Stability, safety and operability of small fishing vessels

Francisco Mata-Álvarez-Santullano a,n, Antonio Souto-Iglesias b,1

a Maritime Accident and Incident Investigation Standing Commission (CIAIM), Ministry for Development and Transport, Government of Spain, Paseo de laCastellana 67, 28071 Madrid, Spainb Model Basin Research Group (CEHINAV), Naval Architecture Department (ETSIN) Technical University of Madrid (UPM), 28040 Madrid, Spain

a r t i c l e i n f o

Article history:Received 4 June 2013Accepted 18 January 2014Available online 7 February 2014

Keywords:SafetyOperabilityFishing vesselSeakeepingIntact stabilityShip motions

a b s t r a c t

In this paper, the relationship between stability, safety and operability for small fishing vessels isinvestigated. To this aim, a relevant set of small fishing vessels is selected. These have similar maindimensions and capsized in stability related accidents between 2004 and 2007. The stability andoperability characteristics of such vessels are confronted with those of the vessels that had beendecommissioned in order to build them, operated by the same crews, in the same areas and using thesame fishing gear types. Such vessels are considered as reference safe vessels since their operationallife came to an end without any hazards. With regard to stability, fulfilment of the intact stabilitycriteria in force when the vessels were designed and built is verified. Operability criteria are selectedand their fulfilment is analyzed for a range of sea states, headings and velocities using linearseakeeping analysis. In light of this analysis, operability is discussed as a valid indicator of ship safety.This step is considered relevant prior to analyzing these sets of vessels through second generationstability criteria under development by the International Maritime Organization, a subject ofimmediate future research.

& 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Commercial fishing is one of the most hazardous occupationstoday, with fatality rates being widely documented in the specia-lized literature. In the US, they are around 130 fatalities per yearfor every 100,000 fishing sector workers, compared to 4 for therest of the sectors (Lincoln and Lucas, 2010); similar alarmingfigures are reported in the UK (UK MAIB, 2010).

In Galicia, the region of Spain where the fishing sector is mostprevalent, with more than one half of the fishing fleet and workersof the whole country, fishing accounts for a very high number offatal accidents during the working day, second only to theconstruction sector which employs a much larger percentage ofthe Spanish working force. It can be seen that most of thecasualties happen in vessel-related accidents (or maritime acci-dents) and among these, losses due to stability problems (capsiz-ing or large heeling) account for half, mainly in vessels of lengthbelow 24 m (Míguez González et al., 2012). This analysis isconsistent with Jin et al. (2001) who demonstrated that capsizingis the type of accident where crewmembers have the highestprobability of dying.

Between November 2004 and September 2007, five Spanish-flagged fishing vessels capsized due to loss of stability resulting ina large part of their crew dead. Examining the five accidents sideby side, it is noticeable that the vessels had similar characteristics,in particular that their lengths ranged between 15 and 24 m andthat they had all been built between 1999 and 2001 in accordancewith a recent tonnage distribution regulation (Mata-Álvarez-Santullano and Souto-Iglesias, 2013).

The shipowners, masters and crews of the capsized vesselswere the same that had been operating the vessels decommis-sioned in order to build the lost ones. The decommissioned vesselsare referred to hereinafter as “predecessors”. Moreover, it can bereasonably argued that the uses of the capsized vessels wereanalogous to the predecessors' since they operated in the samearea, using the same fishing gear and in the same social frame-work. The predecessors had been in service for many years, whilethe five fishing vessels which succeeded them sank, after a shortoperational life, in stability related accidents. It may therefore beinteresting to try and study the differences, at various levels, ofthese two sets of vessels.

All these lost vessels complied with the IMO (InternationalMaritime Organization) stability regulations, implemented in 1970in the Spanish legislation. Despite this fact, the lack of stabilitycaused all accidents. Given the growing complexity and specializa-tion of vessels and also given that the current stability criteriacould not properly cover part of the dynamic phenomena presentin several stability related accidents, the IMO has recognized that

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/oceaneng

Ocean Engineering

0029-8018/$ - see front matter & 2014 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.oceaneng.2014.01.011

n Corresponding author. Tel.: þ34 915977159; fax: þ34 915978596.E-mail addresses: fmata@fomento.es (F. Mata-Álvarez-Santullano),

antonio.souto@upm.es (A. Souto-Iglesias).1 Tel.: þ34 913367156; fax þ34 915442149.

Ocean Engineering 79 (2014) 81–91

the current stability framework can be improved. It has thereforeestablished working groups to develop the so called “SecondGeneration Intact Stability Criteria” (SGISC) (IMO, 2008). Thisnew regulation is still under development and has not yet beenapproved by the IMO. Surely those works will constitute the basisfor future stability regulations, which sooner or later will apply tofishing vessels. The proposed new stability regulation underdiscussion contemplates the use of CFD codes to assess the shipvulnerability to some stability failure modes. Nevertheless, theIMO working group undertaking this task, has not yet agreed onthe stability direct assessment methodologies, that is, the seakeep-ing calculations with CFD codes are put in question by somedelegations, mainly due to the lack of validation of the codes (IMO,2013, 2012).

Mantari et al. (2011) looked at the links between stability andsafety by considering the actions of fishing gear, wind and beamwaves from the perspective of intact stability. They compared theinclining and righting arms for different operational conditionsand evaluated the balance of energy of these external forces. Theycame to the conclusion that the inclining moments deriving fromthe fishing gear are, in many occasions, at least as important as theheeling moments produced by rough weather scenarios.

The links between safety and seakeeping has been analyzed bysome authors under several approaches. Tello et al. (2011) pro-posed a methodology based on seakeeping calculations for theanalysis of fishing vessels operability. They studied several vesselsof the Portuguese fishing fleet, proposing operability criteria withtheir corresponding limiting values. They concluded that roll andpitch criteria are the most often exceeded ones, and identifiedsome trends in hull shape that optimize the fulfillment of thosecriteria.

The idea that arises is that the relationship between safety andoperability needs to be studied.

The masters operate their ships when motions experiencedonboard are below certain levels and interrupt fishing operationsonly when those are surpassed and operation is not possible. Theyare hence the first to assume that a ship with a larger operabilityrange is a safer ship. This relation needs a rigorous assessment,which we aim at conducting in this paper by analyzing theaforementioned reference case studies, namely, by comparingsome stability and operability characteristics between the fivevessels lost and their predecessors. It is worth noting that thiswork does not intend to model specifically the accidents sufferedby the lost vessels, neither to assess their stability in roughweather. The aim of the paper is to investigate the relationshipbetween the regulatory stability and operability of these tworelevant sets of vessels as one necessary step in understandingthe limitations and prospects of the former.

The paper is organized as follows: methodology of the analysis isfirst presented by briefly reviewing IMO regulations on stability andby selecting operability criteria for a seakeeping analysis. Second,the case studies are presented looking at their main dimensions,weights, etc. Third, results for intact stability and operability basedon seakeeping analysis are presented and compared between thefamilies of new vessels and decommissioned ones. Finally a discus-sion is provided concerning the limitations of the IMO transversal

stability criteria with respect to prevention of stability failure andthe suitability of operability based criteria to help in fishing vesselssafety assessment.

2. Methodology

2.1. General

Five fishing vessels that sank due to stability failure are studied.The studied fishing vessels are similar in size, relatively small (LOAbetween 16 and 20 m), built between 1999 and 2001, and werelost from comparable stability causes between 2004 and 2007.Building any of these vessels meant that one or more existingfishing vessels had to be decommissioned refraining the tonnageof the whole fishing fleet from increasing, according to theEuropean fishing effort regulations in force. In the five casesstudied more than one vessel had to be decommissioned, Thelargest among the decommissioned vessels will be the selected“predecessor” of the capsized fishing vessel and will be considereda reference safe vessel for comparisons. Normally, and what wasthe case for the considered pairs of vessels, a fishing vessel and her“predecessor” share many characteristics such as master, crew,fishing zones, gear type, base port, etc., since the shipowner isusually the same person.

For each lost fishing vessel and her respective predecessor, acharacteristic loading condition is established. Each vessel hasbeen studied in one loading condition only, chosen from theavailable information, normally being the full load condition. Inthe case of vessels for which no stability booklets were available(most predecessors) using the best available information a loadingcondition close to the full load is estimated. Then, stability andoperability calculations for both vessels are performed.

2.2. Regulatory stability

2.2.1. GeneralThe vessels stability is checked against the IMO stability criteria

for fishing vessels proposed in the “Code on intact stability for alltypes of ships covered by IMO instruments”, approved by the IMOAssembly Resolution A.749(18).

Spain adopted these stability criteria in 1970 which weretherefore mandatory when the five vessels studied were designedand built. Since it was not mandatory when the five predecessorswere built, the IMO Severe Wind and Rolling Criterion (weathercriterion) has not been considered in the present study. Regardingthe five lost vessels, according to the Spanish stability regulations,compliance with the weather criterion has to be checked only ifthe area under the stability curve up to 301 is below 0.065 m rad inthe most unfavorable loading condition. All five vessels concernedhad larger area stability curves and therefore the previouslymentioned weather criterion did not apply in any case.

Intact stability calculations have been performed with state ofthe art naval architecture software, considering free trim. No freesurfaces in tanks have been considered. The center of gravity hasbeen considered to be at midship. For the cases where the stability

Nomenclature

B breadth [m]D depth to main deck [m]DISF displacement [t]GM transversal metacentric height [m]

KG center of gravity height over base line [m]LBP length between perpendiculars [m]LOA length overall [m]T mean draft [m]RT register tonnageVA vertical acceleration [m/s2]

F. Mata-Álvarez-Santullano, A. Souto-Iglesias / Ocean Engineering 79 (2014) 81–9182

booklet is available small differences have been found between thecalculated cross-curves and the ones in the booklet. These differ-ences are mainly attributed to the hull modeling process whenintroducing hull form data into the naval architecture software.

2.2.2. Stability indexThe stability curves (GZ) give a quick and good idea of the

stability characteristics of a vessel and, in our case, noticeabledifferences have been found between the stability curves of someof the vessels. These differences may be quantified by comparingeach single stability criterion. For comparison purposes, a singlemagnitude to better quantify the differences in regulatory stabilitybetween each fishing vessel and her predecessor is desirable andan index related to the KG margin has been devised.

For each vessel, the largest KG for which all the stability criteriaare fulfilled is computed; this is the Limiting KG. Then, the stabilityindex (SI) used to compare each vessel is the ratio between theLimiting KG and the actual KG of the loading condition studied.This SI is presented in percentage terms, and gives an idea of thereserve of stability of the vessel.

2.3. Seakeeping and operability

2.3.1. GeneralAs mentioned in the introduction, the current stability criteria

might not properly cover part of the dynamic phenomena presentin several stability related accidents. We propose to explore thelink between operability and safety by analyzing the degree offulfillment of several operability criteria through seakeepingcomputations. With this aim, the seakeeping performance of theten vessels has been analyzed. A short-term seakeeping analysishas also been carried out checking ship motions against a series ofcriterion limiting the ship operability. Tello et al. (2011) proposedseakeeping criteria to assess fishing vessels operability. For theoperability analysis presented herein, and their criteria have beenused. These are enumerated in Table 1:

Motions response operators have been calculated using thePRECAL linear seakeeping code. PRECAL computes ship motions inthe frequency domain using a 3D panel boundary element methodformulation (Chow and McTaggart, 1996; PRECAL version 6.6 UserManual, 2010). A non-linear roll damping coefficient of 0.12 hasbeen considered for the roll motion, akin to the value chosen byTello et al. (2011), considering that the same types of vessels areunder analysis in our research. The x, y, z inertia radius ratios vs. B,Lbp, Lbp have been estimated by the PRECAL code, and valuesbetween 0.32 and 0.38 are taken for roll, 0.25 for pitch, andbetween 0.27 and 0.29 for yaw. These latter values are similar tothe ones used by Tello et al. (2011) (0.4, 0.25 and 0.25respectively).

Green water on deck and propeller emergence probabilitieshave been computed according to the formulation given by Lloyd

(1989), on the basis of the relative motions between the vessel andthe sea surface given by the seakeeping code.

For the green water on deck criterion, the probability has beencalculated at two points for each vessel (deck level fore and aft)and the largest value has been chosen. Vertical and lateralaccelerations have been computed at three points for each vessel(working deck fore and aft, and the bridge). The largest valueamongst the three is chosen to check the criterion fulfillment.Calculations have been performed for headings from 01 to 1801(head seas) in steps of 301 and vessel speeds from 0 to 10 knots insteps of 2 knots.

Although, strictly speaking, the proposed criteria limit oper-ability and not safety it is assumed that, for comparative purposes,these operability values may be valid indicators of the vesselssafety. It will later be seen, by comparing capsized vessels withtheir predecessors, that this could in general not be the case.However, it is necessary to be aware of the several limitations ofthe method used, since a linear seakeeping approach cannotcapture some dynamic phenomena in waves – e.g., broaching-toor parametric rolling, and the amplitude of the ship motions inlarge amplitude waves can be questioned. These limitations mustbe taken into account when considering the relationship betweenthe calculated operability and ship safety.

2.3.2. Sea stateThe operability study has been performed in two sea states

defined by the significant wave height and modal wave periodaccording to the standardized scale adopted by NATO (MilitaryAgency for Standarization, NATO, 1983). For all vessels, SSN4 andSSN5 have been studied, corresponding to significant wave heightsof 1.88 m and 3.25 m with modal periods of 8.8 s and 9.7 srespectively.

A Bretschneider sea spectrum has been used. Three of the fivestudied vessels were lost in the Atlantic Ocean, close to theSpanish north coast; one sank in the Gulf of Cadiz, close to theGibraltar strait, while the fifth was lost in the Mediterranean Sea.Those sea states chosen have been found to represent the condi-tions of those areas, where the fleet of small coastal fishing vesselsoperates most of the time. With the RAOs obtained with PRECALand the previously mentioned spectra, the motion spectra havebeen computed and used to obtain the rms values necessary toassess the fulfillment of the operability criteria presented inTable 1.

2.3.3. Operability indexFor comparative purposes, an operability index (OI) has been

defined and is calculated as the percentage of combinationsspeed–heading at which the operating vessel complies with alloperability criteria. The OI can be rigorously established using anauxiliary function Z which depends on speed and heading and isdefined as a Boolean function, taking value 1 when at least onecriterion is not met and 0 for the safe zone:

OI¼ 1�Z π

0

Z 10

0zðθ; vÞdv dθ=ð10πÞ ð1Þ

An OI equal to zero means that at least one operability criterion issurpassed for every combination of ship speed and heading. An OIequal to 1 means the vessel would operate safely at any speed andheading, meeting all operability criteria.

Operability indexes have been obtained for the five pairs ofstudied ships and their predecessors for speeds from 0 to 10 knotsin 2 knot steps and headings from 01 (following seas) to 1801 (headseas) in 301 steps. The values of the operability indexes have beeninterpolated for those points lying inside the intervals defined.This approach is necessary to obtain accurate operability graphs

Table 1Operability criteria.

Criterion Prescribedmaximum value

C1 Roll 61 (rms)C2 Pitch 31 (rms)C3 Lateral acceleration (at the previous three points) 0.1 g (rms)C4 Vertical acceleration (at bridge, working deck fore

and working deck aft)0.2 g (rms)

C5 Propeller emergence 15% (probability)C6 Green water on deck (at working deck fore and

working deck aft)5% (probability)

F. Mata-Álvarez-Santullano, A. Souto-Iglesias / Ocean Engineering 79 (2014) 81–91 83

and hence meaningful OIs. The loading conditions studied are thesame as in the stability study.

3. Fishing vessels studied

As mentioned in the introduction, between November 2004and September 2007 five Spanish-flagged ships capsized due totransversal stability related causes. Main dimensions and othercharacteristics of those vessels are presented in Table 2. In thispaper, they are labeled as F1–F5. According to the Europeancommon fisheries policy, in order to build one fishing vessel, oneor more vessels had to be decommissioned accounting for thesame gross tonnage and power as the ship to be built. The vesselsthat were retired from service to build F1–F5 are referred to as“predecessors”. Table 3 summarizes the main characteristics of thepredecessors, labeled as P1–P5.

The ships in these tables are referred to using the EuropeanFishing Fleet Register (EFFR) code (http://ec.europa.eu/fisheries/fleet/index.cfm?lg=en).

The case studies LOAs range between 16 and 20 m. They areclassified in three categories according to their fishing gear: seines,hook and lines, and gillnets and entangling nets. Predecessorsbelong to the same categories and are slightly smaller in theirmain dimensions (e.g. LOA¼11.3–16 m compared to 16–20 m).

The body plans of the ten vessels are shown (not to scale) inTable 4.

Main characteristics of both sets of vessels and loading condi-tions studied are included in Table 5.

4. Results

4.1. Regulatory stability

The stability curves of the vessels studied are presented inFig. 1. Comparison of stability curves of vessels (F1–F5) and theirpredecessors (P1–P5) have been plotted jointly.

Regarding the stability curves calculated for each vessel, allpredecessors had, in general, a larger GZ and, thus, better dynamicstability than the lost vessels. The capsized fishing vessels had alsoa lower GM than their respective predecessors.

The stability index computed for all vessels is presented inTable 6. As mentioned, the stability index (SI) for comparison is thelimiting KG margin (in percentage terms) over loading conditionKG. The values of the stability index for the lost vessels (SIlv) andfor the predecessors (Sip), as well as the ratio between the twomagnitudes are included. It is remarkable that most predecessorsshow larger stability indexes than the vessels substituting them.For the pairs F1–P1, F3–P3 and F5–P5 the differences are quitenoticeable. It is also remarkable that vessels F1 and F5, althoughcomplying with the criteria, had very little stability margin. Thereason, for the F5 case, is the short range of positive stability,which causes the heel angle corresponding to maximum GZ to bequite close to 251. This is due to a large part of F5 superstructurenot being considered watertight by the designer and not con-tributing to the intact stability. Therefore, the F5 stability curvestops growing when the deck edge gets submerged. It must behighlighted that the marginal compliance with the angle at whichthe maximum GZ occurs may lead to a rapid capsize in case astrong wind gust is applied to the vessel.

In any case, all the case studies and predecessors fulfilled allapplicable IMO stability criteria. Considering that similar opera-tional factors applied for capsized vessels and predecessors, aquestion mark on the suitability of intact stability criteria for thesesets of vessels can be placed. The idea of looking at operabilitycriteria in order to investigate the vessels safety thus arisesnaturally.

4.2. Operability

In this section, the results of the operability calculations arepresented. Figs. 2–6 present a graphical representation of oper-ability with areas where each criterion is exceeded for each vessel,predecessor and sea state. Safe zones where no threshold isexceeded are also shown. Global operability graphics in SSN4and SSN5 considering simultaneously roll, pitch, lateral accelera-tion and vertical acceleration (excluding green water on deck andpropeller emergence) are presented in Figs. 7 and 8. The oper-ability indexes (OI) (Eq. (1)) calculated for all studied vessels arepresented in Table 7.

Table 7 summarizes the calculated operability index (OI) for allvessels, considering ship motions only, excluding green water andpropeller emergence.

Table 2Fishing vessels lost.

Label Boat code in Europeanfleet

Gear type Length overall(m)

Tonnage(GT)

Year ofbuild

Year ofloss

Possible accident causes (from the official investigationreports)

F1 ESP25057 Seines 17.00 34.18 2001 2004 Lack of stability; probably surf-riding and broachingF2 ESP24593 Hook and linesa 16.02 29.97 2000 2004 Lack of stability, probably overloadingF3 ESP24391 Seines 18.00 44.83 1999 2004 Lack of stability, probably surf-riding and broachingF4 ESP24358 Gillnets and

entangling nets20.00 87.03 1999 2006 Lack of stability, probably dead ship condition and fishing

spaces floodedF5 ESP24199 Seines 19.40 59.01 1999 2007 Lack of stability, probably inadequate weight distribution

a When capsized the ship was operating as a pot vessel.

Table 3Fishing vessels predecessors to the lost ones.

Label Boat code in European fleet Gear type Length overall (m) Tonnage (GT) Year of build Year of retirement Notes

P1 ESP16060 Seines 15 17.11 1989 2001 Predecessor to ESP25057P2 ESP11830 Hook and lines 11.3 5.86 1963 2000 Predecessor to ESP24593P3 ESP05969 Seines 14.1 28.7 1978 1999 Predecessor to ESP24391P4 ESP00251 Gillnets and entangling nets 16 47 1983 1999 Predecessor to ESP24358P5 ESP05154 Seines 15.75 29 1959 1999 Predecessor to ESP24199

F. Mata-Álvarez-Santullano, A. Souto-Iglesias / Ocean Engineering 79 (2014) 81–9184

4.2.1. Results analysisRegarding the operability graphs and OI's calculated in the

previous section, some conclusions can be drawn:

(1) Most of the lost vessels had greater operability than theirpredecessors. This difference is most noticeable in SSN4,where significant differences are found for F1, F2 and F4 withrespect to P1, P2 and P4.

(2) For all vessels, operability deteriorates with increasing seastates. In SSN4 all lost vessels (F1–F5) maintain high levels ofoperability, with the OI ranging between 0.81 and 0.96. Thelost vessels operability deteriorates significantly in SSN5,except for F4 and F5 that still have OI of 0.62 and 0.59respectively. These two vessels being the largest ones amongstthe ten studied, them having a larger operability is expected.

(3) Regarding the predecessors, more heterogeneity is found.Apart from P2, the OI of the other four vessels varies from0.47 to 0.90. Regardless of the wave height being considered,excessive pitch responses above the limiting criterion havebeen found for the P2 case, which significantly reduces theglobal operability of the vessel. This might be attributable tothis vessel being the smallest one, with 11 m LOA, well belowthe rest of the studied vessels. Therefore, operability results forP2 must be cautiously taken.

When examining the operability results in Figs. 2–5, whichpresent criteria separately, some points arise.

(1) In general, predecessors present lower operability than thelost vessels regarding the pitch motion criteria.

(2) Green water on deck and propeller emergence are quitesensitive to the sea state. Most vessels fulfill these criteria inany combination of heading and speed in SSN4 (Hs¼1.88 m)but fail to comply with them in SSN5 (Hs¼3.25 m) for mostheading/speed combinations. Therefore, the operability of allvessels regarding this criterion drastically deteriorates withrelatively small increases in wave height. For this reason, greenwater and propeller emergence are not useful criteria forcomparative purposes between the two sets of vessels studied,since in all cases, the drastic operability deterioration occurs ina relatively small wave height interval.

(3) The most often exceeded criteria are pitch, roll, and lateralacceleration.

(4) Pitch becomes a problem for ship headings close to 1801 (headseas) and high speeds. When increasing wave height, the pitchin the following waves (01) also limits operability.

(5) In general, vessels P1–P5 do not present larger roll motionsthan vessels F1–F5. Roll is not a very limiting criterion in SSN4,except for vessel P2. Regardless of the speed of the vesselbeing considered and for headings around 901 (beam waves),roll criterion is a major limiting factor of operability of most ofthe vessels in SSN5. Notice this feature might pose a problemfor vessels that operate at zero speed without manoeuvringcapability, especially, purse seiners while pulling the net andpulling catches onboard. A common scenario for those vessels,while pulling the net in bad weather, is to be pushed by windand waves, ending transverse to the waves and suffering rollsthat can put the vessel in danger (Mantari et al., 2011).

Table 4Case studies and predecessors' body plans.

Fishing vessels (F1–F5) Predecessors (P1–P5)

F1–P1

F2–P2

F3–P3

F4–P4

F5–P5

Table 5Main dimensions of the lost vessels (F1–F5) and their predecessors (P1–P5).

Lbp (m) B (m) D (m) T (m) DISF (t) Freeboard (m) KG (m) RT Nat. roll period (s) GM (m)

F1 13.5 5 2.35 2.127 71.21 0.223 2.191 30.15 5.2 0.579P1 12.6 4.08 1.54 1.05 25 0.49 1.185 13.37 3.3 0.842F2 13.8 4.57 2.15 1.8 53.23 0.658 1.8 26.03 4.4 0.68P2 9.5 4.38 1.57 1.17 21.65 0.4 1.334 10.1 3.5 0.93F3 13.5 5.2 2.35 1.775 80.67 0.336 2.21 23.69 4.8 0.76P3 12.5 5.12 1.7 1.264 43.6 0.436 1.56 26.4 3.3 1.38F4 16.2 5.3 2.3 2 97.82 0.3 2.238 34.46 4.3 0.62P4 14.46 4.7 1.902 1.302 49.2 0.6 1.98 20 3.3 1.1F5 15.5 5.75 2.5 1.73 90.8 0.759 2.17 32.45 4.8 0.96P5 14 5 2.068 1.616 68.14 0.455 1.104 32.45 3.4 1.2

F. Mata-Álvarez-Santullano, A. Souto-Iglesias / Ocean Engineering 79 (2014) 81–91 85

(6) Regarding vessel accelerations, some conclusions can also bedrawn. Degradation of operability due to the lateral accelera-tion criterion being exceeded with increasing wave heights issignificant for most vessels, while vertical acceleration criter-ion being exceeded is less dependent on sea state. Therefore,in general, lateral accelerations limit operability more thanvertical ones. Lateral acceleration responses are significant forvessel headings around 901, which is a foreseable result, beingstrongly correlated with the roll motion.

When looking at operability and taking into account all criteriasimultaneously (Figs. 6 and 7), the following conclusions can bedrawn:

(1) In general, operability of all vessels is more significantlyimpaired when facing bow waves (90–1801) for both sea statesSSN4 and SSN5.

(2) In SSN5 vessels operability region gets confined to quarteringseas (headings from 301 to 901).

Comparing the operability with the stability characteristics ofthe ten vessels studied in Section 4, some conclusions are reached.

(1) The lost vessels have, in general, larger operability but lessstability than their predecessors.

(2) When comparing stability and operability between a lostvessel and her predecessor, large differences in operabilitydo not always imply large differences in stability. For instance,F4 and P4 have stability indexes quite similar (see Table 6) butthe operability differences between F4 and P4 are quitesignificant (see Figs. 5, 7 and 8). This suggests that there isno direct and obvious correlation between stability andoperability.

Summarizing the previous results, it can be stated that oper-ability studies based on linear seakeeping calculations may not beenough to assess ship safety. A consistent relation between shipstability and ship operability, calculated from linear seakeepingmethods, has not been found. Due to the particular nature of thecase studies (all vessels capsized in stability related accidents) theconclusion can be drawn that a direct link between operability andsafety in the five case studies and by extension for these types ofvessels cannot be established.

4.3. Additional remarks

Some additional issues, which do not directly affect the mainconclusions, are worth presenting:

(1) The stability curves for vessels F1 and F3, presented in Fig. 1,have a peculiar behavior. These curves grow regularly between0 and 10–201, then, the curves remain almost horizontal upto 30–401, and finally the growing ratio increases again.The reason is that the main deck is submerged because ofa reduced freeboard, and then the stability increases againwhen the watertight superstructures are submerged as heelincreases.

(2) According to the operability results, in SSN5, for most speedsstudied, ships tend to be more operative in headings of 30–601(following waves). It is interesting to realize that navigatingwith these headings may impose additional risks since thedynamics of vessels in stern seas is not contemplated in theoperability studies. It is well documented (IMO, 2007) that forthose headings and speeds, fishing vessels may be at risk ofsurf-riding and broaching. In addition, with respect to Table 3,

Table 6Stability index (SI) of the lost vessels (F1–F5) and their predecessors (P1–P5).

Lost vessel SIF (%) Predecessor SIP (%) Ratio SIP/SIF

F1 2.7 P1 11.0 4.6F2 9.9 P2 19.7 2.0F3 6.7 P3 51.0 7.6F4 10.1 P4 8.0 0.8F5 2.0 P5 37.0 18.5

Vessels F1-P1

Vessels F2-P2

Vessels F3-P3

Vessels F4-P4

Vessels F5-P5

Fig. 1. Comparison of stability curves of vessels (F1–F5) and their predecessors(P1–P5).

F. Mata-Álvarez-Santullano, A. Souto-Iglesias / Ocean Engineering 79 (2014) 81–9186

SSN4 SSN5

F1

P1

Fig. 2. Operability of vessels F1 and P1, in SSN4 and SSN5 (see Table 1 for criteria definition).

SSN4 SSN5

F2

P2

Fig. 3. Operability of vessels F2 and P2, in SSN4 and SSN5 (see Table 1 for criteria definition).

F. Mata-Álvarez-Santullano, A. Souto-Iglesias / Ocean Engineering 79 (2014) 81–91 87

SSN4 SSN5

P3

Fig. 4. Operability of vessels F3 and P3, in SSN4 and SSN5 (see Table 1 for criteria definition).

SSN4 SSN5

F4

P4

Fig. 5. Operability of vessels F4 and P4, in SSN4 and SSN5 (see Table 1 for criteria definition).

F. Mata-Álvarez-Santullano, A. Souto-Iglesias / Ocean Engineering 79 (2014) 81–9188

it is stated that some of the vessels studied which sank couldhave experimented surf-riding and broaching.

(3) According to the authors' professional experience, it is notuncommon that fishing vessel masters identify stability with a

low level of motions on board. A vessel with high operabilitymay generate false safety perceptions on the crew, and themaster in particular. Small amplitude ship motions andreduced pitch/roll/accelerations may be perceived by the

SSN4 SSN5

F5

P5

Fig. 6. Operability of vessels F5 and P5, in SSN4 and SSN5 (see Table 1 for criteria definition).

F1 - P1 F2 - P2 F3 - P3 F4 - P4 F5 - P5 Vessel

Predec.

Fig. 7. Operability comparison of vessels F1–F5 and P1–P5, in SSN4, considering roll, pitch, lateral acceleration and vertical acceleration.

F. Mata-Álvarez-Santullano, A. Souto-Iglesias / Ocean Engineering 79 (2014) 81–91 89

masters as a symptom of good stability, while according to thefindings of this paper it is not necessarily true; that is to say,there is no strong relation between stability and operability.This suggests that fishing vessel safety cannot be assessedneglecting the human element, only by the study of shipmotions or ship stability. This reinforces the idea that moreinvestigation about ship motions on a seaway is needed. Therelationship between ship stability, safety and motions mustbe investigated. These conclusions also suggest that an ade-quate training in stability for fishing vessels masters is needed.

5. Conclusions

In the present paper, the intact stability and short termoperability of ten small fishing vessels have been studied. Fivevessels built in the same time frame, which sank due to stabilityrelated causes in a short period of time, and the five respectivevessels which were decommissioned to build those (referred to as“predecessors”) have been compared. These predecessor vessels,which ended their service life in a regular way, are considered as asafety reference. It is relevant to stress that the shipowners,masters and crews of the capsized vessels were the same onesthat had been operating the predecessors, in the same fishing

areas, using the same fishing gear type, and in the same socialframework.

The intact stability of each vessel has been characterized by herstability curve in a characteristic loading condition, and by astability index, defined from the limiting KG that allows the shipto fulfill the IMO intact stability criteria. The Weather Criterion hasnot been considered since it was not applicable to the selectedvessels when these were designed and built.

The stability of each lost fishing vessel has been compared withthe stability of her predecessor. It has been found that the newvessels had in general larger stability than the predecessors.Notwithstanding that, the ten vessels fulfilled the IMO stabilitycriteria and hence, from the stability point of view, they can beconsidered equally safe.

Considering that the capsized vessels and their predecessorswere operated in similar contexts, this stability analysis opens aquestion mark on the suitability of intact stability criteria for thesecases and therefore the possibility of analyzing the vessels oper-ability in order to characterize the vessels safety has beenexplored. The masters operate the ships responding to the fulfill-ment of operability criteria and interrupt fishing operations onlywhen those are surpassed and operation is no longer possible.They are hence the first to assume that a ship with a largeroperability range is a safer ship.

Operability of each vessel is established by calculating hershort-term motions in two typical sea states with linear seakeep-ing analysis, and checking these motions against a set of oper-ability criteria. A global operability index has been definedfor comparison purpose. The operability of each pair ofvessels (sunken vessel and predecessor) has been compared,resulting in the capsized vessels having more operability thanthe predecessors.

As a main conclusion, the comparison between the stabilitycharacteristics of two sets of vessels and the comparison betweenthe operability characteristics of the same two sets of vesselsthrow opposite results. While the predecessors had in generalmore stability, the lost ones had in general larger operability. Thus

F1 - P1 F2 - P2 F3 - P3 F4 - P4 F5 - P5Vessel

Predec.

Fig. 8. Operability comparison of vessels F1–F5 and P1–P5, in SSN5, considering roll, pitch, lateral acceleration and vertical acceleration.

Table 7Operability indexes of the lost vessels (F1–F5) and their predecessors (P1–P5).

F1–P1 F2–P2 F3–P3 F4–P4 F5–P5

OI SSN4Lost vessels 0.94 0.81 0.86 0.92 0.91Predecessors 0.47 0.00 0.90 0.52 0.83Ratio OIP/OIF 0.50 0.00 1.05 0.57 0.91

OI SSN5Lost vessels 0.17 0.15 0.30 0.62 0.59Predecessors 0.08 0.00 0.31 0.08 0.32Ratio OIP/OIF 0.47 0.00 1.03 0.13 0.54

F. Mata-Álvarez-Santullano, A. Souto-Iglesias / Ocean Engineering 79 (2014) 81–9190

the masters of the new fishing vessels could have considered themto be safer as they experienced, in general, lower motions andaccelerations, while in fact the new vessels were less stable thantheir predecessors, and might have required a more carefuloperation.

Overall, these results indicate that usual operability criteriamay not contribute much to assess ship safety during designphases. It also suggests that masters should be strongly trained instability, making them able to adequately manage their vessel'sstability regardless of the operability behavior.

As a final remark, taking again into account that the sunkenvessels fulfilled IMO stability criteria and had larger operabilitythan the predecessors, we believe that more effort is neededtowards developing and validating new and more complex stabi-lity criteria, able to capture the reality of the dynamics of fishingvessels at sea.

Acknowledgments

The authors are grateful to Hugo Gee, José L. Cercos-Pita and ClaraMata for assisting in the preparation of the manuscript. The authorsare also grateful to one anonymous reviewer for her/his detailedanalysis of the manuscript and subsequent recommendations.

References

Chow, D.L., McTaggart, K.A., 1996. Validation of SHIPMO7 and PRECAL with aWarship Model (Technical Memo). Defense Research Establishment AtlanticDartmouth, Nova Scotia, Canada.

IMO, 2007. Circular MSC.1/Circ.1228. Revised Guidance to the Master for AvoidingDangerous Situations in Adverse Weather and Sea Conditions.

IMO, 2008. Report to the Maritime Safety Committee (SLF 51 Meetings Period No.SLF 51/17), Stability, Load Lines and Fishing Vessels Sub-committee. Interna-tional Maritime Organization, London.

IMO, 2012. Development of second generation intact stability criteria. InternationalMaritime Organization, London (Report of the Correspondence Group on IntactStability (Part I) (No. SLF 55/3/1), Stability, Load Lines and Fishing Vessels Sub-Committee).

IMO, 2013. Comments on Present Status of Development of Second GenerationIntact Stability Criteria (No. SLF 55/3/7), Stability, Load Lines and Fishing VesselsSub-CommitteeInternational Maritime Organization, London.

Jin, D., Kite-Powell, H., Talley, W., 2001. The safety of commercial fishing:determinants of vessel total losses and injuries. J. Saf. Res. 32, 209–228.

Lincoln, J., Lucas, D., 2010. Commercial fishing deaths—United States, 2000–2009. J.Am. Med. Assoc. 304, 1437–1439.

Lloyd, A.R.J.M., 1989. Seakeeping: Ship Behavior in Rough Weather. Ellis HorwoodLimited, Market Cross House, Cooper Street, Chichester, West Sussex PO19 1EB,England.

Mantari, J.L., Ribeiro e Silva, S., Guedes Soares, C., 2011. Intact stability of fishingvessels under combined action of fishing gear, beam waves and wind. OceanEng. 38, 1989–1999.

Mata-Álvarez-Santullano, F., Souto-Iglesias, A., 2013. Fishing effort control policiesand ship stability: analysis of a string of accidents in Spain in the period 2004–2007. Mar. Policy 40, 10–17.

Míguez González, M., Caamaño Sobrino, P., Tedín Álvarez, R., Díaz Casás, V.,Martínez López, A., López Peña, F., 2012. Fishing vessel stability assessmentsystem. Ocean Eng. 41, 67–78.

Military Agency for Standarization, NATO, , 1983. Standardized Wave and WindEnvironments and Shipboard Reporting of Sea Conditions (STANAG No. 4194).

PRECAL version 6.6 User Manual, 2010.Tello, M., Ribeiro e Silva, S., Guedes Soares, C., 2011. Seakeeping performance of

fishing vessels in irregular waves. Ocean Eng. 38, 763–773.UK MAIB, 2010. MAIB Annual Report 2010. Maritime Accident Investigation Branch,

Government of UK.

F. Mata-Álvarez-Santullano, A. Souto-Iglesias / Ocean Engineering 79 (2014) 81–91 91