Maritime University of Szczecin
Akademia Morska w Szczecinie
2010, 20(92) pp. 134–139 2010, 20(92) s. 134–139
Estimation merchant ship usability for SAR action based
on information from AIS system
Oszacowanie przydatności statku handlowego do akcji SAR
na podstawie danych systemu AIS
Andrzej Starosta
Gdynia Maritime University, Faculty of Navigation, Ship Operation Department Akademia Morska w Gdyni, Wydział Nawigacyjny, Katedra Nawigacji
81-345 Gdynia, al. Jana Pawła II 3, e-mail: Andrzej.Starosta@wp.eu
Key words: safety, merchant ship, SAR action, weather conditions Abstract
Merchant vessels may be used by SAR mission coordinator (SMC) according to SOLAS convention and other international conventions. Some vessels are more suitable than others due to construction, number of person on board, actual stability situation, etc. Actually SMC may use ship which declared readiness to attend SAR action only. Modern information systems give SMC information about other vessels in action area also. SMC is responsible to choose the best vessel for SAR action. The paper presents model of Bayesian network to estimate merchant ship usability for SAR action. Information from AIS system is very useful for this model. Combination AIS system with presented model gives SMC tool which will help to find the best ship in area. Some ship particulars are more important that other and may bring about that the ship will be not suitable for SAR action.
Słowa kluczowe: bezpieczeństwo, statek handlowy, akcja SAR, warunki pogodowe Abstrakt
Zgodnie z konwencją SOLAS i innymi międzynarodowymi konwencjami, koordynator akcji SAR (SMC) ma prawo wykorzystać statek handlowy do prowadzenia akcji poszukiwania i ratownictwa (SAR). Ze względu na swoją konstrukcję, liczbę osób znajdujących się na pokładzie, aktualną sytuację statecznościową itp., niektóre statki będą lepiej nadawały się do przeprowadzenia akcji SAR niż inne. W świetle aktualnych przepisów ko-ordynator akcji (SMC) może zarekwirować jedynie ten statek, który wcześniej zgłosił swoją gotowość do udzielenia pomocy. Na podstawie nowoczesnych systemów wymiany informacji, koordynator posiada infor-macje również o innych jednostkach znajdujących się w pobliżu obszaru poszukiwań. Koordynator jest od-powiedzialny za wybór statku, który będzie najlepszy do przeprowadzenia akcji SAR. W artykule przesta-wiono model służący do oszacowania przydatności statku handlowego do akcji SAR zbudowany z wykorzy-staniem sieci Bayesowskiej. Dane o statku dostępne dzięki systemowi AIS są bardzo użyteczne w tym mode-lu. Połączenie systemu AIS z prezentowanym modelem daje koordynatorowi cenne narzędzie pozwalające bardzo szybko zidentyfikować najbardziej odpowiednie jednostki do przeprowadzenia akcji SAR. Niektóre parametry statku są najistotniejsze w tej ocenie i mogą wskazywać, że statek nie nadaje się do prowadzenia akcji w danych warunkach pogodowych.
Introduction
Each merchant ship may become rescue unit in SAR action. Based on humanitarian considerations and established international practice each ship is responsible to render assistance to a distress vessel at sea. Under long-standing traditions of the sea
every ship master helps other ship in danger on sea. Some years ago, when radio communication was the only way to inform others about our distress situation, only ships in close range was able to help. Masters may refuse to assist other ship when he can’t do it safely, but he did it very rare, because he
know that his ship may be only one who received distress signal.
Nowadays when distress signal is transmitted a lot of people, organizations and vessels are alerted about situation thanks to modern satellite commu-nication and GMDSS system. Every sea countries have special Search and Rescue organizations with specialized equipment and coordination centers established on the strength of the International Convention on Maritime Search and Rescue [1]. Each SAR operation is carried out under guidance of SAR Mission Coordinator (SMC). He is respon-sible for preparing SAR action plan and assign tasks for SAR units. He may used specialized units which are the part of national SAR organization but he may also choose the merchant ships which acknowledged receipt of distress message and are proceeding to the area of distress.
SMC has to decide which ships are the most suitable for action. According to the special algo-rithm he has to evaluate if merchant ship is good for SAR action [2, 3]. He assesses the risk inherent in any SAR action against the chances for success and the safety of SAR personnel and equipment. SMC has to be convinced that merchant ship will be able to sail safely on all courses ordered search pattern. Based on reports from ships he chooses ship which is the most suitable. Decision algorithm is presented in figure 1.
MRCC has more information about others ships in area of accident very often, for example from AIS system. Thank to this system operator in
MRCC has basic information about each ship. Those dates allow estimating ship usability based on model presented in this paper. The paper presents that some common ship particulars like length of the ship are very important to make proper decision.
Bayesian network elements of merchant ship usability for SAR action
Bayesian networks are very useful tool because they are intuitive tool which adds transparency and consistency to the modeling. Within maritime in-dustry they may be used in a wide range of applica-tion, for example inspection and repair planning for offshore structures, grounding risk analysis or plan-ning ship design [5].
To choose the best available merchant ship SMC should take into consideration some elements of ship safety. He should find answer for following questions:
– can the ship handle the weather conditions dur-ing saildur-ing on ordered search pattern;
– what is the risk for cargo carried by the ship; – what is the risk for marine environment in case
of the ship damage or lose the cargo;
– is the ship equipped to detect search objects; – has the ship got equipment to take up survivors
on board.
Based on those elements Bayesian network was constructed [6, 7]. It is presented in figure 2. Information about ship
Calculate search pattern
Change search pattern Release ship Release ship – Ship particulars – Safety speed – Type of cargo – Roll period – Weather conditions (waves parameters) Ship safety checking – Master
Pass search pattern parameters on Use the ship Is the ship safe? Is the ship safe? NO YES YES YES NO NO
Fig. 1. Algorithm of SMC decision to use merchant ship in SAR action [4]
The merchant ship usability for SAR action may be divided into two main elements: ability to find object in danger and ability to rescue survivors, to get people out of the water. Those two elements depend on ship equipment, number of trained crew and ability of the ship to sail in heavy weather con-ditions.
The main problem is ship safety during SAR action. The master of the ship takes decision if his vessel may complete tasks ordered by SMC (Fig. 1) In standard navigation, during storm weather, navigator changes course and speed to find way which will be the safest for his ship. In SAR action rescue craft has to sail according special types of search patterns, for example expanding square search or parallel sweep. It requires ability of safety sailing on four correlated courses no only on one like in normal navigation. SMC should foresee if the ship will be able to use one of search patterns otherwise he will waste time to prepare search plan which will be rejected by the master due to ship safety. For this purposes determine danger courses algorithm based on IMO guidance to the master for avoiding dangerous situations in adverse weather may be used [2, 3]. Knowledge of main parameters of the ship like length, breadth, actual draft, speed, and weather conditions, wave parameters specially, allow calculating probability of finding safe search pattern.
The ship safety is strongly connected with safety of the cargo. Some cargos are more sensitive to heavy rolling than others. For example trucks carried by ro-ro vessel may move and cause list or even capsized of the ship. Deck cargos like containers or wood are lost during heavy weather
very often. Cargos in bulk are more insensitive for rolling but on the other hand damage of the tanker may be very big danger for marine environment. So the type of the ship and carried cargo should be taken into consideration by SMC. Knowing of ship type allows also estimating technical equipment fitted on board and kind of rescue equipment. For example on car carrier rescue boat is located about 30 meters above sea. Launching of the boat is very dangerous for crew and later taking up is almost impossible. Type of the ship and dimensions allow also estimating number of crew – potential observers (look-outs).
Traditionally SMC obtains that information from shim master during communication. But basic information may be obtained from other sources, for example AIS system.
Information from AIS system
From 1st January 2005 all ships of 300 gross tonnage and upwards, engaged on international voyages have to be fitted in Automatic Information System (AIS) equipment according to Regulation 19 of SOLAS Chapter V [8].
The regulation requires that AIS shall provide information (including the ship’s identification, type, position, course, speed, navigational status and other safety-related information) automatically to appropriately equipped shore stations, other ships and aircrafts. Information from AIS is used for traffic control and other safety applications. Signals from ships are received by MRCC or VTS stations but may be received by private users also and published in internet. Screenshot from one of the web site with situation on the Baltic Sea is Ship type particulars Ship
Rescue boat Deck cranes
Damage of cargo Ability to rescue servivors Ability to detect search objects Usability for SAR action Danger for marine environment Safety of ship on search pattern Technical observation equipment Observer effectiveness Persons on Board Weather conditions (wave parameters) Rescue equipment
Fig. 2. Ship usability for SAR action – The Bayesian network Rys. 2. Przydatność statku do akcji SAR – sieć Bayesowska
presented in figure 3. Information about ships from AIS may be used to estimate merchant ships usability for SAR action.
Fig. 3. Ships on the Baltic Sea [9] and their usability for SAR action
Rys. 3. Statki na Morzu Bałtyckim [9] i ocena ich przydatności do akcji SAR
Types of AIS information and where they are using during calculation are presented in table 1. Table 1. Use of AIS information in the network of ship usabi-lity for SAR action
Tabela 1. Sposób wykorzystania informacji z systemu AIS w sieci określającej użyteczność statku do akcji SAR
Type of
information Use in the network calculation
Static information
MMSI number N / A (identification only) IMO number N / A (identification only) Call sign N / A (identification only) Name N / A (identification only) Overall
dimension Length and Breadth are used to estimate danger courses in the algorithm. Type of ship Use to estimate type of equipment and type of cargo.
Voyage information
ETA /
Destination N / A Cargo
type DG
Use to estimate danger for marine environ-ment in case of ship’s damage during action. Maximum
draught Use to estimate danger courses in the algo-rithm. Crew number Use to calculate ability to detect object.
Dynamic information
Latitude /
Longitude Calculate time of arrival to the action area. Navigational
status Ships under way using engine may be used for SAR action only (status 0). SOG
Speed is used to calculate danger courses in the algorithm and time of arrival to the action area.
COG N / A True heading N / A Rate of turn N / A
Example of calculations
Let’s assume hypothetic rescue action on the Baltic Sea. Positions of the ships are presented in figure 3. The place of the ship in distress is marked as “X”. Usability for SAR action of 23 ships in close range was checked. Information obtained from AIS system (except GM which was assumed) and used for calculation are presented in table 2. Table 2. Data of ships used for calculations
Tabela 2. Dane statków użyte do obliczeń
Name of ship Type L B d SOG GM m m m kn m ALEKSANDR SIBIRYAKOV Cargo 132 19 6.1 12.5 0.5 ARESSA Cargo 95 13 4.3 7.5 0.4 AZURYTH Tanker 105 16 6.9 11.8 1.2 BAUGE Tug 30 10 4.2 7.3 2.0 BRITISH ROBIN Tanker 252 44 12.3 14.1 0.9 BRO ALMA Tanker 143 22 6.1 13.2 1.8 CARRY Tanker 185 27 7.6 13.3 2.1 CONTAINERSHIP 6 Cargo 155 22 6.1 18.7 0.5 EMMA Cargo 90 13 5.1 11.4 0.6 FINNLADY Passenger 219 31 7.1 23.9 1.4 GRANIT Tug 64 14 5.5 9.1 1.5 GREAT JADE Cargo 225 32 13.7 13.8 1.4 INGVILD Cargo 99 16 4.8 13 1.7 KING ERIC Tanker 182 27 7.5 15.1 2.0 KRASNODAR Tanker 249 44 11.8 14.4 1.5 LISCO MAXIMA Passenger 199 26 5.5 21.1 0.8 SCARLET STAR Tanker 183 27 5.4 13.5 1.8 SORMOVSKIY 3003 Cargo 114 13 3.1 9.0 1.2 SPICA Cargo 151 20 7.5 19.1 0.6 STK-1007 Cargo 82 12 3.9 8.4 0.9 SUNNAVIK Cargo 124 18 6.8 13.9 1.0 SYAM Cargo 81 11 4 10.2 0.8 VISCARIA Tanker 83 12 4.8 10.2 1.2 In calculation simplified Bayesian network was used. The main element of calculation was safety of ship on search pattern. Based on danger course algorithm probability of finding safety search pat-tern was calculated for each ship and entered to the network. In the network ability of search object and ability of rescue survivors was assumed the same for each ship. Type of ship, number of crew on board and any other aspects wasn’t taken under consideration. Calculation was made in GeNIe 2.0 program and example final calculation is shown in figure 4 [10].
Fig. 4. The network used in example – GeNIE 2.0 [10] Rys. 4. Sieć użyta w przykładowych obliczeniach – GeNIE 2.0 [10]
To calculate safety of search pattern waves parameters are needed. In example wave periods (Tw) correlated with wave lengths was assumed as follow: Tw = 6 s – 4% time of SAR action; Tw = 7 s – 10%; Tw = 8 s – 15%; Tw = 9 s – 60%; Tw = 10 s – 10%; Tw = 11 s – 1%.
Table 3. Ranking of ships usability for SAR action Tabela 3. Klasyfikacja przydatności statków do akcji SAR
Name % Tw [s] 6 7 8 9 10 11 KRASNODAR 52 18 1 1 1 2 3 ALEKSANDR SIBIRYAKOV 49 10 6 2 2 4 2 BRITISH ROBIN 48 1 3 3 3 6 4 FINNLADY 48 2 4 4 4 3 5 LISCO MAXIMA 43 3 2 6 5 10 13 SPICA 39 6 5 7 9 9 14 CONTAINERSHIP 6 38 4 7 5 12 8 7 SUNNAVIK 36 12 11 17 6 17 8 BRO ALMA 35 13 14 13 7 18 15 GREAT JADE 35 7 13 9 10 11 18 ARESSA 34 5 8 8 13 20 9 KING ERIC 34 8 9 12 14 5 17 BAUGE 32 14 19 22 15 1 1 VISCARIA 32 11 18 15 8 19 23 INGVILD 31 19 10 18 16 7 11 AZURYTH 30 15 12 20 11 16 12 CARRY 27 16 16 19 17 21 6 SCARLET STAR 27 9 15 14 18 15 22 GRANIT 24 17 20 16 20 13 20 STK-1007 23 22 22 23 19 22 16 SYAM 22 20 21 11 22 12 19 SORMOVSKIY 3003 22 23 23 10 21 14 21 EMMA 19 21 17 21 23 23 10
After calculation top-ranking of ships was prepared and is presented in table 3. In table are presented places according assumed weather condi-tions and places according all wave parameters separately. The best is tanker “KRASNODAR” but his final mark should be reduced because she is far away from the action area. Closer are general cargo “ALEKSANDR SIBIRYAKOV”, tanker “BRI-TISH ROBIN” and ro-pax “FINNLADY” that have good usability mark also. From those ships FINNLADY as a ro-ro ship with passenger on board should be released because of potential risk for passengers. For action should be used “GRANIT” also, because she is close to action area and she is AHTS (Anchor Handling and Tug Vessel), so her ability to rescue survivors is very big. This shows how important in the network is information about type of ship which wasn’t taken into consideration in this example.
Discussion about results
In calculation probability of finding safety search pattern was determined based on danger course algorithm. The algorithm is constructed based on “Revised guidance to the masters for avoiding dangerous situation in adverse weather and sea conditions” [11]. The guidance is restricted to hazards that may cause capsizing of the vessel or heavy rolling. Other hazards and risks like damage through slamming, longitudinal or torsional stresses, risk of collision or stranding are no addressed in this publication and must be addi-tionally considered by masters. The guidance are focused on following dangerous phenomena: – surf-riding,
– broaching-to,
– synchronous rolling motions – parametric rolling motions, – reduction of intact stability, – successive high-wave attack.
In presented calculation reduction of intact stability wasn’t taken into consideration.
The intact stability can be decreased substan-tially according to changes of the submerged hull form, when a ship is riding on the wave crest. This situation is dangerous in following and quartering seas, because the duration of riding on the wave becomes longer. Therefore the guidance states that the phenomena is danger when the angle of encounter is in the range 135 < < 225. The stability reduction may become critical for wave length within the range of 0.6 to 2.3 ship’s length.
Because of wide range of courses that are determined as danger SMC is no able to find ship
pattern that will be safe. The probability of safety on the search pattern is equal 0, so a ship is no suitable for SAR action. When this phenomena is taken under consideration the number of ships that may be used during action decreases dramatically. The number of ships suitable for SAR action from
example is presented in figure 5. Detailed infor-mation about each ship from example is presented in table 5. In assumed weather conditions, when most waves have period about 9 seconds, SMC should take into consideration only 5 from 23 ships.
Conclusions
Automation of calculation will help SMC to take proper decision in short time. Obtaining informa-tion about ship from captain may be replaced by data from LRIT (Long Range Identification and Tracking) system connected with AIS database and some statistical information. It will help to elimi-nate the worst ships and allowed SMC to ask mas-ters of more usable ships who didn’t decided to proceed to the scene of a distress to reconsider their decision. There are a lot of works to develop and improve the model of Bayesian network of usability ship for SAR action.
The discussion about which phenomena is im-portant should be continued. How much reduction of intact stability is dander for ship is very impor-tant question because if we take into consideration this phenomena only few merchant ship will be suitable for SAR action.
References
1. International Aeronautical and Maritime Search and Res-cue Manual. IMO/ICAO, London / Montreal 1999. 2. STAROSTA A.,BURCIU Z.: Niezawodność obiektu w
syste-mie antropotechnicznym – statku handlowego w akcji SAR. Niezawodność systemów antropotechnicznych, XXXVII Zimowa szkoła niezawodności, Szczyrk 2009. 3. STAROSTA A.: Zakłócenia transportu morskiego – ciężkie
warunki pogodowe – sposób oceny. Logistyka 2009, 4. 4. BURCIU Z.: Bezpieczeństwo w transporcie morskim. Tom
1. Poszukiwanie i ratowanie życia na morzu w ujęciu sys-temowy. Monograph – working version.
5. FRIIS-HANSEN A.: Bayesian networks as a decision support tool in marine applications. PhD thesis. Technical Univer-sity of Denmark, December 2000.
6. STAROSTA A.: Application of Bayesian network to estimate merchant ship usability for SAR action. Journal of KONES 2006, Vol. 13, No. 3.
7. STAROSTA A.: Usage of AIS information to estimate mer-chant ship usability for SAR action. Proceedings of XIII International Scientific and Technical Conference on Ma-rine Traffic Engineering, Malmo, Sweden 2009.
8. Międzynarodowa konwencja o bezpieczeństwie życia na morzu, 1974 – SOLAS ’74, Tekst jednolity, PRS 2006. 9. www.marinetraffic.com
10. GeNIe 2.0, http://genie.sis.pitt.edu
11. MSC.1/Circ. 1228 Revised guidance to the master for avoiding dangerous situations in adverse weather and sea conditions.
Table 5. Ship suitable for assumed SAR action
Tabela 5. Statki przydatne do prowadzenia akcji SAR w zało-żonych warunkach pogodowych
Name L [m] Tw [s] 6 7 8 9 10 11 KRASNODAR 249 OK OK OK OK NO NO ALEKSANDR SIBIRYAKOV 132 OK OK NO NO NO NO BRITISH ROBIN 252 OK OK OK OK NO NO FINNLADY 219 OK OK OK OK NO NO LISCO MAXIMA 199 OK OK OK NO NO NO SPICA 151 OK OK NO NO NO NO CONTAINER-SHIP 6 155 OK OK NO NO NO NO SUNNAVIK 124 OK NO NO NO NO NO BRO ALMA 143 OK OK NO NO NO NO GREAT JADE 225 OK OK OK OK NO NO ARESSA 95 OK NO NO NO NO NO KING ERIC 182 OK OK OK NO NO NO BAUGE 30 NO NO OK OK OK OK VISCARIA 83 NO NO NO NO NO NO INGVILD 99 OK NO NO NO NO NO AZURYTH 105 OK OK NO NO NO NO CARRY 185 OK OK OK NO NO NO SCARLET STAR 183 OK OK OK NO NO NO GRANIT 64 NO NO NO NO OK OK STK-1007 82 NO NO NO NO NO OK SYAM 81 NO NO NO NO NO OK SORMOVSKIY 3003 114 OK NO NO NO NO NO EMMA 90 NO NO NO NO NO NO Fig. 5. Number of ship suitable for SAR action due to wave length
Rys. 5. Liczba jednostek przydatnych do prowadzenia akcji SAR w zależności od długości fali
0 5 10 15 20 25 10 35 60 85 110 135 160 185 210 235 260 285 Wave length [m] Wave lenght [m]
