Marine Accident Analysis With GIS

8/12/2019 Marine Accident Analysis With GIS

1/9


2/9

Marine Accid ent Analysis w ith GIS22

type and accident density, and such marine accidentareas with high risk have been determined.

2. Method

In this study, data pertaining to 585 marine accidentsoccurred during the years between 2007 and 2011,which have been recorded into GISIS, have beenexamined. The vessels include bulk dry, general cargo,container, ro/ro, passenger, passenger ro/ro, oil tanker,gas tanker and chemical tankers. A database has beencreated in GISIS system, having grouped the datarecorded according to accident type, vessel name, flag,accident size and accident coordinates.

In the study, it has been intended to create marineaccident chart for marine accidents, and for such purpose Microsoft Excel 2010 and ArcGIS 10 programs have been used. By means of MicrosoftExcel 2010, the geographical coordinates in degrees,minutes have been converted into numerical valuessuitable with ArcGIS 10 program. To prevent anymistake, the accident data entered into the system have been analysed singly after the conversion. Marineaccident chart has been created with the data processedonce have been transferred into ArcGIS 10 program,and the created marine accident chart has beenseparated into ranges and risky areas have beendetermined.

3. Marine Accident Data

The study includes data regarding the marineaccidents occurred in 2007-2011, which are recordedinto GISIS marine accident and incident database and

classified as very serious, serious, less serious. Veryserious accidents are the accidents including vessel loss,loss of life and solid pollution. In the 37th meeting ofMarine Environment Protection Committee (MESC),solid pollution has been defined as the pollutionaffecting the coast state or flag state, which generatesserious destructive impacts on environment andrequires preventive measures. Serious accidents are theincidents such as fire, explosion, collision, grounding,

contact, heavy weather damage, ice damage, crackswithin the body of vessel, hull flaw which do not havevery serious accident nature. Blockage of main engine,underwater damage, accommodation deck damage,solid structural damages, pollution in small quantities,damages requiring shore assistance or tugboat use areincluded in this group. Less serious accidents are theaccidents which are not as notable as very seriousaccidents and serious accidents and it aims at recordinguseful information [11]. All of the accidents analysedhave been suffered by commercial vessels. Fig. 1contains distribution of vessel accidents in percentagesin terms of vessel type.

A hundred ninety six of such accidents have causedto loss of life, loss of vessel or serious environmental pollution (very serious accident). Three hundredtwenty four of such have resulted in injury, smallscaled environmental pollution or vessels becomingunsuitable for navigation (serious accident) and theremaining 65 accidents are less serious accidents otherthan abovementioned accidents.

Flag state is responsible for that the vessel, whichhas its flag, be fitted out in accordance withinternational life, property safety, national andinternational rules. Therefore, flag state practices arean important factor for marine accidents [10]. Table 1indicates the distribution of vessel accidents recordedin GISIS according to flag state. Among accident datawhich have been transferred to the database, 188accidents have been caused by the vessels with Panamaflag, 83 of them with United Kingdom flag, 30 withDenmark flag, 24 with Malta flag and 23 with Liberia

flag (Fig. 2).Marine accidents have been analysed in seven

categories, including collision, grounding,fire/explosion, flooding/sinking, damages to ship orequipment, occupational accident and other accidents.As a result of examination of marine accident data, ithas been found that the most frequent 3 accident typesare collision, grounding and damages to ship orequipment (Table 1).


3/9

Marine Accid ent Analysis w ith GIS 23

Fig. 1 Distribution of vessel accidents according to vessel type.

Table 1 Distribution of vessel accidents according to accident type [11].

TYPE of SHIPTYPE of MARINE ACCIDENTS

Collision Grounding Damages to ship orequipmentOccupationalaccident

Fire/explosion

Flooding/sinking Other

Bulk Dry/ General Cargo 97 66 52 20 23 10 20Container 41 12 13 19 4 2 5Oil Tanker 14 10 4 9 6 2 4Chemical Tanker 15 6 9 7 4 2 1Gas Tanker 2 1 1 4 3 - -Passanger Ship 4 7 3 4 3 3 6Passanger Ro/Ro Cargo 15 2 5 2 4 1 2

Ro/Ro Cargo 7 8 5 6 7 2 1Total 195 112 92 71 54 22 39

Fig. 2 Distribution of vessel accidents according to flag state.


4/9


4. GIS

GIS is an information system which performs

integratedly the functions of collection, storage, processing and presentation to customer of theinformation, either graphical or non-graphical, whichhave been obtained through data or observations basedon location [12]. GIS has 5 important components.These are hardware, software, data, human and method.Effective use of GIS depends on organized use of allcomponents. The most important component is the datawhich requires the maximum time and cost amongothers [13].

Achievement of GIS project depends on availabilityof data with appropriate structure. In GIS, thegeographical data consists of 2 main groups: verbal andgraphical. In verbal data, the information, whichindicates attributions of geographical data, is stored.On the other hand, in graphical data, the information,which indicates the form and location of objects in theworld, is stored. Since the GIS is based on relationaldata model, table data and graphical data may be

connected to each other. In order to transfergeographical data into computer, to process and displayit in computer, said raw data must primarily beconverted into a form cognizable by computer. Such aconversion is possible through conversion of data intonumerical form [12, 14].

GIS technology has been a popular tool forvisualization of accident data and analysis. Accidentanalysis studies aim at the identification of high rateaccident locations and safety deficient areas [15]. Theuse of GIS is the most effective way of examining andevaluating the results of analyses which use a multitudeof data and different criteria [16]. GIS is used to locateaccidents on a digital chart and realize theirdistribution [17].

4.1 Implementation of GIS in The Study

Marine accidents have been analyzed in 5 stages inArcGIS 10 program. In the first stage, marine accident

chart has been created by entering marine accident datainto ArcGIS 10 program. Fig. 3 indicates the distributionof marine accidents according to vessel type. In the 2nd

stage, the world chart has been divided into 10 degreelatitude and longitude intervals and accident density has been identified for marine accidents.

Accident density chart has been created for generalmarine accidents in Fig. 4, and for collision andgrounding accidents in Fig. 5. In the study, marineaccident areas have been classified as very high risk,high risk, risk, moderate risk and low risk marine area.Definitions are associated with quantity of accidents,depending on the size of accident area. Accordingly,ranges with density of more than 30 accidents are veryhigh risk marine areas; ranges with density of between20 and 30 accidents are high risk marine areas; rangeswith density of between 10 and 20 accidents are riskymarine areas; ranges with density of between 5 and 10accidents are moderate risk marine areas; and rangeswith density of less than 5 accidents are low riskmarine areas.

In the 3rd stage, marine areas where marine

accidents are intensive have been analysed. In thisstage, the North Europe and Far East have beendetermined as the marine areas with the most intensivemarine accidents. In the 4th stage, such marine areashave been focused on. Therefore, such marine areashave been divided into grids with 2 degree latitude andlongitude intervals. So it has been made possible todisplay and interpret the marine accidents in the NorthEurope and Far East comprehensively (Figs. 6, 7, 8 and9). It has been found that such process is required fordetermination of marine accident locations. Suchcreated charts enable assessment of marine accidents interms of location.

In the last stage, marine accident charts have beenassessed and interpreted. In order to providecomprehensibleness of the assessment, ranges with highaccident density have been numerated from 1 to 16 (Figs.6 and 8). Table 2 indicates quantitative distribution ofmarine accidents according to range number.


5/9


Fig. 3 Marine accidents chart for merchant ships.

Fig. 4 Density chart for all of marine accidents.


6/9


Fig. 5 Density chart for collision & grounding accidents.

Fig. 6 Density chart for North Europa Countries.


7/9


Fig. 7 General distribution of types of marine accidents in the North Europe.

Fig. 8 Density chart for Far East Countries.


8/9


Fig. 9 General distribution of types of marine accidents in the Far East.

Table 2 Quantitative distribution of marine accidentsaccording to range number.

Polygon Number

Total Accident Numbers

Collision&GroundingAccident Numbers

For Fig. 6

1 7 42 7 63 12 64 9 55 12 116 8 87 8 5

For Fig. 8

8 11 79 8 410 11 611 12 1112 8 813 7 414 7 615 8 616 7 4

5. Results

Marine areas with the highest risk, in terms ofmarine accidents, are North Europe and Far East.

The marine areas with very high risk and high risk inthe North Europe are Strait of Dover between Englandand France, Belfast shores in Ireland, Hamburg inGermany, east coasts of Denmark and in the southcoasts of Norway the seas surrounding Kattegat, GreatBelt and Copenhagen.

Marine accidents in the Far East are intensiveespecially in the coasts of Japan and China.

In the Far East, marine areas with intensive marineaccidents are Kanmon Strait, Urage Channel andBungo Strait in Japan, Shanghai, Ningbo and HongKong in China.

As a consequence of a general assessment of allaccidents, it has been found that especially collisionand grounding accidents have high intensity in theseareas.


9/9


6. Conclusion

GIS is a basic guide which provides interpretation ofresults on a chart. It enables visual interpretation of

accidents. In this study, marine accident areas withhigh intensity have been visually interpreted. It isimportant for determining measures required to detectmarine accident areas with high intensity and preventmarine accidents in such areas. Coastal areas and straitsare the marine areas where collision and groundingaccidents have the highest density.

GIS is a significant instrument for the follow-up andmapping of marine accidents. In this context, the usageof GIS in the field of maritime must be extended. Itshall be important to determine the common reasons ofmarine accidents in the marine areas, with high densityof accident, for preventing similar accidents.

References

[1] L. H. Ringdahl, Safety Analysis, New York:Taylor&Francis, 2001.

[2] N. Akten, Shipping accidents: A serious threat formarineenvironment, J. Black Sea/MediterraneanEnvironment 12 (2006) 269-304.

[3] S. E. Chapman and N. Akten, Marine casualties in theTurkish Straits A way ahead, Seaways (1998) 6-8.

[4] S. Kristiansen, Maritime Transportation SafetyManagement and Risk Analysis, ElsevierButterworth-Heinemann, 2005.

[5] IMO, STCW including 2010 manila amendments STCWconventions and STCW code, 2011.

[6] N. R. Council, Human error in merchant marine safety,Washington: National Academy Press, 1976.

[7] W. A. ONeil, The human element in shipping, Journal ofMaritime Affairs 2 (2) (2003) 95-97.

[8] M. Celik, S. M. Lavasani and J. Wang, A risk-basedmodelling approach to enhance shipping accidentinvestigation, Safety Science 48 (1) (2010) 18-27.

[9] O. Ugurlu, Risk Analysis of Oil Tanker Accidents, PhD,Karadeniz Technical University, TRABZON, 2011.

[10] IMO, Casualty-Related Matters Reports on MarineCasualties and Incidents, London, 2005.

[11] IMO, available online at: http://www.gisis.imo.org/,2010.

[12] T. Yomralioglu, Basic Concepts and Applications ofGeographic Information System,stanbul: Seil Ofset,2000.

[13] S. R. Somer, Geographic information system in localgovernment: A commentary, PhotogrammetricEngineering and Remote Sensing 53 (10) (1987)1379-1382.

[14] R. Nisanci, The production of pixel based urban land valuemaps with nominal valuation method using GIS,Karadeniz Technical University, Trabzon, 2005.

[15] S. Erdogan, I. Yimaz, T. Baybura and M. Gullu,Geographical information systems aided traffic accidentanalysis system case study: City of Afyonkarahisar,Accident Analysis and Prevention 40 (1) (2008) 174-181.

[16] I. B. Gundogdu, A new approach for GIS-supportedmapping of traffic accidents, in: Proceedings of theInstitution of Civil Engineers-Transport 164 (2) (2011)87-96.

[17] K. S. Ng, W. T. Hung and W. G. Wong, An algorithm forassessing the risk of traffic accident, Journal of SafetyResearch 33 (3) (2002) 387-410.

Documents

Marine Accident Analysis With GIS