or group of compartments may be flooded

(damaged), p.

1
See, SLF39, SLF 40, SLF 41, forthcoming SLF 42
 2. The probability of survival after flooding the ship compartment or group of
compartments under consideration, s.

The total probability of survival is expressed by the attained subdivision index A which is
given by the sum of the products of pi, and si for each compartment and compartment group, i,
along the ship's length:
A = ∑ pi ⋅ si (3)
i
The regulations require that this attained subdivision index should be greater than a required
subdivision index R, which is determined by the number of passengers the ship is carrying and
the extent of life-saving equipment onboard. This value is a measure for the ship’s probability
of surviving a random damage and it is obvious that this value increases with the number of
passengers onboard the ship. The factors in the formula determining R are so selected to
correspond to the mean values of the attained subdivision indexes of a sample of existing ships
with acceptable stability characteristics. This is, of course, a point for lengthy discussions,
because the safety standards of passenger ships have significantly changed over the years,
therefore the basis for the evaluation of R must be updated to account for these changes.

Increasing the survivability of the RoRo passenger ships requires the specification of the
"threats" (or better the "risks") they have to counter. Similar to naval vessels, passenger ships
face mainly two major threats: Flooding and Fire. It is obvious, that for A.265 the survivability
is virtually identical to the vulnerability in case of flooding. The regulation does not take into
consideration either the vulnerability in case of fire or the susceptibility of the ship. However,
it is formally not difficult and probably advisable to incorporate susceptibility to the
survivability of Ro-Ro passenger ships, as it has been suggested for naval ships. This concept
is more or less adopted in the formulation of the "Safety Assessment", or "Formal Safety
Assessment (FSA)”, Spouge (1996). This way we can formulate a unified rational scheme
model (methodology) for assessing the survivability of naval surface combatants, passenger
Ro-Ro ships and any other type of ship. The only difference would be the type of the
anticipated risks each type of ship has to counter. For passenger ships, and merchant ships in
general, we should be considering flooding due to one of the following impact events, Spouge
(1996):

1. Ship to ship (collision) 5. Terrorist act

2. Ship to berth/breakwater (contact) 6. Material Failure
3. Ship to bottom (grounding/stranding). 7. Human Error
4. Explosion

The probability of a passenger ship loss in case of flooding or fire is calculated by the formula:

PL [Flooding ∪ Fire] = PL [Flooding] + PL [Fire] - PL [Flooding] × PL [Fire] (4)

As has been noted above, we should herein discuss only the probability of loss in case of
flooding, namely PL [Flooding].
2. OUTLINE OF POSSIBLE SHIP DAMAGE CONSEQUENCES

It is obvious that between the intact condition and the total loss of a ship there are many
intermediate stages. Though these stages can be defined in various ways a very common
procedure is the one relating them to a functional hierarchy, Ball & Calvano (1994).

According to this in case of a naval ship we may have, in descending order, one of the
following damage extents:

• Total Kill when the ship is considered lost entirely because of sinking (foundering) or
completely damaged by fire (or other phenomenon).
• Mobility Kill if immobilisation or loss of controllability of the ship occurs.
• Mission Area Kill if a mission area (e.g. AAW capability) is considered lost for the ship.
• Primary or Combat System Kill in case of one or more vital systems of the ship, such as a
propulsion engine or a CIWS, are damaged.
• Hull, Machinery or Electrical (HM&E) Support System Kill if one or more components
supporting a primary/combat system of the ship are damaged (e.g. the cooling water
system).

Apparently a combat system kill can lead to a mission area kill or a mobility kill or even a
total kill. Likewise a mission area kill may decrease to a combat system kill after the crew
makes necessary repairs. Our primary target is to confine damage extent to the lowest possible
level. Accordingly, in case of a Ro-Ro passenger ship we may have:

• Loss of stability (Intact or damage ship capsizing)

• Loss of floatability (progressive flooding, foundering)
• Loss of power
• Loss of mobility (controllability)

Loss of stability and of floatability, though they may both result to the foundering of the ship,
they can be addressed separately, because of the different time scales available for evacuation
of the ship. With the loss of power and controllability we refer herein to a damage of the ship's
main machinery and vital equipment compartments, and not of an isolated failure incident of the
ship's equipment.

3. FUNDAMENTALS OF NAVAL SHIP DESIGN

Compared to a merchant ship, the design of a naval ship is a very complicate task. There is a
vast number of requirements -many of them contradicting- and additionally a large number of
constraints. Our effort to develop ships of enhanced survivability imposes further constraints to
the naval ship design. This means penalties associated with cost, weight and impact on other
features. It is generally recognised that during a naval ship development program, though the
expenditure at the early design stage amounts only 6% of the total cost, related decisions
concern 80% of the final realisation cost and should be taken with great care, Van Hees
(1997). Thus it is important in the early stages of design to have a methodology to assess the
survivability of the proposed vessel, to identify shortcomings and to explore feasible
improvements. Furthermore, it seems that simulation methods and the development of virtual
demonstrators will be very handy to ship design in the 21st century, (US National Research
Council study). Within such approaches, the Survivability Performance Analysis is an
indispensable design tool, to be briefly addressed in the following.

The assessment of a naval ship’s the survivability requires the identification of the major
threats it has to counter. Considering that conventional weapons are the most widely used, we
can focus on the consideration of a threat posed by a radar guided missile. The same type of
analysis can obviously be modelled for other type of weapons, sensors and threats in general.

4. SURVIVABILITY PERFORMANCE ANALYSIS FOR NAVAL SHIPS

This analysis is based on the modelling of the event sequence from the enemy's arrival to ship’s
operational area up to the moment at which a hit might strike the vessel. Thus we have the
detection, classification, target acquisition and launch of an enemy attack. The ship’s response
is to jam, attempt to deceive, or to destroy the enemy's incoming weapons. The probability of
ship's detection is a function of the threat's sensor, its range and ship’s signature. A first
estimation of the RCS of a surface ship can be taken from the formula, Rains (1994) :

σ = 52 ⋅ f ⋅ 3 Disp 2 (5)

where ó is the ship's RCS in m2, f is the incident radar frequency in MHz and Disp is
ship's displacement in tons. The range of ship’s detection by the enemy's radar is given by the
equation, Goddard, Kirkpatrick, Rainey & Ball (1996):

1/ 4
 P G 2λ2σ 
R max =  T 3  (6)
 (4π) Pmin 

where Rmax is the maximum detection range, PT the transmitter’s power, G the antenna gain, ë
is the radar’s operating wavelength, ó is the ship’s Radar Cross Section (RCS) and Pmin is
the minimum detectable received signal from the enemy's sensor. Obviously the lower the RCS
of the ship, the closer the enemy has to come for detecting it. An optimisation of the RCS is
nowadays possible by application of STEALTH technology. Pmin depends on the enemy’s radar
characteristics and also on the environmental conditions. By the later we mean temperature, sea
condition as well as jamming. Increase of any of these parameters results to an increase of Pmin
and eventually decrease of the Rmax.

Following the vessel’s detection it is up to the enemy to decide on the tackle of the target he
has picked up. In case he does, classification, targeting and the release of missiles will follow.
Assuming that the missile is radar-guided, its course to the target will also depend on ship’s
RCS. The path, it will follow, depends very much on its accuracy. Their Linear Error
Probability (LEP) can express this property for weapons engaging surface combatants.
Knowing (or assuming) missiles' LEP we may assume that the missile’s position relatively to
ship’s profile follows a normal distribution with standard deviation σ, related to the LEP by
the formula, Przemieniecki (1994):

LEP = 0.6745σ. (7)

At this phase the ship will try (assuming it knows that it is under attack) jamming the missile’s
radar. Because of its higher power, the ship’s jamming device will block the missile’s radar
until it reaches a certain distance from its target. This distance depends on the power ratio
between the radar and the jamming device. Once the missile regains a lock on the ship it
depends on its aerodynamic characteristics (i.e. maximum turning acceleration and speed)
whether it will turn to the vessel’s direction. If not, it will miss the target vessel. To be
successful, the missile's minimum turning radius has to be less than its distance from the ship at
that moment, namely, Rains (1994):

2
Vm
Missile Radius = ≤ R regain (8)
N ⋅g

where Vm is the missile’s velocity, N the maximum turning acceleration of missile in [g] and g
the gravitational acceleration. The range at which the missile will regain a clear picture of the
ship’s location is given by the formula, Rains (1994):

PM σ
R regain = ⋅ (9)
PJ 4π

where PM/PJ power ratio between the missile seeker and the jammer. Thus the effectiveness
of jamming can be expressed by the integral of the normal distribution from the ship’s either
end to a distance RmaxRregain towards the centre of the ship.

The next step in the above ship-missile struggle includes the use of decoys. Their performance
depends also on their RCS. If it is higher than that the ship’s it will allow the deception of the
threat. If it is the same, there is a 50% probability that the missile will follow the decoy. In
case of the ship’s arsenal includes a “close in” weapon system (CIWS) it will try to destroy the
incoming missile. Its probability of success is usually known from trial experiments and it can
be incorporated into the model. If there are also missiles that can be used against the incoming
threat, their contribution must be also included.

Computers can estimate the above relatively easy. The problem is that in order to assess the
survivability of a ship design we have to estimate the ship’s vulnerability and in particular her
single hit kill probability. Therefore we have to calculate:

• the probability of “hit of a particular point of the ship”.

• the probable “damages extent given a hit at that point”.
• the probability of “ship’s survival given the hit point and extent”.

5. VULNERABILITY - ADVANCED PROBABILISTIC DAMAGE STABILITY (APDS)

FOR NAVAL SHIPS

The impact of a hit can be at any point of the ship’s length. The target point of the missile
guidance system depends on its type, sensor type and guidance system characteristics. For
instance, an Exocet missile will aim at the ship’s waterline, while a Harpoon missile will try
to enter the ship higher at the ship's superstructure. Likewise, the longitudinal point of impact
will depend on the shape that the signature of the ship presents to the particular threat sensor.
An IR (infrared) missile will target at the ship’s machinery that is an intense hit source. On the
other hand a radar-guided missile will probably aim at the ship’s radar image centre. For
simplicity and generality, we may assume that the impact location is described by a normal
probability distribution with its centre at the ship’s centre and a linear error probability (LEP)
equal to 0.5⋅LWL.

The damage extent can be taken from a Log-Normal Damage Function. This is given by the
function, Przemieniecki (1994):

r
1  ln 2 (r / a )
d(r ) = 1 − ∫ exp −  ⋅ dr (10)
0 2π β r 
 2 β 2


where:
á = (RSKRSS)0.5 RSK "dead-sure kill radius"
1 R RSS "dead-sure surviving radius"
â= ln( SS )
2 2zSS R SK

RSK and RSS correspond to 98% and 2% probabilities of damage respectively. Their values can
be derived from empirical data for the threat missiles considered. Herein we will assume a
RSK diameter equal to the U.S. Navy standards namely 15%LBP, Surko (1994). The "sure-save"
diameter will be taken as 0.24⋅L as in the A.265 IMO-SOLAS regulations for merchant ships.
This results in a ratio of RSK/RSS equal to 0.625. The variation of the lognormal damage
function is shown in fig. 1. Alternatively a diameter depended on warhead’s weight can be
used.

0.12L 0.075L
 Figure 1: Variation of log-normal damage function

Having defined the first two probabilities namely hit at a particular point and damage extent
given the impact point, we are left to define the probability that the ship will survive given the
damage location and extent. A rational methodology for this evaluation can be based on the
survival criteria of the U.S. Navy, Surko (1994), considering that the latter:

• are based on extensive real damage incidences during the great naval battles of the WWII.
• have been proved quite reliable, because they refer to a large number of events during
almost 40 years of use. Even though several ships of the U.S. Navy suffered serious combat
damages, none of them was lost due to the lack of stability or floatability.
• include dynamic effects though in a “quasi-static” way.

The philosophy for transforming these deterministic criteria into a set of rational probabilistic
approach criteria will be based on A.265 IMO-SOLAS regulations for merchant ships.

It is well established that in all relevant criteria there is an underlying assumption that the sea
conditions at the time of damage are "moderate". This constraint could be lifted if we knew (or
we could assume) a specific operational sea spectrum for the ship design in question. Thus, we
could calculate the probability that waves will not exceed the wave height considered as basis
for the current deterministic U.S. Navy criteria, namely a significant wave height of 8 ft. This
wave height was used for the determination of èroll, namely the roll amplitude due to wave
action. It was also the underlying assumption behind the guidelines for establishing the
watertight features/closures to prevent progressive flooding. Thus, any attempt to change the
wave amplitude must take into account changes in both èroll as well as the margin line. Another
important environmental parameter is the wind speed. Given the small probability of exceeding
the values given by the U.S. Navy standards (namely, about 33 knots for a 3500 tons frigate),
this value could be left unchanged. As a first step we can propose the following guidance for
the formulation of survival criteria to be applied in the frame of a probabilistic approach to
the survivability of naval ships given in table 1. Figure 2 explains the meaning of the various
angles on the righting arm curve. For intermediate stages, interpolant values could be used.
 Figure 2: Survival Criteria

It is obvious, that some systematic experimental and theoretical work is needed in order to
specify in a more rigorous way the calculation of the S value for naval ships. In any case, the
calculation of the probability distributions of wave exceedence in the area of operation is
necessary. For instance, the P(HS<8 ft)= 0.60 holds for the North Atlantic, but P(HS< 8ft)= 0.90
for the Mediterranean Sea. Thus, a combatant, meeting the U.S. Navy criteria, should have,
according to the criteria formulated above, a 60% probability of survival for any 2-
compartment damage in the North Atlantic and a 90% probability of survival in the
Mediterranean Sea (Papanikolaou et. al. (1995), Athanassoulis, Skarsoulis (1992)).

TABLE 1
PROPOSED CRITERIA

S=1 èroll = 25 deg Wind speed = accord. to DDS-079-1.

A1 ≥ 1.4 A2 Min Freeboard ≥ 3in + 0.5×(HS(0.95)-8 ft)
S = P (hS=8 ft) The ship meets the current (deterministic) damaged stability criteria of the
U.S. Navy.
S=0 èroll = 10 deg. Wind speed ≤ 11 knots
A1 ≤ 1.05 A2 Margin line immerses.

The above procedure for assessing the damage stability component of the vulnerability could
form an important element of a new Advanced Probabilistic Damage Stability method for
naval ships. It makes possible the assessment of damage cases, where multiple hits of non-
adjacent compartments can occur. This increases the flexibility of the designer. It makes also
the damage stability of the vessel a property of the design and not just a requirement that has to
be met later on. Its integration with the Survivability performance analysis described above is
illustrated in the following flowchart.
 Initial Hull form Probability density Damage Range
General
START function of aiming of given
Arrangement point along the ship warhead
Combat Outfit

Calculation of
LEP of threat Probability of Ship Hull
SHIP SURVIVAL
Damage at any Survival against
PROBABILITY
Navy longitudinal point Capsizing and
Threat
Requirements Sinking
Assessment

Naval Advanced
Damaged
Stability
Requirements

Figure 3: Survivability Performance Analysis flowchart

6. THE RO-RO SHIP DESIGN PROCEDURE

The existing probabilistic method A.265 is a rational but complicate and non-transparent
procedure for assessing the probability of survival of a ship. On the other hand, the ship
specific survivability characteristics should be reflected in a clear way in its survivability
index for possible design optimisation. By use of modern hardware and software computing
tools, an optimisation appears today feasible, as it will be outlined in the following.

As far as the possible risks for passenger ships are concerned, it has been noted at the stage of
definition of the various possible damage cases for Ro-Ro vessels that multiple stages of
damage prior to the total loss of ship can be considered. The probability of loss of
controllability, or of power (black out) should be actually also considered. Piping transferring
cold water for cooling vital machinery or power transfer lines passing through the damaged
compartment should be identified. The probability of their damage has to be considered and
properly included in the calculations. Additionally, the time required to evacuate the ship must
also be taken into account. This would encourage ship designers to consider arrangements and
efficient equipment for the safe evacuation of passengers and crew, beyond the least
requirements set by the regulations. Recent work by D. Vassalos et al. (1998) stresses the need
for assessing the available evacuation time in case of damage and prior to capsizing or
foundering. Existing regulations specify the maximum time available for evacuation but no
special care is given to consider the ship damage conditions that might affect this time. Large
heel or trim may increase the evacuation time significantly especially for children, elderly or
persons with mobility problems. Thus damage cases with large heel or significant trim have to
receive reduced weights even though the stability requirements are met.

The consideration of the particular significance of the various ship spaces can be incorporated
in the regulations by assigning them so-called "vitality" coefficients as it has been already
done with the variation of space permeability. Thus spaces with vital machinery or equipment
will receive more weight in the survivability analysis. This will account for the fact that even
though the ship stays afloat, in case of damage, the associated risks are greater compared to the
same post-damage condition after damage of non-vital spaces. The Attained Subdivision
Coefficient formula should be therefore modified as following:
 A = ∑ pi ⋅ si ⋅ w i (11)
i

where wi is a properly defined vitality coefficient for compartment i. The vitality coefficients
(wi ≤ 1.0) of the various compartments should be in descending order to their significance. As
defined above, vitality of course depends on the existence of redundant equipment. Also, if for
a damage case involving multiple compartments, both the primary and the emergency
equipment are contained in the damage extent, the vitality coefficient has also to be reduced.
The above procedure calls for the introduction of innovative machinery and equipment
arrangement concepts (multiple machinery spaces, redundancy of equipment, etc.) as, e.g., has
been indicated in recently published work by M. Kanerva (1997). Specific values cannot be
proposed in this paper because more work has to be done in this area. It should involve both
damage statistics and actual machinery arrangements of existing ships. The above revision of
the Attained Subdivision Coefficient formula calling for special consideration of vital spaces
can be incorporated in a more general Ro-Ro ship design optimisation scheme, as outlined in
the attached chart. The procedure considers the optimisation with respect to the Local
Subdivision Index, as proposed earlier by Sen & Gerigk (1992) and properly modified, as
given above to account for the vitality of spaces.

7. CONCLUSIONS

The paper addressed the survivability of naval and merchant ships (here: Ro-Ro passenger
ships) through a common ("harmonized") rational probabilistic procedure. Possible risks for
naval and passenger ships have been outlined. Special attention has been paid to the
formulation of survival criteria for naval ships. At present, at the authors' knowledge, all
known damage stability criteria for naval ships are deterministic. They all assume a specific
extent of damage and require that the ship achieve certain values of metacentric height, or
maximum righting levels (GZ), at certain heel angles. The only probabilistic method known to
the authors at this stage is that of the German Navy, but only for those ships that do not satisfy
the set deterministic criteria. Therefore the authors suggest, herein, specific survival criteria
for naval ships, to be included into a formal probabilistic assessment of survival of naval
ships. Finally, following general design concepts of naval ships, a modification of the formula
of the Attained Subdivision Index of passenger ships has been proposed accounting for the
vitality of specific ship compartments, securing the powering, electric supply and
controllability of the ship after damage. The increased importance of modelling and simulation
methods in modern ship design requires the development of more rational concepts. Though the
above methodology is, at this stage, not complete, it is suggested as a first step to a new design
direction. It is of course obvious that much work is still needed to test and to fully verify the
proposed procedures. A framework for discussion of the suggested design procedures for Ro-
Ro passenger ships is the Thematic Network SAFER-EURORO, Vassalos et. al. (1997),
supported by DG XII of the European Commission. Their financial support is herein
acknowledged.

REFERENCES
 Athanassoulis G., Skarsoulis M. (1992). Wind and Wave Atlas of the North-Eastern
Mediterranean Sea. NTUA-SMHL Publ., Athens.
Ball R. E. & Calvano C. N. (1994). Establishing the Fundamentals of Surface Ship
Survivability Design Discipline. Naval Engineers Journal, January, 71-74.
Goddard C. H., Kirkpatrick D. G., Rainey P. G. and Ball J. E. (1996). How much STEALTH?.
Naval Engineers Journal, 108:3, 105-116.
Kanerva M. (1997). Fundamental Rethinking of Passengership Design for Economic
Operation by Owners and Construction by Shipyards. Cruise & Ferry 97, London.
Papanikolaou A. D. et al. (1995). Study on the Practical Implications of the Proposed New
SOLAS Regulations on Existing Greek Ro-Ro Passenger Ships and Critical Review of the
Proposed New Regulations. NTUA Report, Athens.
Przemieniecki J. S. (1994). Mathematical Methods in Defense Analyses. AIAA, Washington,
DC.
Rains D. A. (1994). Methods For Ship Military Effectiveness Analysis. Naval Engineers
Journal, March, 126-135.
Sen P. & Gerigk M. (1992). Some Aspects of a Knowledge-Based Expert System for
Preliminary Ship Subdivision Design for Safety. Proc. PRADS'92 Vol. 2, 1187-97.
Spouge J. (1996). Safety assessment of passenger/Ro-Ro vessels. Int. Seminar n the Safety of
Passenger Ro-Ro Vessels, RINA, London.
Surko S. W. (1994). An Assessment of Current Warship Damaged Stability Criteria. Naval
Engineers Journal, May, 120-131.
Technology for the U.S. Navy and Marine Corps, 2000-2035:Becoming a 21st Century
Force. National Research Counsil study.
Van Hees M. T. (1997). Quaestor: Expert governed parametric model assembling. PhD
Thesis, Technische Universiteit Delft.
Vassalos D. et al. (1997), The Thematic Network SAFER-EURORO, Ro-Ro Design for
Safety. European Commission Network, DG XII, 1996-2000.
Vassalos D. et al. (1998). Time Based Survival Criteria for Ro-Ro Vessels. The Royal
Institution of Naval Architects, Spring Meetings.
Wendel K. (1960). Die Wahrscheinlichkeit des Ueberstehens von Verletzungen. Journal
Schiffstechnik, 47-61.

View publication stats