Accident Investigation Board of Finland Annual Report 2008

Transcription

1

2 Onnettomuustutkintakeskus Centralen för undersökning av olyckor Accident Investigation Board Osoite / Address: Sörnäisten rantatie 33 C Adress: Sörnäs strandväg 33 C FIN HELSINKI HELSINGFORS Puhelin / Telefon: (09) Telephone: Fax: (09) Fax: Sähköposti: E-post: Internet: [email protected] tai [email protected] [email protected] eller förnamn.slä[email protected] [email protected] or first name.last [email protected] Henkilöstö / Personal / Personnel: Johtaja / Direktör / Director Hallintopäällikkö / Förvaltningsdirektör / Administrative director Osastosihteeri / Avdelningssekreterare / Assistant Toimistosihteeri / Byråsekreterare / Assistant Ilmailuonnettomuudet / Flygolyckor / Aviation accidents Johtava tutkija / Ledande utredare / Chief Air Accident Investigator Erikoistutkija / Utredare / Air Accident Investigator Raideliikenneonnettomuudet / Spårtrafikolyckor / Rail accidents Johtava tutkija / Ledande utredare / Chief Rail Accident Investigator Erikoistutkija / Utredare / Rail Accident Investigator Tuomo Karppinen Pirjo Valkama-Joutsen Sini Järvi Leena Leskelä Hannu Melaranta Tii-Maria Siitonen Esko Värttiö Reijo Mynttinen (Leave of absence) Acting Erkki Hainari ( ) Vesiliikenneonnettomuudet / Sjöfartsolyckor / Marine accidents Johtava tutkija / Ledande utredare / Chief Marine Accident Investigator Martti Heikkilä Erikoistutkija / Utredare / Marine Accident investigator Risto Repo Muut onnettomuudet / Övriga olyckor / Other Accidents Johtava tutkija / Ledande utredare / Chief Accident Investigator Kai Valonen

3 MISSION OF ACCIDENT INVESTIGATION BOARD The Accident Investigation Board of Finland was founded in 1996 within the Ministry of Justice. The tasks of the Accident Investigation Board are specified in the relevant act and decree which also include overall directions on the categories of the accidents to be investigated and the methods of investigation to be implemented. By its investigation activities, the Accident Investigation Board intends to enhance overall safety and prevent accidents. As a result of an accident investigation, an investigation report is produced that contains safety recommendations for the competent authorities and other parties concerned. In fact the safety recommendations translate the investigators views on the means of prevention of similar or corresponding accidents in the future. Moreover the Accident Investigation Board follows up the implementation of the recommendations issued. The investigation work conducted by the Board exclusively focuses on the improvement of safety with no stances taken as for questions of culpability, responsibility or liability for damages. It is the mission of the Investigation Board to investigate all major accidents, serious incidents and aviation, rail, and marine accidents and incidents. The investigation of aviation accidents is based on the relevant European Council Directive (94/56/EY) and the Convention on International Civil Aviation (Treaty Series of the Statute Book 11/49), and the investigation of rail accidents is based on the EU Railway Safety Directive (2004/49/EY). As for maritime accidents, their investigation is based on the European Council Directive 1999/35/EY and 2002/59/EY. Accident investigation focuses on the course of events of the accident, its causes and consequences as well as on the relevant rescue measures. Particular attention is paid to whether the safety requirements have been adequately fulfilled in the planning, design, manufacture, construction and use of the equipment and structures involved in the accident. It is also investigated whether the supervision and inspection has been carried out in an appropriate manner. Eventually any detected shortcomings in safety rules and regulations may call for investigation, as well. In addition to the direct causes of an accident, the accident investigation aims at revealing any contributory factors and underlying circumstances that may be found in the organization, the directions, the code of practice or the work methods. In the decision-making on the commencement of an accident investigation, the level of seriousness of the incident is considered as well as its probability of recurrence. An incident or accident or hazardous situation, with only minor consequences may also require investigation in case it puts several persons at risk and an investigation is assessed as producing important information in view of the improvement of the general safety and the prevention of further accidents. Generally speaking, the Accident Investigation Board does not investigate an incident or accident caused intentionally or by an offence. 1

4 The Accident Investigation Board is also responsible for the maintenance of a contingency to rapidly commence an investigation, the training of new accident investigators, the producing of general instructions on the carrying out of the investigation work and on the drawing up of the investigation reports, and the participation in international cooperation. Finally, the Accident Investigation Board is responsible for the printing and distribution of the investigation reports and their publishing on its web pages, Terms A-investigation B-investigation C-investigation D-investigation S-investigation Investigation categories Major accident Accident or serious incident Incident, damage or minor accident Other incident Safety study L R M Y Accident/incident categories Aviation accidents and incidents Rail accidents and incidents Marine accidents and incidents Other accidents and incidents Investigation identifier Each investigation is designated by an identifier that consists of four parts, such as A1/1998R. The first part refers to the investigation category (A, B, C, D or S). The second part is a sequence number referring to the order of the accident within its accident category in the year in question. The third part refers to the year of the accident. The fourth part indicates the accident category (L, R, M or Y). E.g. A1/1998R refers to the first major railway accident investigation in

5 INDEX MISSION OF ACCIDENT INVESTIGATION BOARD...1 REVIEW INVESTIGATION COMMISSIONS INVESTIGATIONS AVIATION...10 Investigations commenced in Investigations completed in Safety recommendations...24 RAIL...27 Investigations commenced in Investigations completed in Safety recommendations...42 MARINE...44 Investigations commenced in Investigations completed in Safety recommendations...59 OTHER ACCIDENTS...61 Investigations commenced in Investigations completed in Safety recommendations...65 PERFORMANCE...68 Impact...68 Cost-effectiveness...69 Productivity...71 Workplace wellbeing...72 Service quality and performance...73 FINANCES...76 INVESTIGATIONS COMPLETED IN

6 REVIEW 2008 In 2008, investigations were initiated on as many accidents and incidents as in About half of these were classified as so-called minor category D investigations. Clearly fewer investigation reports were filed than in 2007, since the number of completed D-investigations totalled ten fewer than in On the other hand, the number of completed B-level and C-level investigations, which refer to more serious accidents or incidents for which a full investigation report is prepared, was five higher than in The helicopter crash that occurred in the outskirts of Tallinn in 2005 is not included in the statistics because the investigation was led by the Estonian authorities. The investigation of this accident was concluded in August The Accident Investigation Board participated in the investigation in accordance with international agreements and allocated the same amount of resources to it as for a major accident. Moreover, the investigation of the Jokela and Kauhajoki school shootings, in which the Accident Investigation Board is also participating, have involved similar resource outlays. Because these shootings were not accidents but premeditated acts, new legislation was passed with respect to their investigation. The investigation of fatal level-crossing accidents was continued in accordance with the EU Railway Safety Directive, which came into force in Investigations were begun on five levelcrossing accidents in 2008, which was two fewer than in the previous year. Excluding levelcrossing accidents, there was a greater number of serious accidents in 2008 than in 2007, and the total number of serious accidents increased. Hence, the declining trend that began in 2005 came to a halt. For the first time, a tram accident was investigated as a serious incident. Efforts have been made to reduce the time spent on investigations, for example by improving monitoring and investigation guidelines. With regard to serious accidents, i.e. B-investigations, the targeted average investigation period of 18 months was achieved, which, however, must still be improved upon in order to attain the one-year maximum set in EU directives. Efforts to shorten investigation periods have been visible, for example in the fact that at the end of 2008 the number of investigations ongoing for more than one year had been reduced to one-half of the number in 2007, and to a third of the number in The average investigation period varies considerably between investigation branches. In aviation, the one-year target has been reached. The Accident Investigation Board contributes to general safety by issuing briefings on its investigation results, and especially by issuing safety recommendations in its investigation reports, for the benefit of the competent authorities and other key parties. In addition, the Accident Investigation Board monitors the adoption of these recommendations. In 2008, for the first time we achieved our target of keeping track of the implementation status of at least 80 percent of the recommendations issued since 2000, totalling 740 at the end of Of these recommendations, 41 percent had been implemented. Implementation takes its own time, as indicated, for example, by the fact that 74 percent of the 145 rail recommendations issued from had been implemented, whereas the corresponding number for was only 31 percent. The decision not to fulfil the recommendation has been made in about 15 percent of the cases. 4

7 The Accident Investigation Board s expenses grew by 1.5 percent from Salary and rents, which accounted for about 85 percent of the total, were up by nearly 5 percent. Strict budgeting allowed the reduction of other expenses by roughly 14 percent. Cost control focused on training and travel in particular. The amount spent on investigations grew somewhat, which was partly due to the larger than normal costs incurred in the search for parts from an aircraft crash that occurred towards the end of the year. Overall, less investigation funding was used than in As in previous years, our working time monitoring indicated that the greatest single use of staff resources, i.e. more than a third, originated in accident investigation proper. The amount of time spent on support functions has somewhat decreased, accounting for about 20 percent or 2.5 working years of the total 12.8 working years of full-time staff. Management and administration also participated in the investigations proper. In addition to our in-house staff, external experts also engage in accident investigations, primarily as members of the investigation commissions. In 2008, a total of 88 external experts participated in the investigation process. This number has remained more or less unchanged over the years, as has the total number of working years contributed by external experts, amounting to just less than eight working years in As a result, the Accident Investigation Board spent around 20 working years on accident investigation in In each investigative branch, orientation training was provided for new external investigators in 2008, and an additional two-day training session was also organised for aviation investigators. A seminar on the impact of fatigue on the occurrence of accidents was organised for investigators and stakeholders in cooperation with the Finnish Motor Insurers Centre, which is in charge of road and terrain accident investigations in Finland. Our in-house investigators participated in training in Finland and abroad. On EU level, special attention has been paid to the education and competence requirements of accident investigators. The Accident Investigation Board participated in the training of Croatian marine accident investigators in connection with a Twinning project led by the Finnish Maritime Administration. Furthermore, the year was a busy one with regard to international activities. The Accident Investigation Board participated in an International Civil Aviation Organisation conference for the development of international accident investigation norms, which is organised about once every eight years. Towards the end of the year, renewal of the European Council s accident and incident investigation directive was initiated, after a joint understanding on the corresponding directive for maritime traffic was reached by the European Council and Parliament in December. In may, the International Maritime Organisation approved new regulations concerning maritime accident investigations, which will enter into force at the beginning of I wish to express my sincere thanks to all accident investigators for their excellent input in Tuomo Karppinen Director 5

8 INVESTIGATION COMMISSIONS 2008 The Accident Investigation Commissions had the following memberships in Aviation Ismo Aaltonen, Ari Anttila, Markus Bergman, Päivikki Eskelinen-Rönkä, Hannu Halonen, Timo Heikkilä, Juhani Hipeli, Ari Huhtala, Erkki Kantola, Vesa Kokkonen, Ilpo Kopra, Jouko Koskimies, Timo Kostiainen, Martti Lantela, Timo Lindholm, Esko Lähteenmäki, Hannu Melaranta, Jari Multanen, Toni Mäkelä, Asko Nokelainen, Pekka Orava, Markku Roschier, Juha Salo, Tii-Maria Siitonen, Hans Tefke, Sanna Winberg, Tapani Vänttinen Rail Pekka Aho, Jari Auvinen, Aki Grönblom, Raimo Harjunen, Kati Hernetkoski, Henrik Hieta, Matti Joki, Veli-Jussi Kangasmaa, Matti Katajala, Timo Kivelä, Jukka Koponen, Pekka Kuikka, Sirkku Laapotti, Pekka Laine, Ari Murtola, Reijo Mynttinen, Jaakko Niskala, Lasse Nurmi, Petri Pelkonen, Harri Pöysti, Hannu Räisänen, Reijo Sarantila, Veikko Stolt, Jarmo Tuomi, Marko Törmänen, Esko Värttiö, Martti Väänänen, Kari Ylönen Marine Ville Grönvall, Harri Halme, Markku Haranne, Martti Heikkilä, Kaarlo Heikkinen, Olavi Huuska, Jukka Häkämies, Mikko Kallas, Kari Larjo, Jaakko Lehtosalo, Karl Loveson, Hannu Martikainen, Seppo Männikkö, Risto Repo, Pertti Siivonen, Toimi Sivuranta, Matti Sorsa, Juha Sjölund, Pirjo Valkama-Joutsen, Micael Vuorio Other Pekka Aho, Hannu Alén, Tor Erik Ekberg, Markku Haikonen, Heikki Harri, Esko Kaukonen, Timo Kivelä, Kurt Kokko, Jukka Koponen, Marja Kurenniemi, Tapio Leino, Anne Lounamaa, Jaakko Niskala, Pasi Paloluoma, Anssi Parviainen, Hannu Rantanen, Taneli Rasmus, Erkki Reinikka, Seppo Ronkainen, Kai Sjöholm, Esa Virtanen, Jorma Westerholm, Tarja Wiikinkoski 6

9 INVESTIGATIONS 2008 The following tables summarise investigations commenced, completed and in progress in 2008 in the various accident/incident and investigation categories. The investigation categories are defined as follows: A-investigation B-investigation C-investigation D-investigation S-investigation Major accident Accident or serious incident Incident, damage or minor accident Other incident Safety study Summary of investigations commenced in 2008 Accident/incident Investigation category category A B C D S TOT Aviation Rail Marine Other TOTAL Summary of investigations completed in 2008 Accident/incident Investigation category category A B C D S TOT Aviation Rail Marine Other TOTAL Summary of incomplete cases 31 December 2008 Accident/incident Investigation category category A B C D S TOT Aviation Rail Marine Other TOTAL

10 In 2008, eight more investigations were begun than were completed. At the end of the year, the number of incomplete investigations was clearly lower than the number of completed ones. With regard to category B and C investigations, for which a full investigation report is compiled, there were also clearly fewer incomplete ones than completed ones. More B-level and C-level investigations were completed than minor D investigations, for which a brief investigation report of a few pages is prepared. One safety study was finalised, and four were under progress at the end of the year. We achieved our objective of having fewer investigations in progress than could be completed within the duration of a single year. The tables below compare the number of investigations started, completed and in progress in 2008, with corresponding numbers for previous years. Summary of investigations commenced in Year Investigation category A B C D S TOT TOTAL Summary of investigations completed in Year Investigation category A B C D S TOT TOTAL Summary of incomplete cases at the end of years Year Investigation category A B C D S TOT

11 As many investigations were initiated in 2008 as in More category B and C investigations were initiated than in The number of completed category B and C investigations was also higher than in that year. Nevertheless, the total number of investigations completed was lower than in the previous year because the number of completed D-investigations fell from 35 in 2007 to 25 in Starting an investigation ties up resources and this might have had an impact on the number of completed investigations. The completion of a substantial number of category B and C investigations also tied up resources, because a full investigation report is compiled for investigations in these categories. At the end of 2008, the number of incomplete investigations was more or less on a par with the number at the end of 2007, and clearly lower than at the end of Since the beginning of 2007, fatal level-crossing accidents have been investigated in accordance with the EU Railway Safety Directive, which came into force in Of the category B investigations initiated, seven were level-crossing accidents in 2007, and five in If levelcrossing accidents are excluded from the number of category B investigations, for the reason that they have not been investigated on earlier occasions, the number of category B accidents fell evenly from 2004 to The number grew somewhat in 2008 and was on the same level as in 2003 and Investigations completed in S D C B A

12 AVIATION In 2008, three B-level and eight C-level investigations were initiated. An investigation commission was appointed for these investigations. Nine D-level investigations were initiated. One case of material damage occurred in air transport. In ten investigated incidents, a transport aircraft was involved, and a general aviation aircraft in one such incident. One fatal accident and three cases of material damage leading to an investigation occurred in general aviation. Three cases of material damage leading to an investigation occurred in the ultra-light and experimental aircraft category. In 2008, two category B and seven category C investigations were brought to a conclusion. Six category D investigations were completed. In August, the investigation of a helicopter crash in Tallinn Bay in 2005 was completed. The investigation was led by the Estonian investigation authorities, with the Accident Investigation Board participating in accordance with international agreements. Four aviation investigations included investigation authorities from other countries. The category D investigation format has proven to be an effective way of creating safety benefits following the investigation of minor incidents. Efforts have been made since the start of 2008 to enhance the distribution and publication of category D investigation results, with the aim of having most of the investigation results published on the Accident Investigation Board's website early in The Accident Investigation Board participated in an International Civil Aviation Organisation meeting on accident investigation held in Montreal in October. Such a meeting is held about once every eight years, one of its aims being to establish new and improved international norms for aviation accident investigations. Furthermore, in August a meeting of Nordic aviation accident investigators was held in Norway. Three investigators from Finland participated in the meeting. In addition, our investigators participated in several international meetings and seminars over the course of the year, and Cooperation with the European Aviation Safety Agency continued. The process of renewing the European Council Directive on the fundamental principles governing the investigation of civil aviation accidents and incidents was initiated towards the end of For this reason, the European Civil Aviation Conference s Group of Experts in Accident Investigation convened several times in The Accident Investigation Board participated in these meetings as well as participating actively in the Council of European Aviation Safety Investigation Agencies, established in Our investigators gave lectures in several training sessions held for the competent authorities. Lectures were also given during many other aviation events. 10

13 Investigations commenced in 2008 In 2008 the Accident Investigation Board commenced altogether 11 aviation accident and incident investigations. Identifier Date of occurrance Title of the investigation B1/2008L Serious incident at Lappeenranta airfield B2/2008L Collision risk in the vicinity of Pori airfield B3/2008L Aircraft accident in Taipalsaari C1/2008L Serious incident at Lappeenranta airfield C2/2008L Serious incident at Helsinki-Vantaa airport C3/2008L Breach of altitude limits in the Helsinki approach zone C4/2008L Incident at Helsinki-Vantaa airport C5/2008L Incident at Oulun airfield approach zone C6/2008L Incident at Helsinki-Vantaa station C7/2008L Incident in the vicinity of Oulu airfield C8/2008L Serious incident at Kemi airport In addition, in 2008, a total of 9 D-investigations were commenced. Investigations completed in 2008 In 2008 the Accident Investigation Board completed altogether 9 aviation accident and incident investigations. Identifier Date of occurrance Title of the investigation B1/2007L Glider accident at Haapavesi aerodrome B2/2008L An air proximity incident in Pori Terminal Artea C1/2007L Serious incident at Seinäjoki Airport C3/2007L Emergency water landing off Helsinki C5/2007L Incidents at Helsinki-Malmi airport caused by ultralight aircraft C6/2007L Helicopter accident at Pelkosenniemi C8/2007L Helicopter accident at Kangasniemi C9/2007L An unauthorized approach to an engaged runway at Pori aerodrome C3/2008L Violation of separation minimums at Helsinki-Vantaa terminal control area In addition, 6 D-investigations were completed in

14 B1/2007L Glider accident at Haapavesi aerodrome on 9 June 2007 A glider accident occurred at Haapavesi aerodrome on Saturday, 9 June 2007 at 14:14 when a PIK-20 glider, registration OH-465, collided with the ground during an automobile launch takeoff. The glider was completely destroyed and the pilot was killed instantly. During the takeoff run the right wing of the glider made contact with the ground, resulting in the glider yawing approximately 30 degrees to the right. According to procedure, this situation would have called for an aborted takeoff with the pilot pulling the towline release handle. However, the pilot continued with the takeoff. The glider s attitude was extremely abnormal at liftoff; it was in a strong sideslip to the left and rolled to the right. Immediately after liftoff the glider went into a steep climb and began to roll to the left. This was followed by a barrel roll-type manoeuvre, resulting in a collision with the ground 137 metres from the takeoff point. The glider came down inverted and at a steep angle. The towline remained hooked until impact. Investigation revealed that the automobile launch takeoff was executed according to routine procedure and in a pre-briefed manner. The pilot was a relatively experienced glider pilot, even though he had accrued most of his experience from He had lately flown totally different types of gliders. After a hiatus of nine years he had flown twice on the accident aircraft type in He had hardly any experience with automobile launch takeoffs on PIK-20s. Due to the long hiatus in flying, he had little recent experience on the type. The primary causes of the accident were the pilot s decision to continue with the takeoff even after the wing made contact with the ground, the direction of the aircraft changed and the fact that he most probably executed the takeoff at an inadequate airspeed. After the steep initial climb the glider was out of control. Contributing factors included the pilot s limited recent experience on the accident aircraft type, his previous experience as well as the wind conditions. The fact that the wing made contact with the ground and his decision to continue with the takeoff were the results of his insufficient recent experience after a long hiatus. He was not adequately aware of how critical the PIK-20 s aileron control was at takeoff, nor of the differences between the PIK-20 and the other glider types he had recently flown. Wind conditions were the most probable reason for inducing a roll on the takeoff run. The investigation commission made no recommendations. However, the commission wants to emphasize glider pilots personal responsibility in estimating their actual proficiency on different 12

15 types of gliders. The commission especially highlights the pilot s capability for correct and rapid decision-making in takeoff situations as well as the importance of sufficient recent experience on the type at hand. B2/2008L An air proximity incident in Pori Terminal Artea on 10 June 2008 On 10 June 2008 two aircraft belonging to the Finnish Aviation Academy were involved in an air proximity incident close to Pori aerodrome. OH-BBN (BE36 Bonanza) was on an instrument rating check flight, as per the syllabus of the Academy. OH-BSB (BE30 King Air) was on an instrument training flight. At 10:54, during their approach phases in simulated IMC conditions, both aircraft maintained 1700 ft (approximately 500 m) above the initial approach fix (IAF) PITUM and flew towards each other. OH-BSB executed a TCAS manoeuvre. The incident did not result in any damage. Air traffic control cleared OH-BBN from Pori instrument training area B to initial approach fix FR at 2700 ft for an ILS Y approach to runway 30. A moment later the air traffic controller recleared it to initial approach fix PITUM at 1700 ft for an ILS Z approach to runway 30. Air traffic control cleared OH-BSB, departing runway 12, to FR at 1700 ft. A moment later the ATC recleared OH-BSB to PITUM at 1700 ft. Some forty seconds after takeoff OH-BSB reached the clearance altitude 1700 ft. After it passed PITUM the air traffic controller cleared it for an ILS Z approach to runway 30. OH-BBN flew towards PITUM, heading around 200 degrees, and reached 1700 ft just before arriving at PITUM. Both aircraft were now flying at the same altitude. When OH-BSB reached the final approach track, the flight paths of the aircraft crossed at an approximate 90 degree angle. At 10:54 OHBBN crossed the flight path of OH-BSB approximately 2.3 NM ahead of it. After passing PITUM OH- BBN turned towards the heading of 120 degrees, causing the aircraft to now be heading towards each other. Some thirty seconds later the TCAS system on OH-BSB generated a Traffic Advisory (TA), directly followed by a Resolution Advisory (RA). OH-BSB immediately initiated a TCAS descent. 13

16 The pilot in command of OH-BSB reported having seen an approaching aircraft approximately half a mile away. The pilot reported the TCAS manoeuvre to air traffic control, notified that they were clear of conflict and climbed back to 1700 ft. According to their statement neither pilot on OH-BBN spotted the other aircraft at any phase of the incident. Immediately after the incident the air traffic controller realized that he had inadvertently cleared OH-BBN to 1700 ft on the radio, even though he had marked 2700 ft as the clearance altitude on the flight progress strip. Both aircraft followed their clearances and none of the pilots noticed that they had been cleared to the same fix at the same altitude. The erroneous altitude clearance went unnoticed by the air traffic controller in charge as well as by the other air traffic controller who was working the same shift. Both aircraft continued their flights as planned after the incident. The pilots in command of both aircraft as well as the air traffic controller reported the incident as required. The cause of the incident was the air traffic controller inadvertently issuing an erroneous clearance altitude on the radio, even though he had marked it correctly on the flight progress strip. Contributing factors include: The aircrews did not adequately monitor radio communication neither did anybody notice the mistake The locator FR was informed by notam to be unserviceable, but this was not clearly indicated on the flight progress strip board and the air traffic controller, out of habit, used it as a clearance limit. The locator was, however, operating and transmitting its identification code and a bearing indication. The aircrews did not correctly report all positions and altitudes essential in Procedure Control. The investigation commission made no safety recommendations. The investigation commission made no recommendations. 14

17 C1/2007L Serious incident at Seinäjoki Airport on 1 January 2007 An incident occurred in Seinäjoki aerodrome airspace on 1 January 2007 from 17:50 18:30 local time. The incident involved a Finncomm Airlines ATR-42 airliner, call sign WBA205S. The aircraft made several unsuccessful instrument approach attempts to Seinäjoki. Onboard warning systems generated several warnings during approaches. In the end, the aircraft was flown to Vaasa, the alternate aerodrome. The aircraft departed Helsinki at 17:15 and flew to Seinäjoki at FL 200. The first officer was the Pilot Flying. Meteorological conditions at Seinäjoki were: cloud base varying between 500 and 1500 ft; light snow; surface wind 120 degrees KT, gusting to 18 KT; wind at 3000 ft 160 degrees 30 KT. At 17:48 the aircraft passed Seinäjoki outer marker (PSJ) for runway 32 and initiated an NDB approach via locator O for runway 14. On the final approach the aircraft descended too low, resulting in an EGPWS warning. The first officer flew a missed approach profile towards PSJ. Again, the aircraft descended too low during the approach and they received another EGPWS warning. After the second missed approach procedure, during the turn to the final approach course, their airspeed decreased, the autopilot disengaged and the stick pusher activated. The flight crew assumed an electrical malfunction and climbed to 7000 ft to work out the cause of malfunction. When no such malfunction was detected, they returned to the outer marker and began to descend in the racetrack pattern. For the third time, the aircraft went too low, resulting in yet another EGPWS warning. The first officer then flew a missed approach procedure. Soon after this the flight crew noticed that the first officer s altimeter was incorrectly set to hpa. The captain s altimeter was set to Seinäjoki QNH 978 hpa. After that they did an ILS approach to runway 32 and a circling manoeuvre to runway 14. While in the circling manoeuvre, they received yet another warning of an erroneous flight configuration. During the turn to the final approach course the aircraft banked approximately 50 degrees resulting in a bank angle warning. During the missed approach procedure the turn continued and finally they flew at a heading of 050 degrees instead 15

18 of heading 130 degrees towards PSJ. At 18:29 the captain requested and received a clearance to Vaasa, where they landed at 18:50. Investigation revealed a mistake concerning the selection of the flight crew for this flight. The captain had only logged some 50 hours on the ATR and the first officer had approximately 80 hours on the aircraft. Irrespective of the captain s 3500 total flying hours, the crew was inexperienced according to company s policy with this aircraft type. The approach preparations to Seinäjoki proceeded too slowly causing the crew to rush. The approach checklist was inadequately completed and they forgot to change the altimeter settings from standard air pressure hpa to Seinäjoki QNH 978 hpa. The result was that they flew the entire time 950 feet too low. Therefore, during the NDB approach to runway 14 the aircraft descended to 345 ft (105 m) above Ground Level (AGL), which activated the unsafe terrain clearance warning. The same factors generated the second warning. This time they descended to 425 ft (130 m) AGL. As they continued with the approach, they lost so much airspeed and the angle of attack increased so much that the stall warning activated, the autopilot disengaged and the stick pusher activated. The antiicing system was on; therefore, the stall warning and the stick pusher activation threshold took place at a higher airspeed and lower angle of attack compared to normal. This time they descended to 1250 ft (385 m) at minimum. Unsafe terrain clearances generated the EGPWS warnings. Too low airspeed and the rapidly increased angle of attack activated the stall warning and the stick pusher. The first officer s fatigue and loss of concentration caused the excessive bank angle during the circling approach. The captain should have taken over at this stage. The Cockpit Voice Recorder (CVR) data analysis revealed poor Crew Resource Management (CRM), several missing checks, inadequate system knowledge and incomplete following through of approach procedures. In order to establish the underlying factors of the occurrence, the investigation looked into the airline s quality system, training system, organization and management as well as their security culture. Shortcomings were discovered in all of the above. The company had not sufficiently prepared for the challenges of the then ongoing business expansion. Nevertheless, the airline took corrective action during the time of the investigation. The commission issued three recommendations. The incorrect altimeter setting, the flight crew s limited experience on the aircraft type and insufficient situational awareness were considered the causal factors for the chain of events which triggered the incident. Shortcomings in the airline s training system, organization, quality system and security culture were regarded as contributing factors. 16

19 C3/2007L Emergency water landing off Helsinki on 27 May 2007 An emergency landing occurred on Sunday, 27 May 2007 at around 11:40 (Finnish time) off the shoreline of Helsinki. The gearbox clutch of a single-engine Diamond DA40D four-seat aircraft, registration OH-FDA, failed and the aircraft was subsequently forced to make an emergency landing in water. The OH-FDA was owned by the Aviation Club of Helsinki University of Technology. The pilot intended to fly two passengers to Tallinn airport and return to Helsinki Malmi aerodrome in the afternoon. The Full Authority Digital Engine Control (FADEC) unit executed a pre-takeoff engine runup test. The test passed, with the engine providing full power. The takeoff and climb to 1000 ft were uneventful. The pilot was cleared to 4000 ft in Helsinki Terminal Control Area (TMA). When he began to climb he selected full power, at which time engine power fluctuated between % for 5 to 10 seconds. He then remembered that the pilot of the previous flight on the OH-FDA had told him that he, too, had noticed similar power oscillations. However, the power stabilized at 100% and all engine instrument indications were normal. Nevertheless, the pilot felt that the engine did not provide as much power as it had earlier. When he set the autopilot to a 500 ft/min climb, the engine suddenly overrevved (exceeded maximum RPM). The pilot then reduced power and tried to establish what the matter was. Meanwhile, he reported engine troubles to the air traffic control (ATC) and said that he would turn back to Helsinki-Malmi. At that time he was over the sea next to Harmaja Island, approximately 15 km from Helsinki-Malmi aerodrome. Even though the pilot checked the emergency checklist regarding engine trouble, he soon realized that the engine would either idle or overspeed. The pilot deemed that the malfunction involved either the fuel supply or the FADEC. He let the engine run at a high rpm so as to make it back to land. When they were at about 700 ft the pilot realized that they could not make it to the shore. Instead, they would have to make an emergency landing into the sea. The pilot reported the impending emergency landing and his estimated landing spot to the ATC and informed the passengers of the same. He landed the aircraft in the strait between Valkosaari Island and Katajanokka. The aircraft came to a halt approximately 15 metres from Valkosaari shore. 17

20 The pilot and the passengers climbed out onto the wings. Momentarily, a boat arrived next to the right wing, into which they boarded. Soon after, a police patrol boat also arrived at the scene, transporting the pilot and the passengers to Helsinki South Harbour where ambulances were already waiting. The Border Guard fastened the aircraft to Valkosaari shore, where it was encircled with oil spill booms. The aircraft was transported to Katajanokka Quay, where the wings were detached. Then the ircraft was taken to Helsinki-Malmi aerodrome for test and research. The gearbox was disconnected from the engine, at which stage it was noticed that the friction surfaces on the clutch plate were worn. The FADEC was sent to the German accident investigation authority, under whose supervision the engine manufacturer downloaded vital On-Board Diagnostic (OBD) data from the FADEC. The cause of the emergency landing was clutch slippage and clutch plate surface wearing out. Clutch slippage was caused by engine oil which entered through a leak in the crankshaft s lip seal. Fissures were detected in the seal and engine manufacturer assumes they are due to high internal crankcase pressure The investigation commission issued three recommendations. First: The investigation commission recommends that the engine manufacturer design a system indicating changes in the rotational ratio between the crankshaft and propeller, i.e. clutch slippage. Clutch slippage indication should be recorded in the FADEC s On-Board Diagnostics system. Second: The investigation commission recommends that the engine manufacturer redesign the clutch breather orifice so that oil leaks internal to the clutch housing could be reliably distinguished from other, external, oil leaks. Third: The investigation commission recommends that the aircraft manufacturer expedite the time of compliance for mandatory service bulletin to improve the crankcase breathing system. 18

21 C5/2007L Incidents at Helsinki-Malmi airport caused by ultralight aircraft on 12 August 2007 At Helsinki-Malmi airport occured several incidents during unintentional taxiing and take-off on 12th August, 2007 at caused by Skyranger V.Max -type ultralight amphibian aircraft. The pilot s purpose was to taxi the aircraft from apron 2 to apron 1 across runway 18/36 to its own stand. Immediately after engine start-up it went straight to full power and the aircraft started moving. The pilot concentrated taxiing the aircraft and could not stop it. The aircraft taxeed at high speed across active runway 18/36 and continued on apron 1 taxiway. After turning left the pilot decided to make a take off. After take off pilot took radio contact to Malmi tower and returned from flight. The floats were slightly damaged during taxiing. The pilot s flying experience with this aircraft type was little. In addition aircraft cockpit ergonomy was significantly different from other aircraft types the pilot had flown. The pilot did not use checklists and did not use safety belts before engine start-up. The pilot turned the radio ON during the flight. Incidents were caused by throttle lever mistakenly left to full power during engine start-up. Other causal factors were lack of checklists, little experience on this aircraft type, pilot s attitude to taxiing compared to actual flight and wheel brakes insufficient to keep aircraft not moving. The Investigation Commission made two safety recommendations. The Finnish Aeronautical Association is recommended to produce general instructions and checklist model for ultralight aircraft taxiing and run-up. In addition the Commission recommends that during basic and continuous training for ultralight flight instructors the correct use of checklists is emphasized. The Finnish Civil Aviation Authority is recommended to change type rating training part of aviation regulation PEL M2-70, the training according to different cockpit layouts and technical characteristics. 19

22 C6/2007L Helicopter accident at Pelkosenniemi on 16 September 2007 An accident occurred on Sunday, 16 September 2007 at 14:00 local time when a helicopter landed near Pelkosenniemi village centre. The aircraft involved was a Robinson R22 Beta heliopter, registration OH-HMH. The pilot and the passenger sustained minor injuries. During the final phase of its approach the helicopter drifted through a bush at the edge of the landing area. The aft end of the left skid caught the verge of a ditch, causing the front end of the right skid to make contact with the uneven field. This resulted in a dynamic rollover. The helicopter pivoted around the front right corner and rolled onto its right side. The investigators believe that the pilot flew an excessively tight landing pattern. Therefore, the helicopter did not stabilize on the final approach. Rather, the helicopter continued on a right sideslip and banked to the right while descending low with relatively high speed. Because of his limited flying experience, the pilot did not have the skills to recover from the situation. The pilot did not manage to level the helicopter. The speed did not decrease enough and this resulted in an unsuccessful landing. He could have aborted the approach and flown a wider landing pattern. At no stage during the approach did the helicopter descend into its own rotor wash in a vortex ring state. It is the opinion of the investigation commission that the accident was not caused by a technical malfunction or defect. The investigation commission made no recommendations. The draft investigation report was promulgated for comment to the Finnish Civil Aviation Authority, the owners of the accident helicopter as well as to the pilot and the passenger. 20

23 C8/2007L Helicopter accident at Kangasniemi on 8 November 2007 An accident occurred at 10:05 on Thursday, 8 November 2007 at Koittila in Kangasniemi municipality. A Hughes 369D helicopter, registration OH-HLA, collided with the ground. The helicopter in question was maintaining a transmission line by sawing tree branches off with an external saw. The pilot, the sole occupant of the helicopter, was slightly injured, however, the helicopter was completely destroyed. The helicopter departed for the sawing flight approximately one hour and thirty minutes prior to the accident. The helicopter was fitted with a 369 kg heavy external saw that extended 25 m below it. During the flight it was snowing lightly. At times the snowfall turned into heavy sleet, with snow accumulating a few centimetres on the ground. After flying for about an hour the helicopter landed for hot refuelling. After the helicopter became airborne again it continued the sawing operation in light snowfall. While the helicopter was sawing branches in an almost completely stationary hover the engine failed. Because of the external load and the steering input of the pilot there was not a controlled descent, but rather, pivoting around the saw boom, the helicopter yawed into the direction it had come from, collided with trees and finally crashed into the ground. In spite of his injuries and the damage to the helicopter the pilot made it out of the cabin on his own accord. It is known that a jet engine such as the one used in this helicopter may flame out if it ingests enough snow or ice. The engine manufacturer has advised users of this propensity by publishing bulletins explaining the phenomenon. The company had fitted the helicopter with a particle separator at the air inlet and issued restrictions concerning snowfall in order to ensure the safe conduct of helicopter operations. When it was sleeting the heaviest the meteorological conditions did not fulfil the company s weather requirements. While the helicopter was aloft snow and sleet probably built up in the air intake, particularly, during heavy snowfall and sleet. Without warning, the engine probably ingested snow from the air intake, resulting in a flame-out. The helicopter was equipped with an engine warning system that also activates an automatic reignition system, thereby providing an automatic engine restart capability. During the accident flight the system was switched off, which is a violation of the helicopter Flight Manual. Moreover, the investigation 21

24 revealed that there were shortcomings in the company s operations manuals and that the personnel do not always comply with valid regulations. The accident was caused by flying in inclement weather, snowfall, which resulted in an engine flame-out that was probably caused by the engine ingesting wet snow accumulated on the engine air intake surface. The fact that the engine warning system was turned off, effectively eliminating the automatic reignition system, was a contributing factor. The investigation commission recommended that the company update the weather restrictions in its operation manuals so as to correspond with regulations. In addition, the commission recommended that the company take appropriate actions to bring the fleet s Flight Manuals up to date as well as alter the practices of its staff so as to comply with valid manuals and regulations. The commission also recommended increased supervision with regard to staff procedures. C9/2007L An unauthorized approach to an engaged runway at Pori aerodrome on 7 November 2007 An incident occurred on 7 November 2007 at 20:23 UTC (Finnish time -2h) at Pori aerodrome, involving a Pegasus Airlines Boeing 737 charter flight from Copenhagen and an aerodrome maintenance vehicle. As the aircraft was approaching Pori, a maintenance vehicle was assessing the Runway Visual Range (RVR) by counting the number of visible runway lights on the active runway. The air traffic controller reported the RVR to the pilots. Because the RVR was below the landing minimum he cleared them to a holding pattern to wait for RVR to improve. However, the pilots were of the impression that they were permitted to continue the approach to the Decision Altitude (DA) of the ILS approach. The misunderstanding between the air traffic controller and the pilots resulted in a collision hazard between the vehicle and the aircraft. The incident occurred because the pilots flew the approach without the required ATC clearance. Unsatisfactory and unclear radio communications between the air traffic controller and the flight crew were contributing factors. Another contributing factor was that the pilots violated regulations by continuing with the approach even when the reported RVR was below minimum. 22

25 The investigation commission did not make any recommendations because present rules and regulations, if properly observed, suffice in preventing these kinds of incidents from taking place. C3/2008L Violation of separation minimums at Helsinki Vantaa terminal control area on 29 February A significant incident occured at Helsinki Vantaa terminal control area on 29 February 2008 at UTC when a Finnair MD11 on a scheduled flight from Helsinki to Delhi and a Finnair A340 on a scheduled flight from Tokio to Helsinki passed each other on crossing tracks at approximately 2300 meters height so that the required separation minimums were violated. The aircraft received traffic alert (TA) from their collision avoidance systems but not collision avoidance command (RA). Both aircraft were in instrument meteorological conditions and the pilots did not get visual contact with each others aircraft. Finnair 74 approached Helsinki from the northeast according to air traffic control clearance and was descending to flight level 80 (2450 meters). At the same time Finnair 21 departed Helsinki following according to air traffic control clearance a standard departure route, which leads southeast. The air traffic control applied vertical separation due to crossing tracks. Finnair 21 was cleared to climb to flight level 70 (2150 meters) 1000 feet below Finnair 74. Finnair 21 climbed approximately 400 feet (120 meters) above the cleared flight level. At the time of the incident the QNH at Helsinki Vantaa airport was 995 HPA. The pressure differential to the standard pressure, 1013 HPA, was 18 HPA. Since one HPA corresponds to 27 feet height, the total height difference was 486 feet (144 meters). Finnair 21 flew above transition altitude using QNH setting in altimeters which meant that the actual flight level was 486 feet above the indicated altitude displayed by the altimeters. Finnair 21 started to descend before it reached 7000 feet on QNH so altitude difference to Finnair 74 was approximately 600 feet. The vertical separation was violated by approximately 400 feet and horizontal radar separation by approximately 1,2 nautical miles. In this case the vertical separation should have been 1000 feet (300 meters) and the horizontal radar separation should have been 3 nautical miles (5,5 kilometers). The violation of separation minimums did not induce collision risk. 23

26 Due to the deficient cockpit crew work management, the altimeters were not adjusted to the standard pressure setting. The wrong pressure setting went unobserved and the aircraft climbed too high violating separation minimums with the other aircraft on a crossing track. The investigation commission gave two safety recommendations: 1. In the Finnair Airbus and Embraer fleets there is an attention system that safeguards the observation of transition altitude which the other fleets of the company do not have. The commission recommends that Finnair Oyj considers the installation of similar attention system to other aircraft fleets of the company and to the aircraft types that will be obtained in the future. 2. The present Helsinki radar system includes a conflict alert feature (Short Term Conflict Alert, STCA) but it is not in operative use due to nuisance alerts. The radar system updating is planned. One part of the updating is a new STCA feature which serves as a safety net also in parallel runway operations. The commission recommends that with the updating Finavia obtains a STCA feature which is suitable in the Helsinki TMA airspace. Safety recommendations A total of 15 recommendations were given by aviation accident and incident investigations completed in 2008, 4 of which were issued on the basis of a category D investigation. Some of the recommendations were directed at several recipients. The 2008 recommendations were directed at the following recipients: Recipient No Finavia 3 Finnish Civil Aviation Authority 3 Airlines / flight schools 5 Organizations 1 Manufacturers 3 The implementation status of recommendations issued after 2000 is monitored by means of a special monitoring programme. Such monitoring indicated that, of the 269 recommendations issued during , 48% had been implemented by the end of A decision not to carry out the recommendations was made in about 15% of cases. 24

27 Aviation accident investigations Accidents investigated TOT Major accidents (A-investigations) Other accidents (B- and C-investigations) TOTAL Investigations as per type of accident/incident TOT Accidents Damages Losses of separation Other incidents Investigations as per aviation category TOT Commercial aviation Other Investigations as per aircraft category TOT Commercial aircraft General aviation aircraft/helicopter Glider and motorglider Experimental and ultralight aircraft FAF aircraft Personal injuries TOT Pilot Fatally injured Passanger Total Pilot Seriously injured Passanger Total Pilot Slightly injured Passanger Other person Total TOTAL

28 D-investigations in D-investigations as per type of accident/incident Damages 7 7 Losses of separation 3 - Other 9 2 D-investigations as per aviation category Commercial aviation 13 3 Other 6 6 D-investigations as per aircraft category Commercial airplane 12 2 General aviation aircraft/helicopter 4 3 Experimental and ultralight aircraft 2 3 Other

29 RAIL In 2008, the Accident Investigation Board initiated six B-level and six C-level rail accident investigations. One safety study was also initiated. Of the B-level investigations, five involved fatal level crossing accidents and one involved a tram collision investigated as a serious incident. Of the C-level investigations, one involved a truck and shunting unit collision involving serious injury, three were derailings involving the transport of dangerous goods during shunting work, one involved the collision of a locomotive and a maintenance machine and one involved a train traffic incident. The safety study addresses safety deviation in train number automation. The European Railway Agency (ERA) has been notified of investigations initiated in accordance with the EU Railway Safety Directive. In 2008, four B-level investigations and eleven C-level investigations were completed. The B-level investigations were fatal level crossing accidents. The investigation reports on these cases have been sent to the ERA in accordance with the Railway Safety Directive. Our rail accident investigators participated in a meeting of Nordic Rail Accident Investigators (NRAI) in Norway in May. The chief rail accident investigator participated in meetings of the National Accident Investigation Bodies (NIB) of the European Railway Agency (ERA) and those of two of its sub-committees. The NIB meetings were held in February, June and October in Lille, France. A sub-committee meeting on the compilation of NIB annual reports was held in November in Lille. Meetings of a working group addressing training and competence requirements were held in June and November in Lille. A Human Factors training day relating to the working group was held in December, also in Lille. These working groups will reconvene in In January, the Head of the ERA s Safety Unit visited the Accident Investigation Board to learn more about our activities. Furthermore, during the year our rail accident investigators participated in the Finnish Rail Administration s 2008 seminar and in a major accident simulation exercise. Our rail accident investigators gave lectures to their partners on accident investigation and provided workplace instruction and orientation for new part-time investigators. The first orientation course in line with our new training system was held in May. 27

30 Investigations commenced in 2008 Identifier Date of Title of the investigation Basis occurrance (occurrence type, location) for inv. S1/2008R Safety study on safety deviation relating to train number automation in the Lahti area iii B1/2008R Fatal level crossing accident in Laukaa i B2/2008R Collision of two trams on Mäkelänkatu in Helsinki ii B3/2008R Fatal level crossing accident in Viinijärvi i B4/2008R Fatal level crossing accident in Kiuruvesi i B5/2008R Fatal level crossing accident in Suonenjoki i B6/2008R Fatal level crossing accident in Iisalmi i C1/2008R Derailment of five shunting unit wagons in the Heikkilä railway in Turku iii C2/2008R Derailment of a wagon carrying phosphoric acid in Ykspihlaja, Kokkola iii C3/2008R Collision of a shunting unit and a forklift on the Syväsatama Port track in Joensuu iii C4/2008R Derailment of a tank wagon during shunting work in Ykspihlaja, Kokkola, iii C5/2008R Collision of a locomotive and a track tamping machine at the Jyväskylä railyard iii C6/2008R Train traffic incident in Kerava iii Basis for investigation: i = According to the Railway Safety Directive, ii = On national legal basis (covering possible areas excluded in Article 2, 2 of the Safety Directive), iii = Voluntary - other criteria (National rules/regulations not referred to the Safety Directive) 28

31 Investigations completed in 2008 Identifier Date of Title of the investigation Basis Completed occurrance (occurrence type, location) for inv. (date) B4/2007R Fatal level crossing accident in Kiuruvesi i B5/2007R Fatal level crossing accident in Röykkä, Nurmijärvi i B6/2007R Fatal level crossing accident in Perälä, Kempele i B7/2007R Fatal level crossing accident in Lahti i C9/2005R Freight car derailing and member of train crew injured at Tuupovaara iii C3/2006R Derailment of five freight wagons between Tupovaara and Heinävaara, Finland iii C2/2007R Derailment of a wagon in Ylivieska iii C3/2007R Incident at the Tampere railway yard iii C4/2007R Derailment of eight freight train wagons between Saarijärvi and Äänekoski iii C5/2007R Derailment of a freight train locomotive in Talviainen iii C6/2007R Tank wagon loaded with nitric acid tipped over in Siilinjärvi iii C1/2008R Derailment of five shunting unit wagons in the Heikkilä railway yard in Turku iii C2/2008R Derailment of a wagon carrying phosphoric acid in Ykspihlaja, Kokkola iii C3/2008R Collision of a shunting unit and a forklift truck on the Syväsatama port track in Joensuu iii C4/2008R Derailment of a tank wagon during shunting work in Ykspihlaja, Kokkola iii Basis for investigation: i = According to the Railway Safety Directive, ii = On national legal basis (covering possible areas excluded in Article 2 2 of the Safety Directive), iii = Voluntary - other criteria (National rules/regulations not referred to the Safety Directive) 29

32 B4/2007R Fatal level crossing accident in Kiuruvesi, Finland, on 6 May 2007 A fatal level crossing accident took place in Kiuruvesi, at the unprotected level crossing of Pohja. This accident occurred when a car travelling along the Pohja private road drove without stopping under a rail bus running from Ylivieska to Iisalmi. There were two passengers in the car; the driver perished and the front seat passenger was seriously injured. The total cost of the accident amounted to 50,000. The accident was caused by the car driver s failure to notice the train. The level crossing at which the accident occurred does not fulfil the determined level crossing regulations in regard to visibility and the crossing angle. Observation was hampered by the characteristics of the level crossing and, possibly, the driver s health. Familiarity with the level crossing probably diminished the driver s attentiveness. The start of the rescue operation was delayed due to the fact that the emergency alarm call made by traffic control was routed to the wrong emergency response centre with regard to the place of the accident. There were problems with locating the place of the accident, resulting in two of the rescue units driving to the wrong level crossing. In order to avoid similar accidents and alleviate their effects, the investigation commission recommends that the Pohja level crossing be eliminated, since the nearest overpass is located at a distance of only 300 metres. Moreover, the commission recommends that the Finnish Rail Administration analyse similar level crossings that do not entail a great level of risk but that can be removed at a minor cost. In addition, the commission calls for an improvement in the emergency call capacities of traffic control, and reiterates its previous recommendation regarding emergency alarm calls performed at the accident site. 30

33 B5/2007R Fatal level crossing accident in Röykkä, Nurmijärvi, Finland on 13 August 2007 On Monday 13 August 2007 at 3.15 p.m., a level crossing accident occurred in Röykkä, Nurmijärvi, in which a passenger car collided with a freight train en route from Kirkniemi to Riihimäki, resulting in the death of the car s passenger and serious injuries to the car driver. The car was driving in the direction of the track along Korventie road before making a turn on the Leppälammentie road, which crosses the track. The driver stopped before the level crossing, at the STOP sign, and started crossing the track, but the engine stopped and the car became stuck in the middle of the crossing. The driver of the car and the passenger on the driver s right tried to restart the engine. The driver then saw the train approaching from the right and heard the warning whistle. The freight train had departed from Kirkniemi at 2.22 p.m. The train driver noticed that a passenger car had stopped on the track on a straight section after a bend on the approach to the Korpi crossing. The train driver gave a warning whistle 250 metres ahead of the crossing. Emergency braking began about 150 metres before the crossing. The train collided with the right side of the car, at the point where the front wheel was situated. This collision threw the car onto the track embankment about 18 metres from the point of collision. The driver and the passenger were still inside the car. The train stopped 276 metres from the point of collision. The front of the train engine suffered some damage and the car was entirely wrecked. The direct cause of the accident was that the passenger car died in the middle of the level crossing without restarting and the train was unable to stop in time despite applying the emergency brake. The driver of the car possibly made a mistake that led to the engine extinction on the crossing. A technical failure in the car prevented the engine from restarting. The weakening of the passenger s judgement due to intoxication also had an impact on the driver s decision not to leave the car as the train approached. Furthermore, being alarmed by the approaching train affected the decision not to leave the car. In order to prevent similar accidents, the investigation commission recommends that the Korpi level crossing be equipped with half-barriers. 31

34 The investigation commission s observations support earlier recommendations relating to problems rescue operations have in locating the sites of accidents: Level crossings should be equipped with signboards displaying at least the name of the level crossing and its location in the coordinates and relevant track-km. The signboard should be clearly visible in both running directions of the road. [B1/00R/S143] The instructions for the drawing up of an emergency notice should be developed to ensure that whenever urgent aid is needed from the rescue service, also the general emergency number is called from the incident site, in addition to the notifying of the traffic control unit. [B1/05R/S211] The compliance of the localization data used by the railway with the data system of the Emergency Response Centre Agencies shall be ensured, e.g. by installing the track-kilometre data in the data system of the Emergency Response Centre Agencies. [B1/05R/S212] B6/2007R Fatal level crossing accident in Perälä, Kempele, Finland, on 6 October 2007 On Saturday 6 October 2007 at a.m., a car and a Pendolino train en route from Oulu to Helsinki collided on the Perälä level crossing in Kempele, resulting in the death of the car driver. The train staff and passengers were not injured. The car driver s destination was a dog fair in Kempele. The driver left the highway at the Oulunsalo ramp, headed in the direction of Oulunsalo, and then turned left towards Kempele town centre. At the Shell service station the driver turned left onto Sohjanantie. According to an eyewitness, the driver drove along Sohjanantie at a low speed, but did not stop at the level crossing s STOP sign. Pendolino S52 had departed from Oulu towards Helsinki at a.m. As the train approached the level crossing at Perälä, the engine driver noticed a car driving along Sohjanantie on the right. After it became apparent that the car was continuing towards the crossing without stopping, the engine driver started emergency braking. The track speed limit at the crossing was 140 km/h, which was also the speed of the train immediately before the collision. 32

35 The train collided with the left side of the car. The car was caught under the front of the train and was pushed ahead by the train until the train stopped 592 metres from the crossing. The train was not derailed. The car was entirelly wrecked and the train s front structure incurred damage. Traffic along the section of the line came to a halt from a.m p.m. Train traffic between Oulu and Liminka was handled by busses while the track was closed. The accident caused damage totalling 132,000. The direct cause of the accident was that the car driver drived onto the level crossing without stopping. It is likely that the driver failed to make any observation of the train approaching from the left. This was probably because the driver had taken a wrong turn. Furthermore, the driver s attention was probably focused on driving in an unfamiliar environment, looking out for a place to turn around and pedestrians who were walking their dogs along the road. Sohjanantie was not equipped with appropriate warning signs either before or at the crossing. The vegetation between the road and the track also impaired the visibility of the track and the train running parallel to the road. In order to prevent similar accidents, the Accident Investigation Board of Finland recommends the removal of the Perälä level crossing or its replacement with an interchange. Furthermore, as a derailment prevention measure, the Board recommends that the lower front structure of the train be re-designed in such a way that cars cannot be wedged underneath. In addition, the Board proposes the following actions before the implementation of the recommendations: Sohjanantie should be equipped with the appropriate warning signs, the level crossing should be equipped with portals, and undue vegetation between the road and the track should be regularly removed. The Board favours proposals for improved accident localisation with equipping locomotive with a GPS equipment and automatic locating of a mobile phone, as submitted in a statement by the Emergency Response Centre of North Ostrobothnia and Kainuu. B7/2007R Fatal level crossing accident in Lahti, Finland on 21 November 2007 On 21 October 2007 at p.m., a fatal level crossing accident occurred on an unprotected level crossing along Heikinpellontie road in Lahti. The accident occurred when a car on Heikinpellontie road drove without stopping in front of a locomotive en route from Lahti to Heinola. The driver, who was the sole person in the car, died instantly. 33

36 The accident occurred because the driver of the car did not see the train. The level crossing in question meets regulations concerning visibility and crossing angles, but does not meet those concerning wait platforms. It is possible that the driver was not sufficiently vigilant due to familiarity with the crossing and the impression that train traffic was infrequent there. Rescue operations were somewhat delayed because traffic control called the wrong emergency response centre, and not the emergency response centre that was responsible for the accident site. Further delays were caused by difficulties in locating the accident site. To prevent similar accidents in the future, the investigation commission recommends that renovation investments planned for the track be speeded up with respect to level crossing security. Several level crossings along the Lahti Heinola track do not have sufficient visibility, and therefore the investigation commission recommends that the track speed limit be lowered at level crossings to a level that ensures the appropriate safety level. In order to limit the negative consequences of similar accidents, the investigation commission restates earlier recommendations concerning the ability to make a telephone call directly from the accident site, and improving the compatibility of information used by emergency response centres and traffic control to locate the accident site. C9/2005R Freight car derailing and member of train crew injured at Tuupovaara on 31 December 2005 On Saturday 31 December 2005 at 9.14 a.m., a shunting accident occurred in the Tuupovaara railway yard, in which a group of empty wagons for carrying wood products, being pushed by an engine, collided with a derailer, causing the derailment of the first wagon in the direction of travel. The shunting foreman, who was standing on the wagon s left end step, was seriously injured after falling between the tracks and being hit by the left end step of the next wagon as he extricated himself from the moving wagons. The step dragged him for several metres before he was able to break free. The accident occurred because the derailer had not been removed and the shunting foreman did not notice this in time. The non-removal of the derailer, in turn, was possible because the key could be removed from the derailer's safety lock even though the derailer had not been removed from the rail. 34

37 To prevent the occurrence of similar accidents, the Accident Investigation Board recommends that the safety lock's operation be altered in such a way that the safety key cannot be removed before the derailer has been removed from the rail. In addition, the Board recommends that greater attention be paid to safety measures for shunting workers during the ploughing of snow in rail yards. C3/2006R Derailment of five freight wagons between Tupovaara and Heinävaara, Finland, on 13 July 2006 On Thursday, 13 July 2006, at 4:41 pm, a freight train en route from Tuupovaara to Joensuu was derailed about 10 km from Tuupovaara in the direction of Joensuu. The train was carrying timber. There were no casualties. A 100-metre stretch of track was damaged, and five wagons were partially damaged and later scrapped. The accident was caused by the formation of a heat curve on the track. The rail buckled because it had a weak structure made of light rails, wooden railway sleepers and gravel ballast. The heat curve was released below the train and derailed the last five wagons of the train. Repair work at the scene also had a negative impact on track stability. The Accident Investigation Board of Finland is not issuing any recommendations as a result of this accident because the track has little traffic and the risks of a similar incident are very low. 35

38 C2/2007R Derailment of a wagon in Ylivieska on 21 March 2007 On Wednesday, 21 March 2007, at 10:33 am, one wagon of the freight train en route from Oulu to Ylivieska was derailed at the northern turnout of the Ylivieska station, as the train was switching from main track to side track. The top leaf of the spring pack of the derailed wagon had broken and fallen before the derailment. In addition, a wheel bearing was broken, a wheel flat occurred, and brake triangle support screws had fallen. The wagon wheels were damaged while running on ballast, and the bogie and under frame were damaged as the bogie collided with the under frame. Also damaged were the coupling and buffer equipment of the derailed wagon and the wagons connected to it. The derailed wagon broke the electric-motor switch drive of two turnouts. Rail traffic northbound from Ylivieska was blocked for three and a half hours, and eastbound traffic for 24 hours. The total cost of the accident was 24,000 euros. The derailment occurred because the unloaded front wheel of the front-most wheelset did not steer at the turnout, because of the missing spring pack, and therefore the bogie did not turn but tried to continue straight ahead. The spring pack had fallen because the uppermost leaf holding the pack together had broken. This was probably caused by the wheel flat and leaf fatigue. On account of its structure, a spring pack can come apart after the main leaf breaks, and the vibration caused by a wheel flat contributes to this. The Accident Investigation Board of Finland recommends that, to prevent the occurrence of similar accidents, greater care be exercised in statutory freight train inspections, and that any flaws observed be acted upon more quickly than is currently the case. It should be ensured that the inspectors are qualified to identify damage such as that described above. 36

39 C3/2007R Hazardous situation in train traffic in Tampere on 27 May 2007 On Sunday, 27 May 2007, at 6 pm, an incident occurred at the Tampere station, in which a shunting unit passed, without authorisation, a shunting signal that was in the stop position at the south end of the Tampere passenger railway yard. Simultaneously, a passenger train was arriving in Tampere, for which a route had been provided to the station. The train driver noticed that a shunting signal in front of the train had switched to stop and was able to stop the train ahead of shunting unit wagons that were on the track. The cause of the incident was that the shunting foreman did not notice that the shunting signal was in the stop position. Locomotives engines standing on the adjacent track and their tail lights had blurred the shunting foreman s vision as the shunting unit approached the point. The Accident Investigation Board is not issuing new safety recommendations as a result of the incident but stresses that training and guidelines should emphasise the importance of providing relevant additional information during shunting work. C4/2007R Derailment of eight freight train wagons between Saarijärvi and Äänekoski, Finland, on 3 July 2007 Eight wagons of a freight train carrying wood were derailed on 3 July 2007 at 4.01 p.m. Four of the wagons incurred heavy damage, and four minor damage. About 170 metres of track were damaged. 37

40 The accident was caused by the poor condition of the track and the train s excess speed, considering the condition of the track. The first of the freight wagons (the 16th wagon), carrying pinewood, was derailed. Researches show that pinewood is heavier than spruce. As the wagon approached what was possibly the weakest point of the track, the outer rail of the track, which was on a curve, was dislocated. The distance between the rails grew to such an extent that the wagon s wheels fell between the rails. Track support work had been completed at the point of the derailing. This track work, and the small repositioning and sideways movement of the rails that this involved, reduced the stability of the track. To prevent the occurrence of similar accidents, the Accident Investigation Board recommends that segments of the track that are in poor condition be investigated, and that a speed limit of 20 km/h be set for segments that are in poor condition for trains with an axis weight of tons, until the necessary repairs have been completed. In addition, the Accident Investigation Board repeats recommendation S181, which it issued after the occurrence of a similar accident in Huutokoski on 31 May 2002: The track should immediately be repaired and the defective old sleepers be replaced by new ones. Replacement of spike fastening by screw fastening, replacement of the rails by heavier ones, and replacement of the gravel in the railway bed by ballast should be discussed and considered. C5/2007R Derailment of a freight train locomotive in Talviainen, Finland, on 15 July 2007 On Sunday 15 July 2007 at 6.11 p.m., one of the two locomotives of a freight train was derailed after passing a curved turnout in Talviainen station. The derailed locomotive incurred some damage. The derailment occurred because the track was bent out of shape and therefore hindered passage. Contributing to this was the fact that rail construction in the depot had involved deficiencies in planning and implementation. At no point during the construction project had the special features of the rail s unusual geometry been taken into account. The geometry had been called into question during planning, but the matter had not been addressed when new plans were formulated. During planning, no observations had been made that there was insufficient space to even out the cant in the turnout. In order to prevent similar occurrences in the future, the Accident Investigation Board recommends that planning guidelines be formulated for curved turnouts and that demanding construction projects include the measurement of rail geometries with loads before commissioning, in order to ensure that limit values are met. 38

41 C6/2007R Tank wagon loaded with nitric acid tipped over in Siilinjärvi, Finland, on 4 August 2007 At Kemira GrowHow Oyj railway yard an accident occurred on Saturday at 6.24 am, where a tank wagon loaded with nitric acid collided with a derailer, causing the wagon to derail and tip over. The following wagon also derailed. It stayed upright. The total cost of the accident was less than euros. The reason for the accident was that the derailer was not removed before shunting of the wagons and that the derailer that had been left on was not noticed in time. The shunting foreman gave order to shunt without securing the route first. To avoid similar accidents, the Accident Investigation Board of Finland recommends that the right operation of derailers should always be secured so that false operation and leaving the derailer on rail could not be possible. On railway yards, where dangerous goods are handled, should always have dependence between the derailer and the turnout that leads to the rail. C1/2008R Derailment of five shunting unit wagons in the Heikkilä railway yard in Turku, Finland, on 8 February 2008 On 8 February 2008 at 9.53 a.m., three Russian tank wagons and two Russian covered wagons were derailed during shunting in the Heikkilä railway yard in Turku. The track was damaged for about 70 metres. The direct cause of the occurrence was that the track, which was in poor condition and fastened by rail spikes, gave way under the heavy tank wagons. In addition, the dry, non-greased bogie pivots of the wagons placed additional pressure on the track curve. In order to prevent similar occurrences, the Accident Investigation Board of Finland recommends that a 20 km/h speed limit be set for wagons transporting dangerous goods on spike-fastened secondary tracks. In addition, track and railway yard condition monitoring and rail fastening work should place special emphasis on routes and tracks used for the transport of dangerous goods. 39

42 C2/2008R Derailment of a wagon carrying phosphoric acid in Ykspihlaja, Kokkola, Finland, on 1 March 2008 On Saturday 1 March 2008 at 6.12 a.m., a shunting work incident occurred on an industrial track in Ykspihlaja, Kokkola. A group of wagons carrying phosphoric acid drove into a derailer. The bogie of the leading wagon was derailed. The incident occurred because the shunting unit did not stop in sufficient time before the derailer. This was due to an error of judgement made by the shunting foreman and a lack of communication between the shunting foreman and engine driver. Furthermore, snow and ice had accumulated on the brakes of the wagons, thereby weakening the power of the breaks. In order to prevent the occurrence of similar incidents, the Accident Investigation Board of Finland recommends that engine drivers be clearly informed of any blocks (e.g. derailers) along the track during shunting operations. In addition, the Board emphasises that unnecessary risks should be avoided when approaching derailers or other blocks. C3/2008R Collision of a shunting unit and a forklift truck on the Syväsatama port track in Joensuu, Finland, on 30 April 2008 On Wednesday 30 April 2008 at 7.04 a.m., a shunting unit collided with a heavy forklift truck on Joensuu s Syväsatama port track 183. The shunting foreman was seriously injured. One of the freight wagons incurred minor damage and the forklift truck was badly damaged. The accident occurred because the forklift driver did not observe the approaching shunting unit before turning or when turning to cross the track. The driver noticed the shunting unit only upon the collision. In order to fulfil his lookout duty, the shunting foreman was standing on the buffer step on the right side of the first wagon in the direction of travel, which contributed to the injury. 40

43 He was unable to stand on the corner step because of a high loading platform on the right side of the track. In order to prevent the occurrence of similar accidents, the Accident Investigation Board of Finland recommends that storage containers should be placed further away from the track so that they do not impede visibility. No other recommendations have been issued because actions have been taken to improve port safety with the installation of warning lights indicating that a shunting unit is moving along the tracks. In addition, the loading platform next to the track should be dismantled if it is no longer in use. C4/2008R Derailment of a tank wagon during shunting work in Ykspihlaja, Kokkola, Finland, on 15 May 2008 On Thursday 15 May 2008 at 5.28 p.m., one tank wagon carrying a sulphuric acid consignment was derailed in Ykspihlaja in Kokkola. After the shunting unit started pulling the wagons, the last bogie of the second last wagon carrying sulphuric acid moved onto the wrong track. Three wagons incurred damage as a result of the derailment. The track and the turnout were also damaged in the derailment area. The derailment did not cause disorder to the other train traffic. The cause of the incident was forcing open the turnout when shunting wagons. The opened forced turnout switched to its initial position while pulling underneath the wagon and the wagon s other bogie were directed onto the other track. The wagon derailed as a result of directed to two tracks. When shunting the wagons, the lookout was not conducted in sufficient way. The Accident Investigation Board of Finland is not issuing new recommendations as a result of the incident, but reminds operators that a lookout should be placed on the steps of the last wagon when several wagons are being shunted, if the wagon allows for this. If it is not possible to place a lookout on the wagon, the lookout should walk alongside the wagons as they are being shunted ahead. 41

44 Safety recommendations A total of 20 recommendations were issued in Almost all of them were addressed to more than one body or authority. Recommendations were directed at the following: Recipient No. Finnish Railway Agency 8 Finnish Rail Administration 13 VR Group Ltd 4 Ministry of the Interior - Ministry of Transport and Communications - Emergency Response Centre Agency 1 Finnish Road Administration - Municipalities 5 Other 4 Recommendations as per target gategory: Category No. Rolling stock 1 Track equipment 8 Rolling stock and track equipment - Traffic control equipment 3 Operating directions 7 Rescue operations 1 A total of 249 recommendations were issued from the beginning of 1997 until the end of At the end of 2008, the number of recommendations implemented was 154 (62%) and 41 (16.5%) recommendations were decided not to be implemented. The fulfilment of recommendations can take time, as indicated by the fact that, of the 196 recommendations issued from , 145 (74%) had been implemented by the end of 2008 and 32 (16%) were decided not to be implemented. 42

45 Implementation of recommendations during Recommendation implementation status Recommendation issued Implemented In progress Not to be implemented Year [No.] [No.] [%] [No.] [%] [No.] [%] TOTAL Rail accident investigations in Following table contains summary of investigations of accidents and incidents conducted by Accident Investigation Board during the last five years. Accidents investigated TOT Major accidents Collisions (A-investigations) Derailments Collisions Other accidents (B- and C- investigations) Derailments Level crossing accidents Occupational accidents Tram accidents Incidents (B- and C-investigations) Safety studies (S-investigations) TOTAL Personal injuries TOT Passangers Fatally injured Crew Third party Total Passangers Seriously injured Crew Third party Total Passangers Slightly injured Crew Total TOTAL N.B. Since November 2005 also includes third party (level crossing user, not pedestrian); fatally and seriously injured. 43

46 MARINE In marine transport, 26 accident investigations were initiated in 2008, 18 of which were level-d investigations. A fishing accident in the southern Baltic Sea led to the death of one fisherman. Of the investigations in progress, one involved a joint investigation with the UK s Marine Accident Investigation Branch, with Finland leading the investigation. Another international accident is being investigated by the Swedish accident investigation authorities, with an investigator from the Accident Investigation Board participating in the capacity of official observer. Ten category B or C marine investigations were completed in Also completed were 13 category D investigations. Nine investigations involved cooperation with the flag state of the vessel concerned or with the investigating authorities of the state of occurrence. This cooperation is based on International Maritime Organization (IMO) guidelines. Investigations were carried out in cooperation with the accident investigation authorities of Sweden, the Netherlands, Estonia, Germany, Norway and the UK. A safety study was completed on the safety of deck hatch cranes, in cooperation with the Dutch accident investigation authorities. The chief marine accident investigator has been a member of the marine transport advisory board s traffic section and a member of a Ministry of Transport and Communications working group tasked with preparing a government programme for improving maritime safety in the Baltic Sea. The Accident Investigation Board hosted the annual meeting promoting bilateral cooperation with Estonia s accident investigators in Helsinki in June. The executive director of the European Maritime Safety Agency (EMSA) visited the Accident Investigation Board on 27 August Our chief marine accident investigator participated in meetings of a permanent working group of the European Maritime Safety Agency (EMSA), held in Lisbon in March and November. The latter meeting involved the finalisation of a proposal for joint investigation guidelines. Furthermore, the European Marine Casualty Information Platform (EMCIP), an accident database developed by the EMSA, was finalised for launch in 2009, and the Accident Investigation Board investigators participated in training relating to this in Lisbon and in Helsinki. As part of the so-called Third Maritime Safety Package, the joint text was approved by the European Council and Parliament concerning the proposed EU Directive establishing the fundamental principles governing the investigation of accidents in the maritime transport sector. The Accident Investigation Board representatives have participated in the preparation of the Directive as a Ministry of Justice representative, in cooperation with the Ministry of Transport and Communications. 44

47 In May, the International Maritime Organization (IMO) approved a new code for the investigation of marine casualties and incidents, the so-called Casualty Investigation Code. This code will become mandatory once it is appended as part of the International Convention for the Safety of Life at Sea (SOLAS). The code will enter into force on 1 January Both marine accident investigators participated in the 17 th Marine Accident Investigators International Forum (MAIIF) held in October in Malta. The chief marine accident investigator, who had acted as the organisation s Deputy Chairman, was elected as its Chairman at the meeting. The fourth European Marine Accident Investigators' International Forum (EMAIIF4) was held in France. The chief marine accident investigator is also the Chairman of this annual regional MAIIF meeting. A one-day training session, entitled Introduction to Accident Investigation, was organised for new marine accident investigators in Helsinki in January. Our full-time investigators gave lectures and presentations on the investigation of marine accidents and their results in relation to more than 20 occasions in Finland and abroad, including a meeting of the International Safety Panel of the International Cargo Handling Assosiation (ICHCA International), and also to EMSA working groups and MAIIF and EMAIIF meetings. The chief marine accident investigator participated in an EU project, led by the Finnish Maritime Administration, which assessed the EU compliance of Croatia's legislation and maritime administration. In connection with a sub-project of this Twinning project, three meetings were held in Zagreb with the Croatian accident investigation authorities. The sub-project was concluded in Croatia in November, by means of an accident investigation seminar attended by 28 Croatian accident investigators and led by our chief marine accident investigator and marine accident investigator. Investigations commenced in 2008 Identifier Date of occurrance Title of investigation B1/2008M Cargo vessel MS TALI (FIN), grounding near Jössingfjord in Norway B2/2008M Fishing vessel FV MASI FIN-219-V (FIN), drowning off the southern coast of Sweden on 5 November 2008 B3/2008M Cargo vessel MS LEMO (St. Vincent and Grenadines), kitchen fire near Kotka C1/2008M MS OOCL NEVSKIY (LUX), grounding near Harmaja C2/2008M MS SERENA F (RUS), grounding in Hiittinen islands C3/2008M MS ANNE SIBUM (CYP), grounding at Orrengrund C4/2008M MS SEA WIND (SWE), engine room fire on the Åland Sea (participation in the investigation by the Swedish authorities) C5/2008M MS BIRKA EXPORTER (FIN) and FV HENDRIK SR (GBR), collision in the English Channel (joint investigation with U.K. authorities) In addition,18 D-investigations were commenced. 45

48 Investigations completed in 2008 Four of the investigation reports were published in Swedish and four translated into English. Two of the C-level investigations (C7/2006M and C3/2007M) were conducted by Swedish investigation authorities in which a Finnish investigator was involved in the capacity of official observer. These investigation reports are available only in Swedish. Identifier Date of occurrance Title of investigation B3/2004M Trawler SEA GULL, foundering in the Baltic Sea 1 B1/2005M M/S AMORELLA, fire on car deck 1 C2/2004M Trawler NORDSJÖ, heeling on the Northern Baltic Sea 1 C5/2004M MS KRASNOVIDOVO, collision with pontoon bridge in Kyrönsalmi strait C1/2006M MS ESTRADEN and MT WOLGASTERN, collision in the Kiel-Canal 2 C6/2006M Passenger Vessel MS NORDLANDIA, Collision with Quay in Tallinn 3 C7/2006M Loss of M/S FINNBIRCH between Öland and Gotland 4 C1/2007M M/S KRISTINA REGINA and barge CARRIER 5, collision in Danish territorial waters at Kadetrenden 2 C2/2007M MS CLAUDIA, grounding off Tornio 2 C3/2007M M/S ÅLANDSFÄRJAN, grounding off Maariehamn 5 1 (Published also in Swedish) 2 (Published also in English) 3 (Published also in Swedish and in English) 4 (Published in Swedish and English) 5 (Published only in Swedish) In addition, 13 D-investigations were completed. 46

49 B3/2004 M Trawler SEA GULL, foundering in the Baltic Sea on 27 May 2004 The Finnish trawler SEA GULL (AAL-18) departed on May from the Danish port Neksö at Bornholm to fish sprats in the Swedish fishing zone in the Baltic Sea. Later the vessel moved to the Polish fishing zone. The fishing was done by a trawl. When the fishers in the evening of 26 May were lifting the last pull, the vessel suddenly got a starbord list and stayed in that position. The skipper ordered the crew to don immersions suits and to prepare the life-raft. There were attempts to straighten the vessel by pumping fuel to the other side of the vessel by pumping fish cargo into the sea, but without success. The Finnish trawler HELLE KRISTINA, which had been fishing nearby and had been summoned to help, tried to straighten the SEA GULL with the help of its trawl-winch. The life-raft was launched and part of the crew transferred to the HELLE KRISTINA. Despite the attempts to straighten the vessel, the list increased. In the morning of 27 May there were attempts to get the SEA GULL into the HELLE KRISTINA s towage, but this did not succeed. At 8.30 the list of the SEA GULL suddenly increased and the vessel sank. According to the board of investigation the accident was caused by the fish cargo becoming porridgelike, which resulted in a large free liquid surface, and the stability of the vessel was significantly reduced. At the same time the lift of the trawl-bag done at the final stage of fishing caused the vessel to list and the fish cargo started to shift over the longitudinal fish-bulkheads to the starboard side. The list quickly increased to approximately 45 degrees, after which the listing got slower. The bulkheads could also have been damaged when the fish shifted, which contributed to the course of events. The vessel slowly made water and in about 10 hours it finally lost its stability and capsized when the list had increased to degrees. The Investigation Board gives recommendations to the Finnish Maritime Administration on the improvement of fishing vessel inspections especially when it comes to factors influencing the vessels stability and to the updating of stability documents. Common recommendations are given to organisations and schools operating in the fishing trade and to the Finnish Maritime Administration on the development of the contents of fishing vessel stability documents to a more tangible form for fishermen and on studying how the special risk situations connected to sprat fishing could be controlled. 47

50 B1/2005 M M/S AMORELLA, fire on car deck on 5 May 2005 Viking Line The Finnish Car Passenger Ferry ms AMORELLA had a fire on the car deck in the Stockholm archipelago The engine space of a Volvo 740, which was on a car deck caught fire and caused a serious incident. The vessel was on its normal route from Stockholm to Turku. There were 1,077 passengers and a crew of 159 onboard the vessel. There were 1,380 tonnes of cargo. The car decks were full of private cars, busses, trucks and trailers, and motorcycles. When AMORELLA was approaching the Oxdjupet strait in the Stockholm archipelago, the automatic fire detection system gave alarm of a fire at The fire detector gave alarm from the car deck on deck 5, i.e. from the so-called combi-deck. In addition to the car deck there are passenger and crew cabins on the deck in question. The combi-deck was fully laden with 89 private cars and three motorcycles. The alarming of the vessel s own personnel and fire fighting organisation and the messages about the fire to the maritime rescue centre and shipping company were realised according to the alarm list. A security guard with fire-fighting training started the initial fire-extinction efficiently together with a deck watchman who had been on a watch patrol. The security guard was in charge of the initial fire-extinction until the actual fire-fighting personnel came to the scene. The fire was limited and cooled down with the help of jets of water and the sprinkler system. The fire-chief and the fire-fighting chief directed the extinguishing from the car deck. The fire, which generated a lot of smoke, was finally extinguished by smoke divers using foam 29 minutes after the fire alarm. The fire-watch after the fire was extinguished was carried out in an appropriate way. Due to the threat caused by the smoke generation the 1,077 passengers onboard the vessel were evacuated to the outer decks or near the outer doors according to the evacuation plan and in a beforehand practised manner. The evacuation was completed almost at the same time as the fire was extinguished. When the extinguishing of the fire was secured, the passengers returned indoors and the vessel continued its voyage safely to Turku. The investigation has concentrated on the risk that a fire on the car deck of a car-passenger ferry during the voyage can cause to the safety of the vessel. The aim of the investigation has been to identify the risk factors connected with such fires and estimate the possibilities modern technology offers in improving the safety situation onboard. The Turku Police investigated the cause of the ignition of the car. In this case the fire, which totally destroyed the engine space of the car, had started due to a failure in an electrical device. It is impossible to estimate the fire risk of an individual car during the loading of the vessel because the safety of vehicles depends on the level of service and maintenance. The shipping 48

51 company and the vessel cannot influence this risk. On vessels transporting cars the methods for maintaining sufficient fire-safety include the detection of the fire at an early stage as well as correctly dimensioned and to their methods standardised and fast measures to limit and extinguish the fire. The investigation commission gives safety recommendations to shipping companies and training organisations within the maritime field on the organising of leadership, to the Emergency Services College on the development of extinguishing equipment and extinguishing methods in cooperation with the sea safety authorities and shipping companies, to the training organisations within the maritime field and shipping companies on the advancement of the fire-fighters' skills and knowledge, and to the shipping companies, the Ministry of Transport and Communications and the Ministry of the Interior on the development of the cooperation related to rescue at sea. C2/2004M Trawler NORDSJÖ, heeling on the Northern Baltic Sea on 27 January 2004 On 27 January 2004, the Finnish trawler NORDSJÖ (FIN-1107U) was on the Northern Baltic Sea lifting a trawl bag when it heeled degrees. The vessel was fishing in tandem with the trawler WIND. The entire joint catch had been lifted on board NORDSJÖ. NORDSJÖ notified the already departed WIND of the situation, which then made an emergency call. The cruise ferry SILJA SYMPHONY was nearby, coming to NORDSJÖ s assistance and continuing to provide wind lee after WIND had also arrived on the scene. NORDSJÖ s heel was almost fully rectified and it was able to proceed to Kasnäs, where the catch was transferred on board the NEW BALTIC. The investigation commission concluded that the accident was caused by taking too much cargo on board and an increased propensity for the fish to slide about due to water that had accumulated with the catch. The lifting of the trawl bag at the end of the fishing process heeled the vessel and the catch started moving to starboard over the vessel s longitudinal bulkheads. The cargo hold was divided into six compartments and quite full, with the result that the cargo heeling moment came into balance with the vessel righting moment at roughly 30 degrees. 49

52 In addition to the recommendations issued in connection with the SEA GULL capsizing, the investigation commission recommends that the Finnish Maritime Administration monitor the stability data of fishing vessels in connection with all possible cargo scenarios. The investigation report is appended to investigation report B3/2004M, Trawler SEA GULL, capsizing on the Baltic Sea on 27 May C5/2004M MS KRASNOVIDOVO, collision with pontoon bridge in Kyrönsalmi strait Savonlinna 18 July 2004 Timo Härkönen The Russian-flagged dry-cargo vessel ms KRASNOVIDOVO was on on its way to Varkaus nearly fully laden with timber. The Master of the vessel had been changed in St. Peterburg, and the Master now onboard was making this voyage on the KRASNOVIDOVO for the first time. This was also his and the Chief Officer s first voyage to the Saimaa Lake District. Three Finnish pilots had piloted the vessel from the Saimaa Canal to Savonlinna. The pilot carrying out the piloting from Savonlinna to Varkaus had embarked the vessel at 10.16, 20 minutes before the vessel entered the Kyrönsalmi strait which passes the Olavinlinna Castle. When the new pilot took over, the pilots had a brief discussion on the automatic steering of the vessel and the helmsman's skills. The pilot started piloting, and after a while he took over the manoeuvring of the vessel by manual steering. The KRASNOVIDOVO as a vessel type was not familiar to the pilot, and he did not have time to get enough feel for the manoeuvring of the vessel before it reached Kyrönsalmi. The pilot and the crew discussed neither the manoeuvring characteristics of the vessel in greater detail nor the earlier stages of the voyage. When the KRASNOVIDOVO approached the spar buoys located near the lower end of Kyrönsalmi, the pilot noticed that the current was very strong in the strait. The vessel entered the inhomogeneous current field of Kyrönsalmi at an angle after a turn which had been too long. The stopping of the turn and the starting of a counter-manoeuvre were delayed. At the same time the pilot handed the manoeuvring over to the Chief Officer without in advance informing about the change in the manoeuvring responsibility. The Chief Officer took over the helm, and the Master started to handle the engine. After the vessel had started to take a strong turn in the opposite direction, the Master could not stop this turn quickly enough. The vessel first hit the Olavinlinna pontoon bridge, which was open, and then the cliffs of the island at a speed of a little less than three knots. The damages on the KRASNOVIDOVO were minor, and the vessel was soon manoeuvred through Kyrönsalmi. 50

53 The KRASNOVIDOVO entered the Kyrönsalmi fairway at an angle, and the pilot handed over the manoeuvring to the Chief Officer just before the vessel reached the spar buoys without any prior notice. These two incidents together were the direct causes of the accident. In a narrow fairway with a current even slight deviations in the steering of the vessel require good knowledge about the conditions in the fairway and excellent manoeuvring of the vessel. The Kyrönsalmi fairway is one of the most difficult places in the Lake Saimaa deep-water channel system due to its geometry and currents. The fairway cannot be developed because of the historical significance of the strait, and this means that the only option considerably improving structural safety is transferring the fairway to Laitaatsalmi, which has already been planned. In the collision the fastening of the fixed part of the pontoon bridge came partly loose, and it was no longer possible to use the bridge as a traffic route. The first round of the Timo Mustakallio song contest was about to begin in the Olavinlinna Castle. At this point a decision was quickly made to convey passengers on a transport ferry, which had not been inspected as a passengervessel. The KRASNOVIDOVO accident occured in the summer when the use of the Olavinlinna Castle for big events for the general public is most active. The accident also showed that the only planned route for a great number of people back to the mainland from the Olavinlinna Castle is very vulnerable to disturbances. The historical constructions of the Olavinlinna Castle and its current usage as festival premises for thousands of people are inconsistent with current requirements on public safety. The safety arrangements at the Olavinlinna Castle constitute a whole, which is not restricted only to the usability of the pontoon bridge or the vessel accident. In 1989 the maximum allowed number of persons on the castle yard was established to 560. The combined width of the traffic routes to and from the castle is according to the current building regulations too small for the number of persons temporarily allowed at the gallery (2257) or at the whole castle (2777). The decisions on the maximum number of persons are based on a practice which has continued for a long time, and the situation has not been reassessed when there have been rebuilding of the gallery and stage structures in The Investigation Commission s safety recommendations relate to information about the currents in the strait to the users of the Kyrönsalmi fairway, the transfer of the deep-water channel, the development of piloting practices, the special requirements on pilots navigating the more demanding fairway sections, and a thorough investigation of the public safety of the Olavinlinna Castle. 51

54 C1/2006M MS ESTRADEN and MT WOLGASTERN, collision in the Kiel-Canal on 2 February 2006 MS ESTRADEN ( Mann & Son Ltd) MT WOLGASTERN. ( Seafoto Hannu Laakso.) The WOLGASTERN left the Holtenau lock of the Kiel-Canal at on 1st February Due to the 9-meter draught of the vessel, her highest allowed speed was 12 km/h (6.5 knots). The vessel approached the siding area of Audorf-Rade slowing down to let three ships (TURCHESE, ANTJE and ESTRADEN) behind it pass her. The ESTRADEN left the Holtenau lock of the Kiel- Canal at on 2nd February with a draught of 5.9 m. Three vessels were sailing one behind the other with the ESTRADEN as the last one. Their speed limit was 15 km/h (8.1 knots). According to the joint plan worked out in the bridge of the WOLGASTERN, the purpose was, at the straight of Audorf-Rade also to encounter the two smaller vessels, the LENA and the RIROIL 5. The ESTRADEN started to pass the WOLGASTERN from her portside while meeting the LENA. The WOLGASTERN started to turn starboard, but the correction succeeded with manoeuvring measures. At that time, the speed of the ESTRADEN was about 8.9 knots and that of the WOLGASTERN about 4.6 knots. When the ESTRADEN was about half her length ahead of the WOLGASTERN, they encountered the RIROIL 5. At that point, the speeds of the two vessels were about the same: the ESTRADEN 6.0 knots, the WOLGASTERN 6.5 knots and the RIROIL knots. After the meeting, the manoeuvrability of the ESTRADEN weakened, and she started turning to port. The ESTRADEN increased the speed of the vessel. At the same time it was notified from the bridge of the WOLGASTERN that the WOLGASTERN was turning portside and it was suggested that the ESTRADEN further increases her speed. The WOLGASTERN also increased her speed to improve her manoeuvrability as it was noticed that the vessel was restless. At that moment preceding the accident there were three vessels side by side in a part of the canal with a width of m (for that width depth of water was at least 10.5 m). The bow of the WOLGASTERN hit the ESTRADEN midships at about 02.36, whereupon the stern of the WOLGASTERN turned portside and collided with the aft part of the ESTRADEN. At the time of the collision the speeds of the vessels were about 8 knots. Due to the strength of the impact, the ESTRADEN started to turn to starboard and her aft part approached the portside bank. The Master of the ESTRADEN immediately took charge of the steering. Portside engine was reversed while the starboard engine was in ahead-position to prevent the aft from drifting portside towards the bank of the canal. This measure dropped the speed of the ESTRADEN so 52

55 that the WOLGASTERN started to slide along the starboard side of the ESTRADEN. The portside wing of the bridge of the WOLGASTERN hit the starboard deckhouse, mess and Master s cabin of the ESTRADEN. There were no people in these premises so personal damage was avoided. The WOLGASTERN slid past the ESTRADEN and drifted to the portside bank of the canal bow first. The investigation commission considers the reason of the accident the lengthy stay of the colliding vessels too close to each other at too high a speed due to meeting two encountering vessels one after the other. During the overtaking, the WOLGASTERN had to sail close to the right-hand side of the canal, which made its manoeuvrability more difficult. When the manoeuvring of the vessels became more difficult, their speeds were increased, which increased further the interaction forces due to the closeness of the vessels. The investigation commission addresses safety recommendations to the Wasser- Und Shifffahrtsdirection Nord to specify rules concerning overtaking situations in the canal. Safety recommendations are also addressed to the owners operating in the canals and to other bodies in connection with the canal navigation to estimate additional education needs of their personnel concerning the effects of restricted waters. In addition the committee recommends that maritime training institutes should complete their training concerning the effect of confined waters in ship navigation. C6/2006M Passenger Vessel MS NORDLANDIA, Collision with Quay in Tallinn on 28 October 2006 Eckerö Line The Finnish car-passenger ferry M/S NORDLANDIA had departed from Helsinki to Tallinn on at The weather was fine at the time of the departure. The meteorological institutes in both Estonia and Finland had forecasted storm in the Gulf of Finland with westerly-northwesterly winds m/s. The wind increased during the voyage. The option of waiting for the weather to improve was never discussed. When the NORDLANDIA was approaching Tallinn, the master, the officer of the watch and the helmsman were on the bridge. The vessel s wind-meter showed that the wind was from northwest and that the wind speed was over 20 m/s. The chief officer also arrived on the bridge, and the master told him that he had ordered tug assistance. The master was steering the vessel to the basin at a greater speed than usual from the port wing steering place. He tried to get the tug to assist. In the berthing the bow of the vessel hit the quay constructions. 53

56 The port side of the NORDLANDIA s bow was damaged when it hit the quay fender. As to the port constructions, the fender and the covered passenger gangway were damaged. The collision with the quay did not cause personal injuries and the damages did not endanger the safety of the vessel while it was in port. The NORDLANDIA was approaching the port and hit the quay construction in wind conditions which exceeded the performance of the vessel. No information about the vessel's operational limitations had been produced to the master. The speed at which NORDLANDIA entered the port, the track and the lack of advance discussion suggest a traditional and established procedure in good weather conditions. The fact that the circumstantial factors were taken into consideration can be seen mainly in the high speed of the vessel, which was used in order to try to control the effects of the wind. The shipping company has no standard procedures for mooring. Each master has to develop his/her own routines. This means that the advance discussion on the distribution of work and on communication, which is essential for bridge co-operation, becomes more difficult or non-existent. In the same way there should be a common, pre-agreed plan of action for the co-operation with tugs. According to the prevailing practice, the routines can vary within one shipping company and as the masters change, even on one vessel. The responsibility for port manoeuvring has been allocated to the master alone, but he/she has been left without support in the decision-making. The environmental limitations for port manoeuvring have not been set, and there are no minimum requirements as to the steering devices. The SOLAS Convention rule on the operational limitations for a passenger vessel has not been applied to the wind limits of port manoeuvring. The Finnish Maritime Administration has not required this from the shipping companies. The operational limitations can be used as the basis for defining the port-specific wind limits for vessels. In their training, ship officers can be provided with port manoeuvring skills only within a vessel's operational limitations. The general character of the training requirements set in the STCW Convention are the reason for the fact that the present ship officer training does not include adequate requirements to control port manoeuvring. The operational limitations can give the STCW objectives on ship officers' skill levels in port manoeuvring a realizable and realistic framework. The Investigation Commission has issued two safety recommendations to the Finnish Maritime Administration and one to the shipping companies. All recommendations are connected with the SOLAS Convention requirement on the vessels operational limitations and defining them for the purpose of port manoeuvring. 54

57 C7/2006M Loss of M/S FINNBIRCH between Öland and Gotland 1 November 2006 On October 31, the Swedish ro-ro vessel FINNBIRCH departed on schedule from Helsinki for Århus, Denmark. The vessel had a full cargo consisting of trailers and a paper roll consignment in the cargo hold. The weather was poor, with northern winds of 20 m/s that reached up to m/s in gusts. On her journey between Öland and the Gotland islands the vessel experienced repeated inclines to the left in rough seas. Thereafter, the vessel was left with a steep degree incline due to the movement of the cargo. The vessel immediately sent a Mayday signal, which led to a long and arduous rescue operation. The crew gathered on the lifeboat deck wearing rescue suits. However, evacuating the crew to nearby vessels proved impossible, as did rescuing them by helicopter, due to the poor weather conditions. The crew remained stuck on deck until the vessel finally capsized and sunk four hours later. A crew member drowned with the vessel and one was lost to hypothermia. The remaining crew members were rescued by helicopter from the sea. The causes of the accident were as follows: FINNBIRCH s course in rough seas destabilised the vessel from its rear, and steep but not exceptional big inclines caused the cargo, which had been poorly secured, to move. The following factors contributed to the accident: the cargo fastening manual was not up-to-date and instructions were not followed. The shipper had its own cargo fastening system and the vessel s manual was not used. Moreover, the fastening was based on verbal agreements between the shipper and the vessel's officers and the shipping company had not been notified of the shortcomings in the cargo fastening manual. Neither the shipping company nor the supervisory authorities had noticed that the vessel s cargo fastening practices were in major breach of the set requirements. As a result, Haverikommission presented 13 recommendations to the Swedish marine traffic authorities and one recommendation to the aviation authorities. 55

58 C1/2007M M/S KRISTINA REGINA and barge CARRIER 5, collision in Danish territorial waters at Kadetrenden on 29 May 2007 M/S KRISTINA REGINA Proomu CARRIER ( Vesta Marine) The Finnish passenger vessel KRISTINA REGINA was on her way from Amsterdam, Holland to Helsinki, Finland, when she, in poor visibility in the maritime area between Denmark and Germany collided with the Swedish barge CARRIER 5 tugged by tug boat PIONEER sailing under the flag of Comoros on 29 May There was no damage to persons nor did KRISTINA REGINA suffer significant damage. On the other hand, the barge PIONEER lost part of the logs it was carrying and it sustained significant damages in her hull which were later repaired at a shipyard. At the time of the event KRISTINA REGINA was using two radars, one of which was scaled to 6 miles and the other to 3 miles. Due to the sea clutter, the close radar targets could not be monitored with ARPA. KRISTINA REGINA has an AIS transmitter and receiver and its display is located on the bridge next to the radar. The VHF equipment and fog signal device of KRISTINA REGINA were operational and in use at the time of the event. The navigation lights of the vessel were operational at the time of the event. The tug boat PIONEER was using one radar and the watch officer was using various scales between 3 and 12 miles. The other radar was not functioning at the time of the event. PIONEER had no AIS device because the size of the vessel is below 300 GRT. The fog signal device was not functional. The VHF and the navigation lights of the vessel were operational at the time of the event. Onboard neither vessel was the master called to the bridge although the visibility was poor. After KRISTINA REGINA arrived in Helsinki, it turned out that the voyage data recorder of the vessel, VDR, had not recorded any information after 16 December Assistance in the investigation was obtained from the Swedish, Danish, German and British maritime investigation authorities. The tug boat and barge combination had no AIS device; noticing the combination was only based on a radar echo or a visual observation. According to the provisions, a vessel exceeding 300 GT has to have AIS equipment. However, the requirement does not concern a barge combination, only a vessel. When the tug boat is under 300 GRT, the combination need not be equipped with an AIS device even if the combination were tenfold compared to that required. A mistake has 56

59 been made when creating the provision. It has not been known that this type of combinations may make poor radar targets. The investigators recommend that the Finnish Maritime Administration undertake measures to make the AIS requirement concern all vessel combinations exceeding 300 GT. C2/2007M MS CLAUDIA, grounding off Tornio on 23 October 2007 Wijnne & Barends The Dutch break bulk carrier CLAUDIA had loaded a cargo consisting of 4,781.5 tons of stainless steel in Tornio. The vessel departed from the harbour under pilotage along a fairway where fairway construction work was being done. When approaching a fairway part called Portti at the bend of the fairway, the pilot was blinded by the lights of the dredgers dredging in the area so that he could not see the lights of the buoys bordering the narrow passage. The sea clutter caused by the roll of the sea disturbed the vessel s radar picture in such a way that the echoes generated by the buoys could not be discerned from the sea clutter of the nearby area. The turn was short and the vessel went aground and was damaged to its bottom. There were no environmental damages. The CLAUDIA returned to Tornio, where its damages were surveyed and the cargo was unloaded. The vessel was docked in Gdynia, Poland, and it later returned to the Tornio traffic. It came up in the investigation that the instructions with reference to the cooperation during the fairway construction work between the vessel traffic and those performing the dredging operations had not been detailed enough. The dazzling floodlights of the dredgers combined with the vessel s inadequate route planning caused the turn ending at the gate of a narrow fairway to fail. The investigators recommend that the fairway constructors specify traffic principles and procedures in their contracts. It should be made sure in the annual inspections and surveys that vessels have route plans. In addition to this, it is recommended that the Finnish Maritime Administration and the Finnish State Pilotage Enterprise form a workgroup to look into the availability of hand-held computers containing electronic charts suitable for pilots use. 57

60 C3/2007M M/S ÅLANDSFÄRJAN, grounding off Maariehamn on 23 October 2007 ( Viking Line) The passenger vessel ÅLANDSFÄRJAN left Kapellskär on 23 October 2007 at 12.01, on its way to Maariehamn. The vessel approached the fairway leading to Maariehamn and the first mate informed Archipelago VTS that the vessel had arrived in the VTS area. After this, he went to the computer next to the vessel s map table to check the work schedule. The vessel s master was in his cabin. He looked at his watch and noticed that he should already have been called to the bridge because the vessel should have been approaching Marhällan beacon at the time. The master went to the bridge and saw the first mate standing by the map table and the lookout sitting on a chair. The master noticed immediately that the vessel was heading straight towards Marbådan island, which was directly in front of the vessel about 200 metres away. It was clear that the vessel would run aground. The master rushed to the starboard wing conning position and reversed the engines. The vessel s speed was somewhat reduced but it nevertheless ran aground at 13 knots. The accident occurred on 23 October 2007 at on coordinates N 60 01,56 E ,40. The accident investigation was carried out by the Swedish maritime authority and the report is available only in Swedish. 58

61 Safety recommendations The marine accident investigation reports completed in 2008 contained a total of 27 safety recommendations. These recommendations were primarily directed at the Finnish Maritime Administration, shipping companies and training organisations. Several recommendations were addressed to more than one recipient. The safety recommendations are summarised below by recommendation area and recipient. Recommendation area No. Vessel operation directions 8 Pilotiage directions 2 VTS directions - Directions for emergency radio communications 1 Navigation and route planning 4 Navigation channels and their marking 4 Vessel equipment and facilities 1 Vessel stability 6 Boating information campaign - Other 1 List of recommendation addressees. Recipient No. Finnish Maritime Administration 11 Other authorities 6 Finnish State Pilotage Enterprise 1 Shipping Companies 5 Organisations 3 Forwarders - Training organisations 5 Our recommendation monitoring requested and received information on the implementation status of 117 recommendations. A total of 174 recommendations were issued during

62 Marine investigations in Accidents investigated TOT Major accidents (A-investigations) Other (B- and C-investigations) (8) 3 8 TOTAL Safety studies Investigations as per accident category TOT Grounding Fire Sinkning Collision Other TOTAL Investigation as per resulting impact TOT Fatally injured Seriously injured Slightly injured Environmental damage

63 OTHER ACCIDENTS In 2008, investigation work was carried out by four investigation commissions investigating serious incidents. One of these had started in 2006, two in 2007 and one in In addition, eight D-level investigations were in progress. Of these, two were initiated in 2007 and six in D-level investigation results have been published on the internet since In 2008, two investigations of serious incidents were completed. These concerned an explosion at an excavation site in Espoo in 2006 and a fire at a psychiatric hospital in Nokia in January Both investigations included studies of accidents or incidents similar to the ones that had occurred. Also initiated was an investigation of a fire at a wooden house serving as a shelter for people with substance abuse problems. The fire, which was classified as a serious incident, led to the death of five people, and its investigation had not yet been completed at the end of Another investigation, still incomplete at the end of the year, concerned the entry of treated wastewater into the municipal drinking water network in Nokia in November Of the minor category D investigations started in 2008, two involved an internal roof collapse. One occurred at a lunch restaurant and the other at a school cafeteria. Other D-level investigations relating to structural integrity concerned the collapse of the covering of a sports arena and the failure of the support structures of a concentrate unloading funnel in a port. The remain two investigations concerned an explosion at an asphalt station and the buckling and tilting of a gas washer at a steel mill. The school shootings that occurred in Jokela in November 2007 and in Kauhajoki in September 2008 were not accidents but premeditated acts, and therefore their investigation was not possible in accordance with regulations governing accident investigation. Finland s Parliament passed a new law enabling their investigation. The Accident Investigation Board s director and a chief accident investigator participated in the investigation of these cases in

64 Investigations commenced in 2008 Identifier Date of occurrance B1/2008Y Title of investigation The fire in the wooden house functioning as a shelter for people with substance abuse problems In addition, six category D investigations of minor accidents or incidents were begun in Investigations completed in 2008 Identifier Date of occurrance Title of investigation B3/2006Y Accident at an excavation site in Espoo B1/2007Y Fire at Pitkäniemi Hospital in Nokia In addition, 6 category D investigations were completed in B3/2006Y Accident at an excavation site in Espoo on 24 April 2006 A semi-detached house was being built in Friisilä, Espoo. The construction work was in its early stages, and excavation work for the foundation and public utility services was in progress. The site consisted mainly of open rock, which was meant to be removed by quarrying approximately 1,400 cubic metres. Over the preceding month, approximately 70 explosions had been performed, and the plan was to carry out approximately five more. The site was in the middle of a residential area, so the rock was quarried in small fields. On Monday, 24 April 2006, one explosion was carried out at 7:57 a.m., after which the holer continued to drill new holes with the drill carriage. The charger and charger s assistant prepared the next field and moved the explosives, placed in boxes, next to the driveway, approximately 5 6 metres from the field. A blasted rock transportation vehicle (lorry) arrived on the site to transport the quarried rock material away. The lorry driver backed along the driveway near the boxes of explosives, and an excavator began loading rocks on the platform. At the end of the loading, a rock was left precariously atop other rocks on the platform, then slid over the side of the platform and fell on the explosives. This resulted in a heavy explosion, causing the lorry, drill carriage, and compressor to catch fire. The emergency centre received several emergency calls regarding the accident, and, in total, six rescue service units and 10 medical rescue service units were called to the site. Police were responsible for isolating the area and clearing away the explosives. 62

65 The rescue operations of both rescue services and medical services were successful, despite the fact that initially, given the seriousness of the accident, too few resources were called in. Rescue operations were further hindered by the detonators exploding in the ground because of the fire, but, as a whole, the situation was handled well. The direct cause of the accident was an approximately 400-kg rock falling on boxes of dynamite during loading of quarried rock material. A problem was created by the loading taking place in the same narrow location in which the site's explosives were kept. The site was designed inadequately, and managerial decision-making relationships were unclear. The implementation of blasting work was based on experience and oral instructions and agreements. Methodicalness is emphasised in regulations concerning blasting work, but work site visits and other accidents indicate that, in practice, the intended methodicalness and drafting of appropriate plans has not been implemented on small work sites. One background factor for this is that the current notification procedure does not function and thus smaller blasting sites are actually not monitored at all, nor are they even subject to a threat of monitoring. In addition, regulations regarding blasting work have been scattered amongst various administrative sectors provisions and instructions, which have been complemented over the decades. As a result, regulations can be obscure and even give rise to different interpretations. The Investigation Commission recommends improving the prerequisites of blasting site monitoring and implementing this monitoring via annual checks, for instance. To clarify blasting regulations, the commission recommends that different ministries renew, update, and align the legislation related to explosives with an aim to removing details and moving toward independent risk management in the industry. However, renewals must cater for the needs of the industry s broad small-enterprise field. In addition, the Investigation Commission recommends that operators in the quarry industry be mandated to take out sufficient liability insurance to ensure compensation for external sufferers of damage. To develop rescue services, the commission recommends that the Ministry of the Interior, in cooperation with the Association of Finnish Local and Regional Authorities, would find out how the address system and its updating is functioning and if it is needed to give more detailed regulations in addition to the existing guidelines. 63

66 B1/2007Y Fire at Pitkäniemi Hospital in Nokia on 25 January 2007 The Pitkäniemi hospital raised a fire alarm on 25 January 2007 at 11:18 p.m. Outdoor clothing hanging in an open wardrobe located in the corridor of acute psychiatry ward APS 9 had caught fire. Though the fire was detected at an early stage, attempts by the staff to put it out were ineffective, because the clothes in the wardrobe burned easily and created an ample supply of flammable gases. Heat caused fire gases close to the ceiling to catch fire, resulting in rapid and violent escalation of the fire to inner parts of the building, while ample and toxic fire gases limiting visibility spread to all corridor areas as well as patient rooms. In this difficult situation, the staff managed to evacuate 10 of a total of 18 patients, and the fire brigade rescued the remaining eight. Department APS 9 was located in building 17 of the Pitkäniemi hospital, caring for patients suffering from long-term illnesses. The psychological status of the patients and sedative medication administered to them complicated the evacuation carried out by the nurses in a situation in which their own safety was at risk. In addition, the fire brigade rescuers experienced difficulties in trying to communicate with the ill and medicated patients. In total, 18 people were injured in the accident. Of these, 15 were patients in the APS 9 ward, two were nurses from the same ward, and one was the head night nurse of the hospital. The material damage came to approximately EUR 500,000. The police performed a fire cause investigation for the accident site, based on which the fire was found to have been started with a match by one of the patients. On the basis of the accident investigation, the following issues, alongside others, were found to have contributed to the creation and escalation of the accident. Information on the patient s inclination to start fires had not been communicated to the nurses on duty on the APS 9 ward. Ignition of the fire was enabled by the established practice of storing clothes on the ward. Storing clothes in a completely or partly open rack in the corridor created a fire risk that had not been recognised. Had the building been equipped with an automatic fire extinguishing system, the fire would have gone out before spreading outside the wardrobe. A serious accident was avoided because all of the following occurred during the incident: 1) The fire was detected almost immediately by the head nurse who coincidentally arrived at the place where the fire was ignited. Extinguishing attempt and evacuation began without delay and the fire brigade was alerted immediately. 2) The staff managed to evacuate 10 patients after the extin- 64

67 guishing attempts even though the conditions developed life-threatening and difficult in 2 3 minutes. 3) Despite the great power of the fire, the amount of fire load was minor, rendering the intensive phase of the fire quite short. 4) Thanks to the operations of the fire brigade, the eight patients still inside the hospital were rescued at the last minute. To avoid similar accidents occurring in the future, the Pirkanmaa hospital district has decided to equip the Pitkäniemi hospital with an automatic extinguishing system by the end of The Investigation Commission issues five recommendations, related to 1) the development of patient legislation to improve safety, 2) the drafting of risk management instructions for nursing homes and other nursing institutions, 3) equipping nursing homes and other nursing institutions with an automatic sprinkler system, 4) drafting of target-specific overall response times, and 5) improved fire safety for storage of outdoor clothing. A dissenting opinion concerning automatic sprinkler systems left by a member of the investigation commission is at the end of the report. Safety recommendations Two investigations on the threat of a major accident were completed in These produced a total of 9 new safety recommendations, which were targeted as follows: Recommendation type No. Safety in care facilities 3 Explosion and mining work 3 Development of rescue operations 2 Clothing storage facility fire safety 1 Since the year 2000, a total of 168 recommendations have been issued concerning other accidents. The recommendations are classified by accident type as follows: Accident type No. % Road traffic accidents 50 30% Fires/explosions 44 26% Building incidents 34 20% South-East Asia tsunami 27 16% Industrial accidents 13 8% TOTAL % Of the 168 recommendations being monitored, 66 (39 %) were verifiably implemented by the end of the year

68 Other accident investigations in Accidents investigated TOT Major accidents (A-investigations) Incidents of serious accidents (B-investigations) Safety studies (S-investigations) Other (D-investigations) TOTAL Investigation classification according to nature of incident TOT Fires/explosions in industry Military accidents/incidents Structural damage incidents Bus accidents/incidents Natural disasters Fires Explosion accidents/incidents Hazardous substance leakage Waterborne disease TOTAL Investigation classification according to area of activity TOT Commercial or similar activity Home or leisure-time TOTAL

69 Investigation classification TOT Tank according piping to site or object of accident Exhibition/sports hall Medium-sized shop or building of similar level of public use 1 1 4* Large shop/shopping centre Indoor swimming pool or bath Bus/coach Explosives or weapons Natural disaster Residential building Institution (nursing/medical/welfare) Hotel TOTAL *one of the four cases involves investigations into structural damages at eight different localities Personal injuries and fatalities during (category D investigations not included) Injuries / fatalities TOT Employee Fatally injured Seriously injured Slightly injured Passanger, customer or resident Total Employee Passanger, customer or resident Total Employee Passanger, customer or resident > > Total TOTAL In the South-East Asia natural disaster 179 Finns died and 250 were injured. Cases of serious or mild injury are not separately classified in the investigation. The Konginkangas accident fatalities included the bus driver and 22 passengers. A further 14 passengers were seriously injured. 2 Military service conscripts 3 Over 8000 people got ill by the waterborne disease in Nokia. 67

70 PERFORMANCE Impact The Accident Investigation Board has an impact on safety through the communication of its investigation results, especially the safety recommendations contained therein. The Accident Investigation Board also follows up the adoption of the recommendations. The safety recommendations issued in are summarised in the table below. Total recommendations and average recommendations per investigation TOT Aviation Rail Marine Other TOTAL as per investigation 1,2 1,4 1,2 1,3 1,1 The number of safety recommendations issued in 2008 declined from 2007, returning to the level of previous years. In 2007, the scope of rail accident investigations was expanded to cover fatal level-crossing accidents, which has increased the number of investigations and recommendations. In addition, a safety study on level-crossing accidents was completed in 2007, and included a substantial number of safety recommendations. The recommendations issued under the other accidents classification varies considerably because the number of investigations completed is lower and the variation higher. The number of recommendations issued per investigation was lower than in previous years, indicating that the policy of exercising prudence in the issuance of recommendations has been continued, as set down in the Accident Investigation Board s operating manual. The implementation status of safety recommendations are summarised in the table below. 68

71 Total recommendations and their implementation status ( ) from 2000 onwards Issued Implementation status known Implemented Will not be implemented Aviation % 48 % 15 % Rail % 50 % 15 % Marine % 26 % 19 % Other % 39 % 3 % TOTAL % 41 % 13 % In recent years, a major focus has been placed on the follow-up of recommendations, and in 2008 we achieved our target for the first time of keeping track of the implementation status of at least 80 percent of recommendations issued. We also have more accurate information on recommendations that have been implemented and those for which a non-implementation decision has been made. These make up 54 percent of all recommendations, which means that 40 percent are either being implemented, are under review or are pending. Objectives have also been set on the safety recommendation implementation percentage in various investigation branches. The objective was 55 percent for rail traffic and aviation, 30 percent for marine transport and 40 percent for other accidents. This objective was not fully achieved by any investigation branch. In total, 41 percent of the recommendations issued from had been verifiably implemented. The fulfilment of recommendations can take time. Of the recommendations issued from for rail traffic, for example, 74 percent of recommendations had been implemented and a decision taken not to implement them in 16 percent of cases. The Accident Investigation Board s investigation reports are an important means of influencing people's conduct, and are also available on the Accident Investigation Board s website. The aim is to distribute the reports in such a manner that their impact on safety is as great as possible. In 2008, the Accident Investigation Board s investigators gave presentations or lectures in more than 60 seminars and training sessions, some of these involving several presentations. The investigators participated actively in Nordic cooperation and EU working groups for improving accident investigation performance among aviation, rail and marine accident investigation authorities. In addition, the Accident Investigation Board participated actively in the preparation of an EU directive on marine accident investigation and the development of an International Maritime Organisation accident investigation code, both of which were completed in Preparatory work on a new directive on the investigation of aviation accidents was also initiated last year. The Accident Investigation Board participated in this preparatory work. Cost-effectiveness A key accident investigation outcome is the accident investigation report. Cost-monitoring cannot be used to accurately calculate the cost of an individual investigation, however. A rough estimate of the unit cost of B-level and C-level investigations, i.e. accidents and minor incidents, is nevertheless given below. The calculations are based on the operating expenses of the Accident Investigation Board, the expenses of the investigation commissions and expenses incurred from hiring 69

72 outside experts. Only the funds used for the investigation of major accidents, i.e. A-level investigations, have been excluded from this sum. There were no Investigations of major accidents during the years The table below summarises cost indicator per completed category B and C investigations, and also per commenced and completed category B and C investigations. This unit cost is greater than the actual cost per category B and C investigation because it includes all pre-investigation expenses, training expenses and international cooperation expenses. Summary of accident investigation costs for B-level and C-level investigations Year Commenced B- and C-level investigations Completed B-and C-level investigations Total costs Costs per completed B- and C-level investigation completed Costs per B- and C-level investigation commenced and completed The costs per investigation in 2008 were somewhat lower than in 2007 because more B-level and C-level investigations were initiated and completed in 2008 than in Costs in 2008 were on the same level as they were in There was a distinct jump in costs from the level of to the level of , partly due to the completion in of several investigations that had been ongoing for a long period. Some of the difference is also explained by the introduction of the minor category D incident practice in 2004, which reduced the number of broad investigations, but increased the total number of investigations. The total costs of all accident investigations, i.e. including level-d investigations and safety studies, are summarised in the table below. Summary of accident investigation costs for all investigations Year Investigations commenced Investigations completed Total costs Costs per investigation completed Costs per investigation commenced and completed 2005 * * Exluding level-a investigations 70

73 The expenses per investigation rose somewhat because fewer investigations were completed than in 2007 due to a reduction in the number of D-level investigations completed. In addition, expenses rose by about 5 percent. The costs of individual B-level and C-level investigations completed in 2008 varied considerably, from a few thousand to tens of thousands of euros. In addition, the search for aircraft parts following an accident towards the end of the year resulted in expenses of about EUR 100,000. Productivity Working time monitoring, as registered by the TARMO working time monitoring system, indicated that a total of 12,77 working years of personnel resources were spent on the following activities: Activity working years % working years % working years Accident investigation 4,72 41,4 4,50 36,7 5,25 41,1 Investigation development 0,35 3,1 0,67 5,5 0,78 6,1 International co-operation 0,86 7,5 0,93 7,6 1,15 9,0 Other stakeholder co-operation 0,60 5,2 0,73 6,0 0,77 6,0 Support operations 1,64 14,3 1,87 15,3 1,62 12,7 Financial administration support 0,78 6,8 0,75 6,1 0,78 6,1 Personnel administration support 0,14 1,2 0,13 1,1 0,12 0,9 Personnel development 0,96 8,4 1,00 8,1 0,51 4,0 Other 0,06 0,5 0,08 0,7 0,07 0,6 Absences 1,33 11,6 1,58 12,9 1,72 13,5 TOTAL 11,44 100,0 12,24 100,0 12,77 100,0 % 71

74 Distribution of Working Time 2008 Accident investigation Investigation development International co-operation Other stakeholder co-operation Support operations Financial administration support Personnel administration support Personnel development Other Absences Working Years As in earlier years, the major share of staff resources, i.e. more than a third, was devoted to accident investigation proper, whose share recovered from a dip in 2007 to the same level as in The number of working hours spent on developing the investigation process continued to grow and was around twice as high as in Investigation development includes activities such as recommendation monitoring and handbook development. Most of the working hours spent on accident investigation, i.e working years, were spent on drawing up accident investigation reports. The second greatest amount of time, i.e working years, was spent on investigation work. International cooperation in the field has substantially increased in recent years, and the amount of time spent on this activity increased somewhat from Investigators also spent a large amount of time preparing and giving presentations on accident investigation and safety. These were recorded as staff development and stakeholder cooperation. It is therefore somewhat surprising that stakeholder cooperation remained on the same level as in the preceding two years and the number of hours recorded under staff development fell by half. In 2008, external investigators received compensation for 14,673 working hours from the allocated investigation budget, which is the equivalent of around 7.73 working years. The table below indicates that the working time of both in-house staff and external investigators, expressed in working years, has grown somewhat from From 2006, the working time of in-house staff grew by one working year. This is explained by the fact that a vacant chief investigator s position was filled in May 2006, increasing the staff number from 10 to 11. Workplace wellbeing The number of employees has been 11 since 1 May 2006, before which the number was 10. The latest staff changes took place in 2007, with the appointment of a new chief aviation accident 72

75 investigator and a new aviation accident investigator. This somewhat reduced the average staff age as shown in the table. Many staff members will reach retirement age over the forthcoming years. Preparations are being made in this regard, for example by developing routines and documenting working practices Average age of employees 54,0 53,1 51,8 52,8 Education level index 4,9 5,1 5,3 5,3 Sick leave days 61,0 17,0 32,0 63,0 Sick leave days / 1 working year 6,5 1,5 2,3 4,6 Staff health services are offered through Terveystalo. The employment contract provides for an annual health check up. Absences due to illnesses per working year were twice as high as in 2007, but still quite low considering the average age of staff. The target of achieving an average of three sick-leave days due to illness per working year was not reached. Due to the low number of staff, even one extended sick leave can have a significant impact. In 2008, a total of EUR 404 per working year was spent on health care and occupational health. The corresponding amount was EUR 298 in 2007, and EUR 512 in In 2008, a total of 88 external experts were hired on an hourly basis. This number has not varied significantly in recent years, given that the corresponding number was 92 in 2007 and 82 in In 2008, a one-day training session was organised for new external researchers in each of the four investigation branches. In addition, a two-day training session was organised for external aviation investigators, with more than 30 participants. The aviation accident investigator and an external investigator participated in an investigation course for the Finnish Air Force. Furthermore, the aviation accident investigator completed a two-week basic course for aviation accident investigators organised by the U.S. National Transport Safety Board, and held in the U.S. The chief aviation accident investigator participated in a course on aviation safety command systems in the Netherlands. Finally, the chief accident investigator for other accidents enrolled in a safety leadership training programme lasting until 2010, the programme being organised by the Helsinki Institute of Technology s TKK Dipoli training centre. The Accident Investigation Board organised a seminar on the role played by fatigue in accidents, together with the Finnish Motor Insurers Centre. A total of 87 persons participated in this seminar. Service quality and performance The Accident Investigation Board s objective, as set by the Ministry of Justice, is to complete serious incident or accident investigations (category B) in one year, and minor accident investigations or incidents (category C) in six months. The following tables summarises the durations of category B and category C investigations for the years

76 Average investigation duration for completed category B investigations Completed investigations Average duration in months Completed investigations Average duration in months Completed investigations Average duration in months Aviation 4 12,0 2 12,4 2 7,7 Rail ,5 4 11,1 Marine 3 20,3 5 29,7 2 38,5 Other 1 11,0 3 15,1 2 19,7 TOTAL 8 14, , ,6 Average investigation duration for completed category C investigations Completed investigations Average duration in months Completed investigations Average duration in months Completed investigations Average duration in months Aviation 8 8,5 4 10,5 7 8,0 Rail 8 20,4 9 13, ,8 Marine 5 17,5 4 15,5 8 25,5 Other TOTAL 21 15, , ,4 The Ministry of Justice has set the objective of the average duration of category B investigations not exceeding 18 months. This objective was met, although the average duration of investigations with respect to marine traffic clearly exceeded the target. The duration of investigations of aviation and rail traffic continued to shorten. Because a relatively low number of B-level investigations are completed each year, a few lengthy investigations can have a significant impact on the average. EU directives have set a time limit of one year for accident investigations, and this is the objective being sought. The average duration of category C investigations has not been reduced, despite efforts to do so, and in 2008 the duration fell significantly short of the set target of nine months. The target was only achieved in aviation. In practice, the scope of category C investigations does not differ significantly from a category B investigation. Indeed, the investigation report for both types of investigations follows the same format. Efforts have been made to shorten the investigation period by means of improved investigation monitoring, which has possibly contributed to the decline in investigations ongoing for more than one year, as indicated in the table below. With respect to marine traffic, the situation has also improved, given that two of the oldest ongoing investigations were completed in 2009 and the remaining ones were all begun in 2008, except for three safety studies included in the figure in the table (5). 74

77 Number of accident investigations in progress for more than one year At the end of 2006 At the end of 2007 At the end of 2008 Aviation Rail Marine Other TOTAL The Accident Investigation Board s operating handbook and investigator s instructions were updated during 2008, as the need to do so arose. The Accident Investigation Board s quality handbook has been in use since Internal audits were initiated in 2008 in accordance with the auditing system, with an audit each spring and autumn. The aim is to continue the process in such a way that all systems are audited over the next three years, with everything completed by the end of

78 FINANCES The Accident Investigation Board s budget allocation has remained more or less unchanged since the year 2000, comprising a very tight budget. The 2008 allocation was EUR 1,030,000, which was supplemented by an additional allocation of EUR 52,000 in the autumn. Salaries accounted for EUR 803,770. The increase on 2007 is explained by statutory salary increases for government employees. Thanks to strict budgeting, around EUR 13,000 of the budget allocation was carried over into Salaries and rents, which accounted for about 85 percent of all expenses, increased, but other expenses fell from 2007 by about EUR 26,000, i.e. around 14 percent. The accident investigation appropriation grant was raised from 500,000 euros to 1,000,000 euros in 2006 and remained the same in In 2008, the total allocation expenditure amounted to EUR 1,181,632. The search for aircraft parts in the wake of the flight school crash in Taipalsaari was a demanding effort involving substantial additional costs. The investigation costs of the Jokela and Kauhajoki school shootings, which were investigated based on specially drafted legislation, were settled from the appropriation, although these costs are not included in the table below. No major accident investigations were underway in 2007 or The size of the annual allocation for investigation work is difficult to estimate and predict because the allocation depends on the nature and number of accidents. The costs of individual investigations vary significantly depending on their nature and scope. Expenditure of the Accident Investigation Board and its accident investigation activites in (financial statement data) Year Salaries Rent Other expences Accident Investigation Board Investigation allowance TOTAL

79 Total Expenses of the AIB Salaries Rent Other expenses Investigation allowance

80 INVESTIGATIONS COMPLETED IN 2008 Identifier Date of occurrance Title of investigation B1/2007L Glider accident at Haapavesi aerodrome B2/2007L An air proximity incident in Pori Terminal Artea C1/2007L Serious incident at Seinäjoki Airport C3/2007L Emergency water landing off Helsinki C5/2007L Incidents at Helsinki-Malmi airport caused by ultralight aircraft C6/2007L Helicopter accident at Pelkosenniemi C8/2007L Helicopter accident at Kangasniemi C9/2007L An unauthorized approach to an engaged runway at Pori aerodrome C3/2008L Violation of separation minimums at Helsinki-Vantaa terminal control area B4/2007R Fatal level crossing accident in Kiuruvesi, Finland B5/2007R Fatal level crossing accident in Röykkä, Nurmijärvi, Finland B6/2007R Fatal level crossing accident in Perälä, Kempele, Finland B7/2007R Fatal level crossing accident in Lahti, Finland C9/2005R Freight car derailing and member of train crew injured at Tuupovaara C3/2006R Derailment of five freight wagons between Tupovaara and Heinävaara, Finland C2/2007R Derailment of a wagon in Ylivieska C3/2007R Hazardous situation in train traffic in Tampere C4/2007R Derailment of eight freight train wagons between Saarijärvi and Äänekoski, Finland C5/2007R Derailment of a freight train locomotive in Talviainen, Finland C6/2007R Tank wagon loaded with nitric acid tipped over in Siilinjärvi, Finland C1/2008R Derailment of five shunting unit wagons in the Heikkilä railway yard in Turku, Finland C2/2008R Derailment of a wagon carrying phosphoric acid in Ykspihlaja, Kokkola, Finland C3/2008R Collision of a shunting unit and a forklift truck on the Syväsatama port track in Joensuu, Finland C4/2008R Derailment of a tank wagon during shunting work in Ykspihlaja, Kokkola, Finland B3/2004M Trawler SEA GULL, foundering in the Baltic Sea B1/2005M M/S AMORELLA, fire on car deck C2/2004M Trawler NORDSJÖ, heeling on the Northern Baltic Sea C5/2004M MS KRASNOVIDOVO, collision with pontoon bridge in Kyrönsalmi strait Savonlinna C1/2006M MS ESTRADEN and MT WOLGASTERN, collision in the Kiel-Canal C6/2006M Passenger Vessel MS NORDLANDIA, Collision with Quay in Tallinn C7/2006M Loss of M/S FINNBIRCH between Öland and Gotland C1/2007M M/S KRISTINA REGINA and barge CARRIER 5, collision in Danish territorial waters at Kadetrenden C2/2007M MS CLAUDIA, grounding off Tornio C3/2007M M/S ÅLANDSFÄRJAN, grounding off Maariehamn B3/2006Y Accident at an excavation site in Espoo B1/2007Y Fire at Pitkäniemi Hospital in Nokia