maritime risk - dnvresearch.dnv.com/skj/present/marrisk-skj.pdf · maritime risk an overview rolf...

MARITIME RISK

An Overview

Rolf Skjong, Dr, Chief scientistChalmers, April 21, 2009

N62°30’12.83’’ E6°06’07.94’’

© Det Norske Veritas AS. All rights reserved Slide 2

Contents1. Introduction-Background

2. FSA – Method

3. FSA, RBD, GBS -relations

4. FSA – Results

5. General conclusions

6. Outlook

7. References


Purpose of FSA (1/2)

FSA is intended to be a tool for rule-making at IMO:• To make the decision process at IMO more rational, reduce ad-hoc

proposals/implementation,give less room for politics

• To provide a proactive, holistic approach, comprising technical as well as operational aspects


Purpose of FSA (2/2)To generate information achieved in a way which is structured, systematic, comprehensive, objective, rational, auditable and documented

To demonstrate that suitable techniques have been applied and sufficient efforts have been made to identify hazards and to manage the associated risk


FSA – A risk Based ApproachDefinition of Goals, Systems, Operations

Hazard Identification

Cause and Frequency Analysis

Consequence Analysis

Risk Summation

Risk Controlled?

Options to decrease Frequencies

Options to mitigate Consequences

Cost Benefit Assessment

Reporting

NoNo

Yes

Scenario definition

Preparatory Step

Step 1Hazard Identification

Step 2Risk Analysis

Step 3Risk Control Options

Step 4 Cost Benefit Assessme

Step 5 Recommendations for Decision Making


Introduction

Risk Based Regulations - Status At IMO • UK proposed FSA at IMO in 1995

• Referred to safety case for offshore• Resulted in FSA as risk assessment process to justify Rules/Regulations

• FSA Interim Guidelines, v0, 1997• FSA Guidelines, v1, 2002• FSA Guidelines, v2, 2007


Should FSA become mandatory?

GUIDELINES ON THE ORGANIZATION AND METHOD OF WORK OF MSC AND MEPC AND THEIR SUBSIDIARY BODIES (MSC-MEPC-Circ.1)

2.10.5 ‘has the analysis of the issue sufficiently addressed the cost to the maritime industry as well as the relevant legislative and administrative burdens?’

2.14 ‘a higher priority should be assigned to items that can be shown, or estimated, to have the greatest effect on safety of life, prevention of serious injury, protection of the marine environment and the highest ratio of benefit to be gained from the implementation of the proposal compared with the cost of its implementation’

2.23.2.2.4 ‘do the benefits justify the proposed action?’


Risk based Rules & Risk Based Design

Definition of Goals, Systems, Operations




Risk Summation

Risk Controlled?




Reporting

NoNo

Yes

Scenario definition

Risk Assessment Method

Generic ShipSpecific Ship

(in e.g. specific environment)

Risk Based Rules Risk Based Design

MSC86/5/4


Risk Based Regulation: GBS - SLA

Tier I

Tier III

Tier V

Tier IV

Tier II

Applicable Industry Standards & Codes of Practice

Prescriptive Regulations & Class Rules

VerificationProcess

FunctionalRequirements

Goals





Risk Summation

Risk Controlled?




Reporting

NoNo

Yes

Scenario definition

RA to Justify


Risk Based Design (RBD)

Tier I

Tier III

Tier II

RBD VerificationProcess

FunctionalRequirements

Goals





Risk Summation

Risk Controlled?




Reporting

NoNo

Yes

Scenario definition

RBD


FSAs

First FSAs High speed craftHelicopter landing area on passenger shipsBulk carrier safety

Other FSAsNavigation – Large Passenger ShipsMandatory Carriage Requirements for ECDISInert Gas System for Chemical Tankers < 20.000 dwt


FSAs by SAFEDORPurpose of submitting FSAs to IMO from SAFEDOR

Document safety level – for reference in Risk Based DesignShip types selected for FSA same as Risk Based Design projects

Contributions from SAFEDORFSA Container ships (MSC83)-GL, AMTW, DÖHLE,SSPAFSA Liquefied Natural Gas ships (MSC83)- DNV, NAVANTIA, IST, LMGFSA Cruise (MSC85)-DNV, CARNIVALFSA RoPax (MSC85)-SSRC, FSG, LMG, COLOR, DNVFSA Oil Tankers (MEPC58) - NTUA, GL, DNV, SSRC, DMA, ASMEFSA-Step 1: Container fire on deck (FP53)FSA Dangerous Goods in Open-Top Container Vessels (Not Submitted yet)


FSAs by SAFEDOR

FSAs have been carried out by different teams

All use the IMO FSA Guidelines as basis

FSAs are therefore quite similar, but for reuse of models more standardisation is beneficial

- Partial models for probabilities, consequence models, reliabilities, availability, vulnerability

- Software need clearly defined interfaces- Relevant both for FSA & RBD


Standardization

6 accident scenarios cover 98.8% of serious ship accidents

98,84.38.810.7 19.524.031.5Ratio of total (%)

1331357611671420259131994195Total

3755323414356639691256Tanker

619237380105141190Roro

4202927967486108Reefer

103423154180162160348Pass./Ferry

15741738172256Offshore

37583901923097648531198Dry Cargo

86586993124322244Container

20952519473376Combination

249662269170635613719Bulker

TotalFndCntF/EW/SColH/M


Standardization

6 accident scenarios cover 99.4 % fatalities in ship accidents

99.411.211.612.633.839.2Ratio of total (%)

295029353413719981157Total

402143616333Tanker

1934210Roro

5071011418Reefer

170018260221553657Pass./Ferry

2511201Offshore

485113653531247Dry Cargo

7613342Container

615Combination

187965408144Bulker

TotalCntH/MW/SColFndF/E


FSAs by SAFEDOR

One clear benefit of standardization- Decision Parameters and Risk Acceptance- All FSAs present Individual risk, Societal Risk (FN),

NCAF, GCAF, as suggested in FSA Guidelines, CATS (if relevant) – easy to read and compare results

- In www.safedor.org ‘Risk Evaluation Criteria’ a table of 50 RCO and their cost effectiveness is listed

- This report has been downloaded > 25.000 times!


FSA – Ship Types


Individual Risk 2007 data


Individual Risk – 2007 data


Loss – 2007 data


Individual Risk – as reported in FSA

10-6Negligible

0,65 10-4RoPax

0,80 10-4Cruise

2,10 10-4Tanker

6,45 10-44,90 10-41,55 10-4LNG

2,25 10-4Container

10-3Intolerable

Total Personal AccidentShip Accidents

Individual Risk For Crew


FSA Container

Total Risk for Container Vessels

1E-05

1E-04

1E-03

1E-02

1E-01

1 10 100Number of fatalities [N]

Freq

uenc

y [F

] of N

or m

ore

fata

litie

s pe

r sh

ip y

ear

SAFEDOR MSC72/16 upper ALARP boundary lower ALARP boundary


FSA Container

4.079.87Implementation of BRM guidelines 32

< 01.14Track control system31

6.6212.27 ECDIS 30

79.6785.47Additional officer on the bridge (on demand)25 b)

191.45197.25Additional officer on the bridge (always )25 a)

< 00.22AIS integrated with radar22

< 05.27Improved bridge design15

189.02203.02Reduced amount of undeclared dangerous goods11

383.23407.12Bow camera system (including night vision)10 b)

85.34109.35Bow camera system (standard)10 a)

5.7211.66Improved navigator training5

< 015.32Second bilge alarm in cargo holds (open-top design)4 d)

25.7176.83Second bilge alarm in cargo holds (conventional design)4 c)

< 0 1.72High bilge level alarm in cargo holds (open-top design)4 b)

< 08.64High bilge level alarm in cargo holds (conventional design)4 a)

< 0 28.67Increased efficiency of bilge system (open-top design)3 b)

96.69143.72Increased efficiency of bilge system (conventional design)3 a)

NCAF[106 $]

GCAF[106 $]Risk Control OptionRCO

No.

GCAF and NCAF values associated with each risk control option.


FSA - LNG


Risk Control Options (RCO)

Cost-effective RCOs exist – all related to navigational safety

Risk based maintenanceECDISTrack control systemAIS integrated with radarImproved bridge design


Lack of qualified LNG crew

New LNG trades – e.g. cold climate

New LNG operators

New trading patterns – e.g. more short-term contracts

Bigger and faster ships

New propulsion systems – with on-board reliquefactionsystems

FSA LNG: Safety critical trends and developments

Current trends that may influence future safety level of LNG shipping (not analyzed):


Prior perception – LNG vessels associated with high level of safety – substantiated by the FSA:

Risk levels within ALARP areaMost RCOs not cost-effective

However, cost-effective RCOs identifiedAll related to navigational safetyShould be implemented to make risk ALARP

Conclusions and recommendations


FSA Tankers

The present analysis covers crude oil tankers of the following types:PANAMAX (60,000 DWT – 79,999 DWT)AFRAMAX (80,000 DWT -119,999 DWT)SUEZMAX (120,000 DWT -199,999 DWT)Very Large Crude Carriers (VLCC; 200,000 DWT -320,000 DWT)Ultra-Large Crude Carriers (ULCC; more than 320,000 DWT).

Major oil trade movements


Risk analysis of Large Tankers Frequency assessment

Only the six (6) events that potentially lead to LOWI (Loss Of Watertight Integrity) are taken into account.

The full risk model should include except LOWI accidents, also "machinery failures", "failures of hull fittings" and "unknown reasons.

TankerCasualties

Collision Contact Grounding Fire Explosion NASF MachineryFailure

Failure ofHull Fittings Other

Potential LOWI Events


Risk analysis of Large Tankers Frequency assessment

DH ships: 1.93E-03All ships: 5.74E-03NASF

1.90E-03Explosion

3.68E-03Fire

7.49E-03Grounding

3.72E-03Contact

1.03E-02Collision

Frequency of accidentInitial event

Frequency by incident category, Covered period 1990-2007

Frequency of incidents, Historical Data

0.00E+00

2.00E-02

4.00E-02

6.00E-02

8.00E-02

1.00E-01

1.20E-01

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Incident Year

Freq

uen

cy p

er s

hip

year

Collision Contact Grounding Fire Explosion NASF

Source: NTUA-SDL

Oil TankersHistorical Data, Covered Period 1990-2007

Collision, 265, 32%

Contact, 96, 11%

Grounding, 193, 23%

Fire, 95, 11%

Explosion, 49, 6%

NASF, 148, 17%

Source: NTUA-SDL Tanker casualty databaseSample data: 846 incidents


Risk analysis of Large Tankers Risk assessment results (2)

Potential Loss of Cargo (PLC)

Oil Tankers (DWT>=60000)

Results of Risk AnalysisEstimated Potential Loss of Cargo, in tonnes per shipyear

1.09E+00 1.44E+00

1.23E+011.30E+01

3.17E-01

1.86E+01

6.16E+00

2.38E+01

0.00E+00

5.00E+00

1.00E+01

1.50E+01

2.00E+01

2.50E+01

Collision-Struckship

Contact withfixed installation

Contact withfloating object

PoweredGrounding

Drift Grounding Fire Due tointernal source

Explosion-Operational

phase

NASF



Individual risk for crew

A crew of 30

Fatality rate of 1.27E-02 per ship-year

Individual risk of 4.23E-04 per year.

Three shifts of crew alternate -> individual risk becomes 1.41E-04.

- ALARP Area



F-N curve for tanker crew

Risk level, Oil Tankers DWT>60000Historical data, Period 1990-2007

1.00E-06

1.00E-05

1.00E-04

1.00E-03

1.00E-02

1.00E-01

1 10 100

Number of fatalities (N)

Freq

uenc

y of

N o

r m

ore

fata

litie

s

Total risk level Collision Fire Explosion


Risk analysis of Large Tankers Main Conclusions -Identified high risk areas

In terms of potential loss of crew life, three areas of main concern or generic accident scenarios were identified:

- Collision scenarios of the struck ship- Fire scenarios due to internal source initiation- Explosion scenarios.

In terms of potential loss of oil cargo, four areas of main concern or generic accident scenarios were identified

- Collision scenarios of the struck ship- Powered grounding- Fire scenarios due to internal source initiation- Explosion scenarios.


Analysis Results(1)


Cost Benefit analysis Results (2) –RCO 7

RCO 7: Ship Design Modifications were analyzed for every ship type and as such the results are presented in separate tables


Recommendations


FSA Cruise

36 mBreadth

8.5 mDraft

290 mLength

4,000Passengers + Crew

1,200Crew

2,800Passengers

22 knotsSpeed

110,000 GRTSize

ValueShip parameters

High level FSA – No specific ship designIn some cases parameters are taken to Represent a more specific ship-type/size etc: Post Panamax


FRA Cruise: Individual risk


Cruise FN

FN Cruise ships

1E-05

1E-04

1E-03

1E-02

1E-01

1E+00

1 10 100 1000 10000

Fatalities [N]

Freq

uenc

y of

N o

r m

ore

fata

litie

s pe

r sh

ip y

ear

ALARP

NEGLIGIBLE

INTOLERABLE


Cruise FN

FN Cruise ships

1E-06

1E-05

1E-04

1E-03

1E-02

1E-01

1E+00

1 10 100 1000 10000

Fatalities [N]

Freq

uenc

y of

N fa

talit

ies

or m

ore

per

ship

yea

r

Collision

Contact

Grounding

Fire/Expl

Lower

Upper

93% from collision & grounding85% from catastrophic


FSA Cruise

6) Increased steel weight 1200-1500 t7) Increased deck area approx. 4500 m28) Increased deck area approx. 2500 m2

1) Increased steel weight 50-100 t2) Increased deck area approx. 100 m2 per deck3) Increased steel weight 50-100 t4) Increased deck area approx. 60 m2 per deck5) Reduced deck area approx. 2500 m2 on bulkhead deck

8)6)0.8992.52.711.233.2285

RCO 1+2+3: Increased Freeboard

0.5 mReserve buoyancy on

bulkhead deckIncreased breadth 1mIncreased GM 0.5 m

60% add. Deck

7)6)0.8752.52.210.733.2285

RCO 1+3: Reserve buoyancy on

bulkhead deckIncreased breadth 1 m Increased GM 0.5

mOne additional deck

5)0.8362.02.210.732.2285

RCO 3: Reserve buoyancy on

bulkhead deck

4)3)0.852.02.711.232.5285

RCO 2: Increased freeboard

0.5 m

2)1)0.852.52.210.732.7285RCO 1: Increased GM

0.5 m

0.802.02.210.732.2285As is

(vessel in Table 2)

Benefit

factors

Cost factor

s

Attained

Subdiv. index A

GM (m)

Freeboard (m)

Freeboard

depth (m)

Breadth (m)

Subdiv.

Length (m)Configuration

Stability RCOs, alterations to the original ship design.

2) Net present value, 5% interest rate, 30 years

1) Per ship per lifetime, assumed 30 years

-39 900 0003 340 000162 000 00012 500 0003.75RCO 1+2+3:Combined Bouyancy,

GM and Freeboard

- 97 800 0004 390 000291 000 00012 500 0002.85RCO 1+3: Combined

Buoyancy & GM

- 205 000361 000540 000344 0000.95RCO 27: BRM

238 900 000239 100 000286 000322 800 0001.35RCO 3: Added

buoyancy

- 2 770 0001 120 0008 160 0002 350 0002.10RCO 2: Increased Freeboard

- 5 260 0001 120 00013 400 0002 350 0002.10RCO 1: Increased

GM

$$$$# of saved lives1)

NetCAFGrossCAFBenefit2)

ΔBCost2)

ΔCRisk reduction

ΔR


FSA Cruise

The following risk control options were shown to be effective in this particular design study and may warrant further investigation:

- Improved bridge design (above SOLAS)- ECDIS - Electronic Chart Display and Information System- Increased Simulator Training for Navigators- Improve the damage stability


RoPax – Historic Risk

1.00E-06

1.00E-05

1.00E-04

1.00E-03

1.00E-02

1.00E-01

1 10 100 1000

Number of Fatalities N

Freq

uenc

y of

N o

r Mor

e Fa

talit

ies

(per

yea

r)

NEGLIGIBLE

ALARP

INTOLERABLE

World-Wide 1994 - 2006

NW Europe1978 - 1994


RoPax Improvement since NWE project

Collision. 40% frequency reduction

Grounding. 52% frequency reduction

Impact. 74% frequency reduction

Flooding from other causes. Unchanged.

Fire. 17% frequency reduction

Overall Frequency. 57% frequency reduction


RoPax – Risk Model

1.00E-06

1.00E-05

1.00E-04

1.00E-03

1.00E-02

1.00E-01

1 10 100 1000

Number of Fatalities N

Freq

uenc

y of

N o

r Mor

e Fa

talit

ies

(per

yea

r)

NEGLIGIBLE

ALARP

INTOLERABLE


General Cargo Ships

On-going work in IACS

Identified serious problem of underreporting

LRFP results is not possible to believe. Need for alternative sources of data to estimate degree of reporting

100≤ gt<500 500≤ gt

<1,000 1,000≤ gt<20,000 20,000≤

gt Total

IACS, ModernIACS, Old

non-IACS, Modernnon-IACS, Old

UNKNOWN, Modern

UNKNOWN, Old

Total

0.0000

0.0100

0.0200

0.0300

0.0400

0.0500

0.0600

0.0700

0.0800

Accident rates


Conclusions

All SAFEDOR FSAs point at measures to improve navigational safety (preventing collisions & groundings) as cost effective measures

- Electronic Chart Display and Information System- Agreed at MSC85, for adoption at MSC86- Supported also by other FSA studies and ENC Coverage

studies: MSC78/4/2, NAV50/11/1, NAV51/10, MSC81/24/5, NAV54/14


Conclusions

FSAs on Cruise Ships and RoPax concluded that damage stability can be improved cost-effectively.

Collision risk is dominating (small probability, potentially large consequence)

Answers from new EU Project: GOAL based Damage Stability (GOALDS). - Project involves yards, owners, classification societies and academia (17

partners)

FSA Tanker also demonstrate cost effective ways of reducing oil outflow- Final conclusion depend on criteria (CATS)


Outlook

FSA should be made mandatory for new and major revisions of instruments (need to define criteria for use)

Flag States need to improve accident reporting to improve quality of FSA (GISIS)

Issues of underreporting

Occupational risk has too low priority compared to other risk to life –often left out of the analysis

Need for incentives to carry out FSA- FSAs carried out in 1+ years- FSA review 2-3 years?


Underreporting

Most FSAs use LRFP (Only ship accidents)- Fatalities (personal) > Fatalities (Ship accidents)

Estimated 441 accidents in NOR/NIS tanker fleet

40,5%

30%12%

41,5%


Outlook

Standardized risk-based tools are and will be developed

Standardization of risk models needed- Reuse of information, tools etc

Risk criteria need to be maintained, preferably by IMO


CATCH – Cost of Averting a Tonne of CO2-eq Heating effect


References

MSC85/INF.3MSC85/17/2RoPax

MSC85/INF.2MSC85/17/1Cruise

MEPC58/INF.2MEPC58/17/2Crude oil Tanker

MSC83/INF.8MSC83/21/2Container

MSC83/INF.3MSC83/21/1LNG

Full FSAMain SubmissionFSA

Other FSA reports at http://research.dnv.com/skj/

maritime risk - dnvresearch.dnv.com/skj/present/marrisk-skj.pdf · maritime risk an overview rolf...

Documents