ANNEX II PASSENGER VESSEL EVACUATION RISK DATA

Transcription

1 ANNEX II PASSENGER VESSEL EVACUATION RISK DATA

2 DET NORSKE VERITAS Report No Page No.. FSA of HLA on PassengerVessels II.i ANNEX II - PASSENGER VESSEL EVACUATION RISK DATA Contents II.1 EVACUATION EXPERIENCE...1 II.1.1 Data Sources...1 II.1.2 Cruise Ship Evacuations...1 II.1.3 Merchant Ship Evacuations...1 II.2 EVACUATION FREQUENCIES...5 II.2.1 General Approach...5 II.2.2 Total Losses of Passenger Ships...5 II.2.3 Evacuations of Cruise Ships...5 II.2.4 Evacuations of Ro-Ro Passenger Ferries...5 II.2.5 Total Losses of Merchant Ships...6 II.2.6 Evacuations of Merchant Ships...6 II.3 EVACUATION FATALITIES...7 II.3.1 Causes of Evacuation...7 II.3.2 Overall Fatality Rates...8 II.3.3 Effect of Accident Scenario on Fatality Rates...9 II.3.4 Causes of Death...10 II.3.5 Time Available For Evacuation...13 II.3.6 Rescue by Helicopter...15 II.3.7 Rescue Ship Availability...17 II.4 REFERENCES...18

3 DET NORSKE VERITAS Report No Page No.. FSA of HLA on PassengerVessels II.1 II.1 EVACUATION EXPERIENCE II.1.1 Data Sources Although no comprehensive list of passenger vessel evacuations is known, a selection of such events is given in Appendix I, based on the following sources: Hooke, N. (1997) : Maritime Casualties, , LLP, London. This lists all total losses for Watson, M.H. (1995) : Disasters at Sea, 2nd edition, PSL. This lists most major accidents involving passenger vessels. Lloyds Casualty Reports. These list all significant marine accidents. Since 1992, the summary of major casualties identifies most passenger vessel evacuations. II.1.2 Cruise Ship Evacuations A list of 48 cruise ship evacuations is given in Table II.1.1, summarising the descriptions in Appendix I.1. This covers the major events since 1963, but is probably more comprehensive since It covers any case where passengers were taken off a cruise ship at sea or in port following an accident. II.1.3 Merchant Ship Evacuations Pyman & Lyon (1986) reported a previous analysis of merchant ship evacuations conducted by Technica. This lists 169 evacuations of merchant ships, mainly from , where lifeboats or liferafts were used. It also lists 87 cases involving fishing vessels, either where the ship was abandoned or where the crew went down with the ship. This is not a comprehensive collection, as it only covers events where reporting was adequate to identify the means of evacuation and the number of people evacuating. Table II.1.2 gives a more recent collection of 34 merchant ship evacuations (excluding fishing vessels), based on Lloyd's Casualty Reports, covering the period Jan-Mar It includes any case where passengers or crew were taken off a ship over 100 GRT (except fishing vessels) in an emergency, unless the ship was in port, or where they sank with the ship. It is believed to be a comprehensive collection for the period, although it may omit events where evacuation was not mentioned in the Lloyd's Casualty Reports.

4 DET NORSKE VERITAS Report No Page No.. FSA of HLA on PassengerVessels II.2 Table II.1.1 Cruise Ship Evacuations, DATE SHIP EVENT NUMBER EVACUATED* EVACUATION TYPE FATAL- ITIES 22 Dec 63 Lakonia Fire 1036 Emergency Nov 65 Yarmouth Castle Fire 549 Emergency 89 8 Apr 66 Viking Princess Fire 494 Emergency 2 21 May 68 Blenheim Fire? Emergency 0 19 Jul 70 Fulvia Fire 719 Emergency 0 8 Jan 71 Antilles Grounding 650 Emergency 0 28 Aug 71 Heleanna Fire 1094 Emergency Sep 72 Caribia Grounding 880 Precautionary 0 3 May 73 Knossos Fire 186+ Emergency 0 12 Sep 74 Cunard Ambassador Fire 290- Emergency 0 7 Jan 76 Prinsess Sissy Grounding 439 Precautionary 0 18 Jan 76 Mecca Fire 1181 Emergency 0 23 Dec 76 Patra Fire 441 Emergency Jun 77 Rasa Sayang Fire 993 Emergency 5 30 Mar 79 Angelina Lauro Fire? Emergency 0 4 Oct 80 Prinsendam Fire 524 Emergency 0 20 Aug 81 Syria Grounding 300 Precautionary 0 28 Aug 82 Mediterranean Star Fire 1000 Emergency 0 9 Mar 84 Scandinavian Sea Fire 946+ Emergency 0 29 Jul 84 Columbus C Impact 620+ Emergency 0 12 Feb 85 Chidabaram Fire 888 Precautionary Feb 86 Mikhail Lermontov Grounding 738 Emergency 1 31 Aug 86 Admiral Nakhimov Collision 1234 Emergency Aug 86 North Star Grounding? Precautionary 0 11 Nov 86 Turkmeniya Fire 300+ Emergency 0 18 May 88 Priamurye Fire 388 Emergency Oct 88 Jupiter Collision 585 Emergency 4 20 Jun 89 Maksim Gorkiy Impact 325? Precautionary 0 2 Jun 91 Pegasus Fire 247 Emergency 0 4 Aug 91 Oceanos Flooding 581 Emergency 0 7 Aug 92 Queen Elizabeth 2 Grounding 1815 Commercial 0 14 Aug 92 Seabourn Spirit Fire 326- Emergency 0 23 Aug 92 Royal Pacific Collision 534 Emergency 9 1 Mar 93 Ocean Princess Grounding 530 Emergency 0 4 Mar 94 Sally Albatross Grounding 1259 Emergency 0 19 Aug 94 Regal Empress Fire 1500 Emergency 0 30 Nov 94 Achille Lauro Fire 979 Emergency 4 10 May 95 Renaissance Six Grounding 183 Precautionary 0 22 May 95 Albatros Fire 565 Commercial 0 18 Jun 95 Celebration Fire 1760 Precautionary 0 23 Jun 95 Star Princess Grounding 1550 Commercial 0 22 Jul 95 Regent Star Fire Precautionary 0 27 Jul 96 Universe Explorer Fire 1003 Emergency 0 4 Aug 96 Gripsholm Grounding 600 Commercial 0 29 Aug 96 Hanseatic Grounding 153 Commercial 0 16 May 97 Albatros Grounding 504 Precautionary 0 14 Jul 97 Hanseatic Grounding 145 Precautionary 0 4 Oct 97 Romantica Fire 673 Emergency 0 * Includes fatalities prior to evacuation

5 II.3 Table II.1.2 Merchant Ship Evacuations, Jan-Mar 1997 DATE SHIP TYPE EVENT LOCATION NUMBER EVACUATED* FATAL- ITIES EVACUATION ROUTE HELICOPTER PRESENT? 2 Jan 97 Nakhodka Tanker Foundered Sea of Japan 32 1 Lifeboat helicopter Yes 4 Jan 97 Mehmet Kirbas Gen cargo Collision Mediterranean 10 0 Ship 5 Jan 97 Khromov Research Grounding Antarctic 36 0 Ferry 7 Jan 97 Prvic Gen cargo Grounding Algeria 0 Gen cargo ship 8 Jan 97 Sun Richie 3 Gen cargo? Foundered Japan 22 0 LPG carrier Yes 8 Jan 97 Onur K Gen cargo Foundered Mediterranean 11 5 Lifeboat ship/helicopter Yes 9 Jan 97 Cliff Wayne Eserman OSV Foundered Gulf of Mexico 0 Lifeboat tug 9 Jan 97 F E Trader Gen cargo Fire South China Sea 21 0 Lifeboat ship 10 Jan 97 Lawas Mewah Gen cargo Foundered South China Sea 16 0 Lifeboat? ship 10 Jan 97 Ru Yi 2 Gen cargo Collision Yangtze River Water? river craft? 11 Jan 97 Rauma 1 Tug Fire Baltic 5 0 Liferaft Ro-Ro cargo ship 29 Jan 97 Meratus Mas Gen cargo Foundered Indonesia 23 0 Lifeboat/raft container ship 30 Jan 97 Ahmet Akdeniz Gen cargo Foundered Black Sea 9 3 Lifeboat? 31 Jan 97 Calma Gen cargo Foundered Mediterranean 7 0 Fishing vessels 31 Jan 97 Toshin Gen cargo Foundered East China Sea 9 0 Cargo ship 2 Feb 97 Armenistis Gen cargo Fire Mediterranean 8 0 Coast Guard vessel Yes 4 Feb 97 Super Naias Ro-Ro pass Grounding Greece Coast Guard vessel 8 Feb 97 Leros Strength Bulk carrier Foundered North Sea 20 0 (Fatalities trapped inside) Yes 16 Feb 97 African Star Ref cargo Fire North Atlantic 19 0 Liferaft cargo ship 18 Feb 97 Albion 2 Bulk carrier Foundered North Atlantic Sea? 18 Feb 97 Finno Gen cargo Flooding North Sea 6 0 Helicopter Yes 18 Feb 97 Rajah Mas Gen cargo Grounded Malaysia 22 2 Sea? shore? 21 Feb 97 Galapagos Gen cargo Foundered Caribbean 0 Ship? 22 Feb 97 Avatapu Pass/cargo Fire Pacific 25 0 Liferafts tanker 28 Feb 97 Kangson Gen cargo Collision Yangtze River 41 3 Water? river craft? 4 Mar 97 Sapphire Gen cargo Fire Mediterranean 9 0 Cargo ship 5 Mar 97 Bintang Perkasa Pass/cargo Foundered Indonesia 20 8 Cargo ship 5 Mar 97 Sea Wind Ro-Ro pass Grounding Stockholm Pass ferry Yes? 6 Mar 97 Vikartindur Container Grounding Iceland 19 0 Helicopter Yes 8 Mar 97 Sea Boekanier Gen cargo Foundered Caribbean 0 Liferaft Coast Guard craft

6 II.4 9 Mar 97 Disarfell Container Foundered North Atlantic 12 2 Sea helicopter Yes 22 Mar 97 Golkap I Gen cargo Fire Black Sea 10 1 Ship? 25 Mar 97 Myo Hyang San Gen cargo Grounding Japan 22 0 Lifeboat Coast Guard craft 26 Mar 97 Cita Container Grounding E Canada 8 0 Shore lifeboat? * Includes fatalities trapped inside vessel

7 II.5 II.2 EVACUATION FREQUENCIES II.2.1 General Approach There are no published statistics on the frequencies of evacuations from ships. Available statistics on total loss rates may be used to give rough estimates. However, many passenger ships are evacuated from events that do not cause total loss. On the other hand, some vessels become total losses while they are laid-up or in repair, without needing evacuation. Hence, in general, the best estimate of evacuation frequency is obtained from surveys of actual evacuations. However, precautionary evacuations from minor events may not be reported widely, and so this approach may give an under-estimate of the true frequency. II.2.2 Total Losses of Passenger Ships The overall total loss rate for passenger vessels world-wide during was 1.4 x 10-3 per vessel year (LR 1995, 1996). This is based on 11 total losses (i.e. 3.7 per year) and a fleet size of approximately 2700 vessels (LR 1997). Of these, 7 out of 11 events (i.e. 64%) were actual total losses. This covers all passenger ships over 100 GRT, many of which are small. For passenger ships over 500 GRT (i.e. excluding small ferries) there were 9 losses (i.e. 3 per year) among approximately 700 vessels, i.e. a total loss rate of 4.3 x 10-3 per vessel year. For passenger ships over 5000 GRT (i.e. mainly cruise ships) there were 3 losses (i.e. 1 per year) among approximately 220 vessels, i.e. a total loss rate of 4.5 x 10-3 per vessel year. These suggest that the total loss frequency (and hence the evacuation frequency) for larger passenger ships may be greater than the average for all sizes of vessel. II.2.3 Evacuations of Cruise Ships Table II.1.1 lists 48 evacuations of cruise ships during , i.e. 1.4 per year. Of these, 40 involved ships over 5000 GRT, i.e. 1.1 per year. Assuming the fleet of 186 cruise ships over 5000 GRT (based on the Fairplay database for 1996) remained constant during this 35 year period, the evacuation frequency would be 6.0 x 10-3 per vessel year. However, this survey may be incomplete. During the period , Table II.1.1 lists 11 evacuations of cruise ships, of which 10 were over 5000 GRT, i.e. 3.3 per year. Based on a fleet of 186 passenger vessels over 5000 GRT, the evacuation frequency would be 1.8 x 10-2 per vessel year. Although this is based on a small sample, it is believed to be the most accurate estimate. It is approximately 4 times the total loss rate for these ships. II.2.4 Evacuations of Ro-Ro Passenger Ferries Appendix I.2 lists 37 evacuations of Ro-Ro passenger ferries during Coverage is probably incomplete, but is probably best for the period when it gives 32 evacuations, i.e. 1.9 per year. The world-wide fleet of passenger/ro-ro cargo vessels was 1900 in 1989 (based on Lloyd s World Fleet Statistics for 1989). Assuming this represents the average during this 17 year period, the evacuation frequency would be 1.0 x 10-3 per vessel year. However, this survey may be incomplete.

8 II.6 Coverage may have been more complete for NW Europe. During the period , Appendix I.2 lists 8 evacuations, i.e. 0.5 per year. The average exposure of vessels over 1000 GRT in this region has been estimated as 205 vessels, spending an average of 5100 hours at sea per year (DNV Technica Project C6185 MR App II.4.3). This gives an evacuation frequency of 2.3 x 10-3 per vessel year, or 4.5 x 10-7 per vessel hour at sea. II.2.5 Total Losses of Merchant Ships Table II.2.1 gives the total loss rates for different ship types, based on a combination of losses for 1994 and 1995 (LR, 1995, 1996), since the rates tend to fluctuate from year to year for some ship types. The highest values are for general cargo ships, and the lowest values are for tankers and passenger ships. Table II.2.1 Total Loss Rates for Ships over 100 GRT, World-Wide, SHIP TYPE TOTAL LOSSES (per 1000 ship-years) Liquefied gas carrier 1.5 Chemical tanker 1.0 Oil tanker 1.4 Bulk carrier 2.3 Oil/bulk/ore carrier 2.2 General cargo ship 4.0 Refrigerated cargo ship 3.7 Ro-Ro cargo ship 3.0 Passenger/Ro-Ro ship 1.2 Passenger ship 1.6 Fishing vessel 2.6 TOTAL 2.3 These can be used to indicate the evacuation rate for different ship types, although they may be under-estimates, as for passenger ships II.2.6 Evacuations of Merchant Ships Table II.1.2 lists 34 evacuations of merchant ships of all types except fishing vessels during Jan-Mar One of these was a tug, for which the operational pattern and the evacuation risks may be different from seagoing ships. The 33 evacuations of cargo, passenger, research, and offshore support vessels imply approximately 132 evacuations of seagoing merchant ships per year. Of the world fleet of 84,300 vessels over 100 GRT, 23,800 are fishing vessels, 8000 are tugs and 1100 are dredgers (LR, 1997), giving a corresponding fleet of 51,400 vessels. Hence the evacuation frequency would be approximately 2.6 x 10-3 per ship year. This is slightly higher than the total loss rate, as expected, and is considered to be a good estimate of evacuation frequency.

9 II.7 II.3 EVACUATION FATALITIES II.3.1 Causes of Evacuation II Cruise Ships Table II.3.1 categorises the events causing evacuation in the 48 cruise ship evacuations during in Table II.1.1. Fire and grounding are the main causes. The dominance of fires is rather less marked in recent years. For example, in the period , only 8 out of 18 evacuations were due to fires, i.e. 44%. Table II.3.1 Causes of Cruise Ship Evacuations, CAUSE NUMBER % Collision 3 6 Grounding Impact 2 4 Flooding 1 2 Fire TOTAL II All Merchant Ships Table II.3.2 categorises the events causing evacuation in the 34 merchant ship evacuations during Jan-Mar 1997 in Table II.1.2. The pattern is significantly different to cruise ships. Foundering and grounding are the main causes. Table II.3.2 Causes of Merchant Ship Evacuations, 1997 CAUSE NUMBER % Collision 3 9 Grounding 8 23 Foundering Fire 7 21 TOTAL Table II.3.3 categorises the events causing evacuation in the 169 evacuations of merchant ships during from Pyman & Lyon (1986). Although different hazard categories are used, foundering (including losses in storms, flooding and capsize) is still dominant. This older data shows more evacuations from fires than grounding, as with cruise ships.

10 II.8 Table II.3.3 Causes of Merchant Ship Evacuations, CAUSE NUMBER % Collision 15 9 Grounding 16 9 Impact Storm Flooding/capsize Structural failure 10 6 Cargo shift/list 3 2 Fire/explosion Dangerous goods leak Unknown 2 1 TOTAL II.3.2 Overall Fatality Rates II Cruise Ships The 48 cruise ship evacuations during caused a total of 848 fatalities, i.e. 18 per evacuation. However, this may be an over-estimate, since many non-fatal evacuations may have been omitted. Considering the period , there have been only 13 fatalities in 18 evacuations, i.e. an average of 0.7 per evacuation. This suggests that the current risks per evacuation may be as much as 25 times less than the historical average. However, this may be an under-estimate, since a single major accident could significantly increase it. On balance, the value of 18 per evacuation is preferred as a cautious estimate, with 0.7 per evacuation as an optimistic sensitivity test. Table II.1.1 indicates that approximately 33,000 people were evacuated in the 45 evacuations where numbers are known, i.e. an average of 730 people per evacuation. This suggests that the fatality rate in evacuation (i.e. the individual risk of death) for the period was approximately 18/730 = 2.5% of people evacuated. The current value may be up to 25 times lower. II All Merchant Ships The 34 evacuations of merchant ships during 1997 in Table II.1.2 caused a total of 82 fatalities, i.e. 2.4 per evacuation. There were at least 615 people evacuated (including those killed), giving a fatality rate of 82/615 = 13% of people evacuated. The 169 evacuations of merchant ships during from Pyman & Lyon (1986) caused a total of 1851 fatalities, i.e. 11 per evacuation. If passenger ships are excluded, this becomes 1235/161 = 7.7 per evacuation. Based on a smaller sample of 131 incidents with better information, Pyman & Lyon (1986) estimated a fatality rate of 23% of people evacuated. These relatively high figures may be due to including mainly larger vessels in the data, or may reflect the improvements in lifesaving equipment and rescue facilities since that time.

11 II.9 II.3.3 Effect of Accident Scenario on Fatality Rates II Cruise Ships Table II.3.4 shows the breakdown of fatalities by cause of evacuation in the 48 cruise ship evacuations during Clearly, fires and collisions are the types of accident that cause most fatalities. The fatalities in collisions are dominated by the largest single event, the 423 fatalities in the Admiral Nakhimov accident. The table also shows the breakdown of fatality rate as a fraction of the people evacuated. Clearly, collisions have the highest fatality rate, although the actual value is unreliable, since it is so dominated by the Admiral Nakhimov accident. Table II.3.4 Fatalities Rates in Cruise Ship Evacuations, CAUSE FATALITIES % OF ALL FATALITIES NUMBER EVACUATED FATALITY RATE (%) Fire Grounding Collision Impact Flooding TOTAL II Ro-Ro Passenger Ferries A previous analysis of fatality rates in accidents on Ro-Ro passenger ferries, based on 28 evacuations world-wide during (DNV Technica C6185 MR) indicated a substantial effect of the accident scenario on the fatality rate. For flooding accidents, the following fatality rates were estimated by averaging the fatality rates for evacuations in similar scenarios: Fast aground/beached 0% Slow sinking in shallow water 0.2% Rapid capsize in shallow water 23% Slow sinking in deep water with prompt/accurate Mayday 2% Slow sinking in deep water with delayed/inaccurate Mayday 80% Rapid capsize in deep water 72% For fires, the fatality rates were: Fire on sea around vessel 99% Fire on vehicle deck 36% Fire in accommodation 8% Fire in machinery spaces 0.7%

12 II.10 II Merchant Ships The 34 merchant ship evacuations during 1997 in Table II.1.2 include too few fatal events to form a distribution, but it is clear that most fatalities resulted from foundering. Table II.3.5 shows the breakdown of fatalities by event in the 169 evacuations of merchant ships during from Pyman & Lyon (1986). Fires and losses in storms are the main causes. Fires on passenger ships dominate the fatalities in fires. Excluding these, losses in storms are the main cause of fatalities. The structural failure category is dominated by bulk carrier losses. The storm category is dominated by general cargo ship losses. Table II.3.5 Fatalities in Merchant Ship Evacuations, CAUSE TOTAL FATALITIES EXC % FATALITIES PASSENGER SHIPS Collision Grounding Storm Flooding/capsize Structural failure Fire/explosion Unknown TOTAL Pyman & Lyon (1986) analysed 131 cases where merchant ships were evacuated by lifeboat or liferaft and found no particular effect of incident cause on the fatality rate, except for events that were correlated with rough weather. The effects of weather on the fatality rate were: Calm/moderate weather (Force 6 or less) 5% Rough weather (Force 7 or more) 35% Overall 13% For ships that sank rapidly (within a few minutes) the fatality rate was 86%. Fatalities due to fires/explosions themselves were excluded. The value for calm/moderate weather was strongly influenced by one passenger ship evacuation in a fire. Excluding this, the fatality rate for calm/moderate weather would be 2%. II.3.4 Causes of Death II Fires on Cruise Ships Information about the causes of death in fires on cruise ships during is as follows: Lakonia, 22 Dec fatalities, assumed to be split as follows 48 trapped on board, 20 killed launching lifeboats, 60 drowned awaiting rescue. Yarmouth Castle, 13 Nov fatalities, assumed to be split as follows 50 trapped on board, 39 drowned awaiting rescue.

13 II.11 Viking Princess, 8 Apr fatalities from heart attacks. Heleanna, 28 Aug fatalities, apparently drowned awaiting rescue. Patra, 23 Dec fatalities, apparently drowned awaiting rescue. Rasa Sayang, 2 Jun crew fatalities, presumably in fire-fighting operations. Chidabaram, 12 Feb fatalities, presumably due to fire/smoke on board. Priamurye, 18 May fatalities due to fire/smoke on board. Achille Lauro, 30 Nov fatalities of which 1 was killed boarding a lifeboat, 2 died from stress-induced illness, and 1 was missing, presumably trapped on board by smoke. Universe Explorer, 27 Jul crew fatalities due to smoke. These give a breakdown of cause of death as shown in Table II.3.6. Table II.3.6 Causes of Death in Fires on Cruise Ships CAUSE OF DEATH FATALITIES % Fire-fighting 5 1 Trapped on board by fire/smoke Stress-induced illness 4 1 Accident on lifeboat 21 5 Drowned awaiting rescue TOTAL II Collisions on Cruise Ships Information about the causes of death in collisions on cruise ships during is as follows: Admiral Nakhimov, 31 Aug fatalities, assumed to be split as follows 200 trapped on board, 223 drowned awaiting rescue. Jupiter, 21 Oct fatalities, apparently all during rescue from the water. Royal Pacific, 23 Aug fatalities, of which 6 were presumably trapped on board, and 3 drowned awaiting rescue. These give a breakdown of cause of death as shown in Table II.3.7.

14 II.12 Table II.3.7 Causes of Death in Collisions on Cruise Ships CAUSE OF DEATH FATALITIES % Killed by impact 0 0 Trapped on board by flooding/capsize Stress-induced illness 0 0 Accident on lifeboat 0 0 Drowned awaiting rescue TOTAL This is based on a small sample, and it might be desirable to supplement it with data from passenger/ro-ro ferry accidents. II Foundering of Merchant Ships Information about the causes of death in foundering of merchant ships during 1997 is as follows: Nakhodka, 2 Jan fatality (missing), assumed drowned awaiting rescue from capsized lifeboat. Onur K, 8 Jan fatalities of which 4 found dead in the water and 1 missing - assumed all drowned awaiting rescue from the sea. Ahmet Akdeniz, 9 Jan fatalities of which 1 found dead in the water and 2 missing - assumed all drowned awaiting rescue from the sea. Leros Strength, 8 Feb fatalities, assumed all trapped inside hull. Albion 2, 18 Feb fatalities, assumed all trapped inside hull. Bintang Perkasa, 5 Mar fatalities, assumed all drowned awaiting rescue from the sea. Disarfell, 9 Mar fatalities, apparently drowned awaiting rescue from the sea. These give a breakdown of cause of death as shown in Table II.3.8. Table II.3.8 Causes of Death in Foundering of Merchant Ships CAUSE OF DEATH FATALITIES % Trapped on board by flooding/capsize Drowned awaiting rescue TOTAL This is based on a small sample, and it would be desirable to supplement it with further data.

15 II.13 II.3.5 Time Available For Evacuation II Fires on Cruise Ships The time available in the event of a fire before evacuation becomes essential is estimated from historical events as follows. The start of the time is taken as the point where the fire starts or the fire alarm is sounded (if later). The last possible time for evacuation to start is taken as the time where evacuation actually started, or the time the fire extends along the whole length of the ship (if this is sooner). Information about the time available for evacuation in fires on cruise ships during is as follows: Lakonia, 22 Dec 63. Approx 10 minutes for fire to spread across the whole of the upper deck. However, this event is unrepresentative for modern ships due to the poor standard of fire protection on this ship. Yarmouth Castle, 13 Nov 65. Approx 40 minutes from fire detection to ordering abandon ship. Subsequently 5 hours till capsize. Although the time was long, this vessel is excluded for the same reason as the Lakonia. Heleanna, 28 Aug hours from fire detection to ordering abandon ship. Prinsendam, 4 Oct 80. Evacuation started 6 hours after fire started. Chidabaram, 12 Feb 85. Fire controlled after 16 hours. Passengers disembarked later. Priamurye, 18 May 88. Evacuation virtually complete within 15 minutes. Regal Empress, 19 Aug 94. Evacuation started 30 minutes after fire started. Achille Lauro, 30 Nov 94. Evacuation started 7½ hours after fire started. Albatros, 22 May 95. Fire controlled quickly. Passengers disembarked a day later. Celebration, 18 Jun 95. Fire controlled in 45 minutes. Passengers disembarked 2 days later. Regent Star, 22 Jul 95. Fire controlled in 2 hours. Passengers disembarked after 2½ hours. Romantica, 4 Oct 97. Evacuation started 50 minutes after fire started.. These give a breakdown of time available for evacuation as shown in Table II.3.9.

16 II.14 Table II.3.9 Time Available for Evacuation in Fires on Cruise Ships TIME FOR EVACUATION EVENTS % Short (< 1 hour) 3 30 Medium (1 3 hours) 2 20 Long (> 3 hours) 5 50 TOTAL II Collisions on Cruise Ships Information about the time available for evacuation in collisions on cruise ships during are as follows: Admiral Nakhimov, 31 Aug 86. Capsize within 8 minutes. Jupiter, 21 Oct 77. Immediate heel and evacuation. Sank after 40 minutes. Royal Pacific, 23 Aug 92. Sank within minutes. Clearly, all sinkings due to collision were within 1 hour. II Foundering of Merchant Ships The time available in the event of a foundering before evacuation becomes essential is estimated from historical events as follows. The start of the time is taken as the point where the Mayday is sent. The last possible time for evacuation to start is taken as the time where evacuation actually started or the ship capsized or sank. Information about the time available for evacuation in foundering of merchant ships during 1997 is as follows: Nakhodka, 2 Jan 97. A Mayday was sent when the ship was listing to 20 o. Assumed several hours before sinking. No reports until after ship had sank, 7 hours later. Onur K, 8 Jan 97. No reports of Mayday. Assumed no warning of sinking. Ahmet Akdeniz, 9 Jan 97. Several Mayday messages sent. Assumed several hours before sinking. Leros Strength, 8 Feb 97. Communication lost during Mayday. Liferafts not used. Assumed no time for evacuation. Albion 2, 18 Feb 97. Vanished without Mayday. Assumed no warning of sinking. Bintang Perkasa, 5 Mar 97. No reports of Mayday. Assumed no warning of sinking. Disarfell, 9 Mar 97. A Mayday was sent when the ship was listing to 90 o. Assumed no warning of sinking. These give a breakdown of time available for evacuation as shown in Table II.3.10.

17 II.15 Table II.3.10 Time Available for Evacuation in Foundering of Merchant Ships TIME FOR EVACUATION EVENTS % Short (< 1 hour) 5 71 Medium (1 3 hours) 2 29 Long (> 3 hours) 0 0 TOTAL This is based on a small sample, and it would be desirable to supplement it with further data. II.3.6 Rescue by Helicopter II Helicopter Availability Helicopters were used to assist the evacuation of cruise ships in the following cases: Heleanna, 28 Aug 71. No information about helicopter role. Possibly used to pick up injured already on shore. Prinsendam, 4 Oct 80. Helicopters used to transfer survivors from lifeboats to a tanker with a helideck. Turkmeniya, 11 Nov 86. No information about helicopter role. Possibly used to land fire party on ship. Maksim Gorkiy, 20 Jun 89. Helicopters used to transfer injured survivors from rescue vessels to shore. Oceanos, 4 Aug 91. Helicopters winched some passengers from ship and transferred them ashore. Royal Pacific, 23 Aug 92. Helicopters helped search for survivors. Sally Albatross, 19 Aug 94. Helicopters brought rescue staff. Achille Lauro, 30 Nov 94. Helicopters transferred supplies to rescue ships. Romantica, 4 Oct 97. Helicopters used to land fire party on ship. Thus, helicopters evacuated people direct from the ship in only one case (Oceanos). Helicopters arrived at the scene in a total of 14 cases, but in the 5 not mentioned above they arrived too late to be of any service. Nevertheless, the ability of helicopters to arrive, even at a late stage, may be used to indicate the probability of being within range of helicopters, given an evacuation. On average for cruise ship evacuations during , this was 14 out of 48, i.e. 29%. However, recent reporting is more comprehensive, and helicopter provision may be greater now than in the 1960s. Hence, a better indication of current helicopter availability is given by the evacuations during , when helicopters attended 9 out of 18, i.e. 50%.

18 II.16 For the merchant ship evacuations during 1997 in Table II.1.2, helicopters attended 9 out of 34, i.e. 26%. Assuming that in some other cases the incident may have been over before the helicopters had taken off, the total probability of being within range might be about 30%. The possible variation of this probability with distance offshore is investigated in Table II.3.11 for the cruise ship evacuations during , using the distances offshore from Table II There have been too few evacuations more than 200 nm from shore to show any variation in helicopter attendance probability. Such variation may be less than expected due to the possible presence of ship-borne military helicopters at long distances from the shore. Table II.3.11 Helicopter Attendance in Cruise Ship Evacuations DISTANCE OFFSHORE (nautical miles) EVENTS HELICOPTERS ATTENDED % ATTENDED In port 4 0 On shore (grounding) nm nm nm nm 0 0 >300 nm Unknown TOTAL II Helicopter Operability Helicopter operations may be limited by adverse weather conditions such as icing, high wind (preventing start-up) and low visibility. No such circumstances have been identified in the 14 cruise ship evacuations where helicopters attended. Meteorological data for the North Sea suggests that typical helicopter operability criteria are exceeded around 4% of the time (DNV Technica C3490 App VI.3.2). Operability of helicopters in precautionary evacuation has been estimated by Statoil to be 98.7% (Cullen, 1990). World-wide SAR experience indicates that less than 5% of missions are cancelled or shortened due to weather conditions. In an emergency, pilots could be expected to disregard operational limits (which are implicitly based on tolerable risks for daily operation), and fly to the limits of air-worthiness. Thus better operability than above should be achieved. A judgemental estimate of operability based on the above data is therefore 97%. The above reasoning assumes that adverse weather occurs independently of the need to evacuate. This is a reasonable assumption for most scenarios. However, for flooding in extreme weather, it is not valid. In these cases, it would be appropriate to assume that helicopter evacuation is not possible.

19 II.17 II.3.7 Rescue Ship Availability Other ships have provided the first available rescue facilities in most evacuations. Available information on their distance from the ship at the point when the Mayday was sent is given in Table II Where arrival times are known but not distances, these have been converted assuming a speed of 15 knots. Groundings are excluded because information on nearby ships is rarely available and may be misleading for these. Collisions are also excluded as these inevitably have another vessel close by. Table II.3.12 Nearby Ships in Cruise Ship Fires DATE SHIP EVENT DISTANCE OFFSHORE (nm) NEAREST SHIP TYPE ARRIVAL TIME (hours) DISTANCE AWAY (nm) 22 Dec 63 Lakonia Fire 160 Merchant 5½ Nov 65 Yarmouth Castle Fire 13 Merchant <6 Moderate? 8 Apr 66 Viking Princess Fire 60 Merchant Moderate? 21 May 68 Blenheim Fire Merchant Close? 19 Jul 70 Fulvia Fire 140 Merchant Moderate? 28 Aug 71 Heleanna Fire 12 Coastal Close? 3 May 73 Knossos Fire Merchant Close? 12 Sep 74 Cunard Ambassador Fire 40 Merchant Moderate? 18 Jan 76 Mecca Fire 17 Merchant Moderate? 23 Dec 76 Patra Fire 48 Merchant Moderate? 2 Jun 77 Rasa Sayang Fire 48 Merchant Close? 30 Mar 79 Angelina Lauro Fire 0 Harbour craft Close 4 Oct 80 Prinsendam Fire 120 Tanker >6 Remote 28 Aug 82 Mediterranean Star Fire Merchant Moderate? 9 Mar 84 Scandinavian Sea Fire Merchant Moderate? 12 Feb 85 Chidabaram Fire 300 Merchant Remote? 11 Nov 86 Turkmeniya Fire 60 Fishing May 88 Priamurye Fire 0 Harbour craft Close 20 Jun 89 Maksim Gorkiy Impact Coast Guard Moderate? 2 Jun 91 Pegasus Fire 0 Harbour craft Close 4 Aug 91 Oceanos Flooding 2 Merchant 6¾ Aug 92 Seabourn Spirit Fire 12 Tugs Close 19 Aug 94 Regal Empress Fire 5 Merchant 30 Nov 94 Achille Lauro Fire 100 Tanker May 95 Albatros Fire 60 Merchant 18 Jun 95 Celebration Fire 35 Merchant Close 22 Jul 95 Regent Star Fire Passenger 2½ Jul 96 Universe Explorer Fire Merchant Remote 4 Oct 97 Romantica Fire 60 Passenger 12 An analysis of the distances to the nearest ship at the time of the mayday from Table II.3.12 for cruise ship evacuations during is given in Table II.3.13.

20 II.18 Table II.3.13 Distance to Nearest Ship in Cruise Ship Fires DISTANCE FROM SHIP EVENTS % (nautical miles) Close (0 49 nm) Moderate (50 99 nm) Remote ( 100 nm) 5 19 TOTAL Other passenger ships were the nearest in 2 out of these 26 events, i.e. 8%. Both of these were in fact other cruise ships. For the merchant ship evacuations during 1997 in Table II.1.2, other ships were used for the rescue in 25 events, and 2 of these were passenger ships. This is the same percentage as for cruise ship evacuations, but in this case both were passenger ferries. II.4 REFERENCES LR (1997) : World Fleet Statistics 1996, Lloyd s Register, London. LR (1996) : World Casualty Statistics 1995, Lloyd s Register, London. LR (1995) : World Casualty Statistics 1994, Lloyd s Register, London. Pyman, M.A.F. & Lyon, P.R. (1986) : Casualty Rates in Abandoning Ships at Sea, Trans RINA, London.