Model Course No: 3.11 Marine Accident and Incident Investigation '- Training Manual First published in 1988 by the INTERNATIONAL MARITIME ORGANIZATION Albert Embankment, London SE1 7SR Revised edition 2000 Printed by Ashford Open Learning Ltd 10 '_ ISBN 92-801-5095-2 IMO PUBLICATION Sales number TA311 E Copyright © IMO 2000 All rights reserved No part of this publication may, for sales purposes, be produced, stored in a retrieval system or transmitted in any form or by any means, electronic, electrostatic, magnetic tape, mechanical, photocopying or otherwise, without prior permission in writing from the International Maritime Organization CONTENTS Foreword v Part Course Introduction Part International Shipping - Shipping Casualties and Public Perceptions Part Investigations Part International Conventions, Requirements, etc Part Setting Up an Investigation 110 Part Interviewing 116 Part Elemel1tsof a Casualty 132 Part Human Factors 147 Part Collecting Evidence 177 Part 10 Determining Occurrence Sequence, Analysing Evidence 183 Reporting 196 Part 11 17 Foreword These notes have been compiled as a reference for this course and also as a working document for use when!§ced with the need to investigate The thrust of the course is directed towards achieving a system of investigation aimed at establishing the circumstances and causes of a marine casualty, to learn and promulgate the safety lessons The course does not deal with blame or liability and only briefly refers to more formal court or inquiry procedures The course is centred on the International Maritime Organization Assembly Resolution A.849(20), the Code for the Investigation of Marine Casualties and Incidents and amendments to the Code as annexed to Resolution A.884(21) The course also will stress the need for reporting to IMO under the provisions of MSC Circ 827/MEPC Circ 333 No two accidents are ever precisely the same However, the course sets out to show that while the causal factors may be diverse, the underlying generic structure of any casualty or accident, whether marine, road, rail or aviation, can be seen to have a common structure Much of the course is centred on role playing and case studies Participants should use their experience and judgement to make the roles as realistic as possible; by so doing participants should complete the course with a clear concept of systemic investigations and their role in improving the safety of life at sea and protecting the marine environment Course Introduction 1.1 Introduction Annexed to IMO Resolution A.847(20) are guidelines to provide flag States with a means to establish and maintain measures for the effective application and enforcement of IMO Conventions Part of the annex refers to flag State investigations "7 FLAG STATE INVESTIGATIONS 7.1 In addition to providing qualified surveyors, the flag State should provide qualified investigators Consistent with article 94.6 and articles 217.4,5 and of UNCLOS and with the provisions of the relevant IMO conventions, investigations should be carried out following a marine casualty or pollution incident The flag State should ensure that individual investigators have a working knowledge and practical experience in those subject areas pertaining to their normal duties Additionally, to assist individual investigators in performing duties outside their normal assignments, the flag State should ensure ready access to expertise in the following areas, as necessary: navigation and the Collision Regulations flag State regulations on certificates of competency causes of marine pollution interview techniques evidence gathering evaluation of the effects of the human element 7.2 Any accident involving personal injury necessitating absence from duty of three days or more and any deaths resulting from occupational accidents and casualties to ships of the flag State should be investigated, and the results of such investigations made public Ship casualties should be investigated and reported on in accordance with UNCLOS, relevant IMO Conventions, and the Guidelines currently being developed by IMO Casualty investigations should be conducted by suitably qualified investigators, competent in matters relating to the casualty.The report of the investigation should be forwarded to IMO together with the flag State's observations, in accordance with the Guidelines referred to above." It is important, therefore that flag and coastal State administrations have in place an effective and internationally acceptable system for investigating marine accidents 1.2 The purpose of the course The purpose of this course is to introduce you to the philosophy, processes and procedures required to support a marine casualty investigation in accordance with IMO Assembly Resolution A.849(20) and the Code for the Investigation of Marine Casualties and Incidents The Code seeks to introduce to the international shipping community uniform objectives and procedures for investigating casualties which occur in this most international of industries The international aviation industry has subscribed to such a code (Annex 13 of the International Civil Aviation Organization Convention) for some years and the systems approach into air accidents is credited with contributing to the safety of the civil aviation industry The IMO Code is aimed at a safety outcome to identify the circumstances under which a casualty occurred and to determine the causes of such an accident This course is designed to introduce potential flag State investigators with an introduction to accident investigations and accident investigation methodology The course will be focused on the IMO Code The course also provide administrators with an insight into what is required in conducting a casualty investigation, supporting the investigator in the field and what an investigation should achieve Investigations are an "official function of a flag or coastal State" When investigating in your own country your own national laws will apply You will have to bear these laws in mind when dealing with certain parts of the course, this will apply particularly to Part 1.3 Course structure The course is divided into eleven learning objectives or "parts", as outlined in the program It will be centred on group activities and lecture sessions Recommended reading and references Reason, J (1990) Human Error, Cambridge University Press, Cambridge (UK) ISBN 0521 314194 Reason, J (1997) Managing the Risks of Organizational Accidents, Ashgate, ISBN 84014 1050 Reason, J Corporate Culture and Safety NTSB Symposium on Corporate Culture and Transportation Safety, Washington, 1997 Moore-Ede, M., The Twenty-Four Hour Society, Random House Australia, 1993 ISBN 09 182755 Perrow, C., Normal Accidents - Living with High-Risk Technologies, Basic Books, Inc New York Elizabeth F Loftus, Eyewitness Testimony, Harvard University Press, 1996.) Rushbrook.R., Rushbrook's Fire Aboard, Third Edition, 1998, Brown, Son & Ferguson ISBN 0851746594 DeHaan J.D., Kirk's Fire Investigation, Fourth Edition, Prentice Hall ISBN 8359 5056 (NOTE THERE MAY BE A 5th ED - GET THE LATEST) Inc, 1997 O'Leary, M & Chappell, S Confidential incident reporting systems create a vital awareness of problems ICAO Journal 51 :11-13, 1996 Ainsworth et aI, (1996) Proceedings of a Research Workshop on Fatigue in the Maritime Industry, Seafarers International Research Centre for Safety and Occupational Health, University of Wales, Cardiff, Executive Summary Colquhoun, W P., Rutenfranz, J., Goethe, H., Neidhart, B., Condon, R., Plett, R & Knauth, P (1988) Work at sea: a study of sleep, and of circadian rhythms in physiological functions, in watch keepers on merchant vessels I.Watchkeeping on board ships: a methodological approach International Archives of Occupational and Environmental Health, 60: 321-329 Siros, W.G., (1995), Alertness Assurance - The key to reducing Fatigue and Human Error in the Marine Industry, American Petroleum Institute/US Coast Guard International Shipping - Shipping Casualties and Public Perceptions '- 2.1 Marine accidents as seen by the public and the media Shipping was once seen as the safest and most environmentally friendly way to transport goods by sea In any analysis it is the only practical way for the vast bulk of the world's trade Shipping accidents, particularly those that involve pollution, bring into question the safety of shipping and the quality of ships and their crews Whether or not such questions are justified, it is shipping accidents that provide an image of the industry, which attracts considerable attention (Video "Scandal at sea" or equivalent Approx 30 minutes) 2.2 Shipping safety issues What is risk? The risk is the probability consequences of event multiplied by the exposure times the ",>- Is there such a thing as an acceptable accident rate? The need is to set the risk as low as possible and at the same time make the operation practical It is possible to design all manner of engineering solutions to minimise risk The idea of "redundancy" is one that is widely used If one has three systems operating in parallel each with a possible failure rate of 1:100, the likelihood of all three failing at once is 1:1,000,000 Mechanical and electrical systems are increasingly reliable and with redundancy the chances of a casualty through equipment failure is minimal It is often stated that human error is the cause of about 80 per cent of accidents This course will examine this statement and introduce some basic consideration of human factors It will show that all accidents bar "acts of God" have some human element in the chain of events leading to the accident The aim of marine casualty investigations is to prevent similar casualties Open, systemic and fair investigations, which are not aimed at apportioning liability or blame, have proved effective in the aviation industry Such an international approach in the marine world may have helped prevented unacceptable losses at sea This course aims to introduce participants to a system of investigation, which may help to prevent future casualties Most casualties involve the smaller range of general cargo ships and fishing vessels The two types of ship that have caught the public's attention are bulk carriers and passenger ferries Bulk carrier losses The loss of these ships reflected a worldwide trend According to Lloyd's Register of Shipping, (J M Ferguson, Bulk Carriers - The Challenge, 14 May 1991) between 1980 and 1991, 64 bulk carriers were lost under circumstances where structural failure may have been a factor At the peak, 20 bulk carriers of over 15,000 gross tonnage were lost in a 20 month period, apparently as a result of some failure in their structure, between December 1990 and August 1991, with the loss of hundreds of seafarers MARINE ACCIDENT AND INCIDENT INVESTIGATION: TRAINING MANUAL Bulk carriers make up about 15 per cent of the world fleet in numbers of ships over 1000 deadweight tonnes The table below shows the number of bulk carriers (and OBOs) lost through having foundering or having been posted as missing compared to all vessels (excluding fishing vessels, tugs and dredgers) that have been reported as having foundered or as missing The column of lives lost shows all lives lost on bulk carriers1 The following table, from Lloyd's Casualty Returns and Lloyd's Annual year Books, shows that the safety of bulk carriers remains an important issue The following table is focused on bulk carriers that sank or suffered known structural damage It does not include bulk carriers (or bulk carrier deaths) from fires, explosions etc Year All losses over 500 g rt 1989 1990 1991 1992 1993 156 141 183 136 121 Bulk carriers lost through sinking 12 14 3., Lives lost through sinking Year All losses over 500 grt 66 121 153 28 74 1994 1995 1996 1997 1998 59 112 93 113 Bulk carriers lost through sinking 7** 3** Lives lost through sinking 123 3* 49 45 69 *Note: Lloyd's Casualty Returns changed the format in which they showed figures for loss of life These figures are open to revision as reports are received Anna Spiratu sank after collision with loss of 26 lives, on 15 June 1996 ICL Vikraman ** sank after collision with loss of 24 lives, on 26 Sept 1997 The loss of bulk carriers between 1987 and 1994 led to an initiative by the Secretary General at MSC63, following which a six person correspondence group of competent persons addressed the issues of bulk carrier safety This group reported to MSC, which drafted a new chapter to SOLAS 74 This Chapter was the subject of a Diplomatic conference in November 1997, which adopted the new chapter 12 to the SOlAS Convention on additional safety measures for bulk carriers Passenger ferry losses Ferries account for about 5.8 per cent of the total number of the world fleet of ships over 1000 tonnes deadweight Ferry accidents, particularly involving loss of life, are major incidents requiring the most rigorous investigation Prior to 1989 a number of ferry casualties created considerable concern These included: Dona Paz involved in collision with the tanker Vector south-east of Manila on 20/12/1990, the death toll has been put at 4,386, though the official figure is lower at about 3148 Herald of Free Enterprise capsized and foundered after leaving Zeebrugge on 3/3/97 with the loss of 193 lives Dona Marilyn capsized in a typhoon in the Philippines on 24/10/88 with the loss of as many as 350 lives 1Source Lloyd's Casualty Return INTERNATIONAL SHIPPING - SHIPPING CASUALTIES AND PUBLIC PERCEPTIONS Since 1989 the losses have continued: Salem Express on passage Jeddah-Suez struck a reef and sank on 14/12/91, 464 deaths; Jan Heweliusz - capsized in the Baltic, 14/1/93, 55 deaths; Estonia - capsized in the Baltic 28/9/94, 852 deaths; Cebu City - sank after a collision in Maila Bay on 2/12/94 with the loss of 140 lives; Gurita, stranded and sank on 19/1/96, Banda Archipelago (Indonesia) with loss of 338 lives; Bukoba capsized on Lake Victoria in May 1996 with loss of 869 lives; Princess of Orient- sank in tropical storm Vicki on 21/8/98, at least 150 deaths Harta Rimba - sank in central Indonesian Archipelago on 6/2 99 with loss of approx 300 lives The loss of the Estonia and the ferry tragedies over the previous years prompted the Secretary General and the Chairman of the Maritime Safety Committee to seek approval of the 18th Assembly for the formation of a group of experts to consider the issues of ro-ro safety This group met on occasions resulting in a Diplomatic Conference in 1995, which approved 30 amendments to SOLAS 74 dealing with ro-ro safety 2.3 Accidents as an iceberg Action ratio study" What can we learn from accident investigations and analysis? Which incidents should be investigated? Quite obviously no marine authority has the resources to investigate all minor accidents and near misses (sometimes referred to as "incidents") However, in an ideal world we would look at even the most minor incident Accidents can be seen as the part of an iceberg above the sea surface Beneath the surface there are a myriad of minor accidents and near misses which are largely unreported and whether they are investigated or not may rely on the company safety culture and ethic A study of industrial accidents undertaken by Frank E Bird, Jr and George L Germain (1969)2 analysed 1,753,498 accidents reported by 297 companies from 21 industrial groups over covering a period of billion man hours In comparing the severity of accidents they discovered that for every reported major injury (death, disability, lost time or medical treatment) there were 9.8 minor injuries requiring first aid Bird F and Germain G., (1986) Practical Loss Control, Leadership Institute Publishing, Loganville, Georgia 10 Determining Occurrence Sequence, Analysing Evidence 10.1 Introduction This part of the course is designed to assist investigators in conducting investigations in accordance with IMO Resolution A.849(20), the Code for the Investigation of Marine Casualties and Incidents, and IMO Resolution 884(21), Amendments to the Code for the Investigation of Marine Casualties and Incidents The course will provide the participants with a systematic approach to marine casualty and incident investigations At the end of the course the participants will be confident in following the systematic As stated in Section Res A 884(21) (Unit 4.6 - section 2) the essential sequence of events for investigating any accident are the following: 10.2 Principles of accident analysis The one overriding principle is that any analysis must be firmly based in the evidence gathered in the investigation This section of the course will help you arrange evidence and analyse it systematically We have already established that any accident is a sequence of events, which can be shown as a time line or series of parallel time lines The start of the time line can be problematic; the termination of the time line, at least in the first instance, is the time of the casualty itself 10.3 Building a hypothesis One of the main reasons for using investigators skilled in the maritime disciplines is that they are able to empathise and will understand how an accident may have occurred This allows an investigator to build a hypothesis as evidence is gathered Two very important points: • any hypothesis must be based on the evidence before the investigator • remember the human factor and that investigators are just as susceptible to decision making errors such as "false hypothesis" or "confirmation bias" The need to keep an open mind while exploring any hypothesis is essential There is always a temptation to favour facts that lean towards the way the investigator is thinking For this reason the ability to discuss and debate any hypothesis with peers within an administration can aid in preventing an investigator embracing a false hypothesis The techniques in this section will help in following objective procedures 10.4 Causal Factor Analysis and events and condition charting At this stage it will be useful to review the content and discussion of section of this course Professor James Reas.on in a paper to the 22nd Annual International Society of Air Safety Investigators in 1991 wrote:' "Many decades of air accident investigations have created well stocked data bases that, in history at least, should establish the relative significance of such causal factors as pilot error, mechanical failure, weather, inadequate maintenance and other personnel failures But a glance at some recent statistical analyses shows that this is far from being the case A report to the Flight Safety Foundation in 1986 claimed that mechanical failure preceded by faulty maintenance was the principle cause of air accidents (Forman, 1990) In 1987 the Chairman of the US National Transportation Safety Board (NTSB) told the press that bad weather near airports caused 64 per cent of major crashes in the preceding five years The Lufthansa World Accident Survey (1989) found that cockpit crew errors were the prime contributor, accounting for 76 per cent of all causal factors Whom should we believe? In my view, we should believe none of them All of these figures misrepresent the causal reality because they under-emphasise the stochastic, organisational and combinatorial nature of aircraft accidents Failures in maintenance, air traffic control and aircrew performances are not uncommon, but only very rarely they cause accidents Such factors are necessary but insufficient causes The sufficiency is supplied by a malign chance (sods Law) that combine with these elements in a moment of system vulnerability Neither the errors nor mechanical failures nor the weather conditions are individually important, rather it is their chance conjunction with other causal factors to breach, circumvent or remove the aircraft's defences that is of primary significance Such complex interactions are not always emphasised ·" Reason proposes that on analysis any accident can be reduced to eleven 'general failure modes'1:, these may be paraphrased as: • Deficient or inadequate plant and equipment • Poor design/design failure • Flawed maintenance policy and planning • Poor procedures • Error-enforcing conditions • Inadequate planning and management of resources • Incompatible goals • Poor communications • Deficient organization • Deficient training • Inadequate or ineffective defences Also defences are designed to serve one or more of the following seven functions2: • To create understanding and awareness of hazards • To give guidance on how to operate safely • To provide alarms and warnings in the event of imminent danger • To restore the system to a safe state in the event of a threat • To interpose safety barriers between the hazards and the operation • To contain and eliminate hazards • To provide a means of escape and rescue should all else fail A means of analysing the effectiveness or otherwise of defences is shown in the matrix below Function/ mode feature Engineered safety controls Standards, policy, supervision Procedures, instructions, drills Training, briefings, equipment Awareness Detection warning Guidance Protection Recovery Containment Escape Reason, J., Managing the Risks of Organizational ibid., pg Accidents, pg 136,Ashgate, 1997 Personal protection MARINE ACCIDENT AND INCIDENT INVESTIGATION: TRAINING MANUAL 10.5 The SHEL MODEL (Edwards 1972 as modified by Hawkins 1975) Any operational system is made up a number of major elements, one of which is the human element, and all of which interact in such a way that their total effect is larger than the sum of their parts The SHEL model helps to aid the understanding of human factors The name is taken from the initial letters of its component parts: • • • • Software documentation, procedures, symbols, etc Hardware machinery and equipment, etc Environment both internal and external to the workplace Liveware the human element Liveware In the centre, the hub, of the model is a person, the most critical as well as the most flexible component in the system However, people are performance as between individuals, or on an individual stress factors to which an individual may be subjected other components in the system to the Liveware in order lead to a failure subject to considerable variation in basis depending upon the differing It is important to try and match the to avoid stress which will eventually In order to achieve such a match the following characteristics are important: 1) Physical size and shape The design of the working environment and equipment and general ergonomic principles These principles may have to be adopted to factors such as gender, age and ethnic characteristics 2) Physical wellbeing 3) Input characteristics People need food, water, air and sleep Humans have a sensory system for collecting information by sight, hearing, touch and smell, all of which, either singularly or in combination help them to respond to external events and carry out tasks These senses are liable to degradation depending on the physical and psychological performance over time 4) Information processing Humans have severe limitations in short and long term memory and in some mental processing activities In poorly designed instrumentation these have led to ambiguity and inappropriate action and/or deductions 5) Output The appropriate physical response, once the senses system have initiated a response and the brain has processed the perceived information, relies on all of the above and the environment DETERMINING 6) OCCURRENCE SEQUENCE, ANALYSING EVIDENCE Environmental tolerance Temperature, humidity, noise, time of day, light, darkness, atmospheric pressure, aroma all reflect on performance A boring, stressful working environment can be expected to degrade human performance Liveware/Hardware This interface is the one most often considered when considering the human-machine (equipment) system Because humans are adaptable and can make allowances for less than optimal design, deficiencies in Hardware may not be identified until after a disaster This does not make the deficiencies any less real Liveware/Software This interface is between the human and system procedures (manual and check list formats, symbology and computer programs) In an accident mismatches here are often difficult to identify, but often stem from contradictory "software" or misunderstood "software" Liveware/Environment Mismatches between the human and the environment can easily be identified at sea Ships operate a 24 hour society all over the world The ship's movement, particularly in bad weather, vibration, engine noise taken with temperature, humidity, the need for the O.o.Wto operate in darkness are examples of such environmental factors Liveware/Liveware This is the interface between people Between individuals on a ship, between the bridge team and pilot, between the ship's staff and shore management, between port officials and the master It is a highly complex, but important element in all human performance 10.6 Causal Factor Analysis and events and condition charting It is always important to keep in mind that accidents occur as a chain of factors containing unsafe events or acts, which combine to cause an accident The chain often continues beyond the casualty itself into a period of degradation or recovery It is also important to remember that elements of the casualty may have been lying dormant within the system for many years, until the "malign chance" of circumstances come together When investigating a casualty or incident there are a number of techniques that may be used to chart the sequence of events to help arrive at a logical sequence of events "Fault tree analysis" is one such technique, but it does require a special understanding of the symbology used and training in the technique Simpler, less time consuming and highly practical for the investigation team is a simple chronological charting of "events and conditions" The following text has been taken (or adapted) from the United States Department of Energy publication "Accident/Incident Investigation Manual" Second Edition (DOE/SSDC 76-45/27) Accidents are investigated to identify the causes of their occurrence and to determine the actions that must be taken to prevent recurrence It is essential that the accident investigators probe deeply into events and the conditions that create accident situations, and the managerial control systems that let these events and conditions so that the root causes can be identified Ludwig Benner suggests two principles which are helpful in defining and understanding these sequences of events, conditions, Accidents are the result of a set of successive events that produce unintentional harm The accident sequence occurs during the conduct of some work activity MARINE ACCIDENT AND INCIDENT INVESTIGATION: TRAINING MANUAL The key points are that an accident involves a sequence of events that occurs in the course of a normal operation but that culminates in unintentional injury or damage Implicit here, too, is the existence of contributing causative factors, such as existing conditions, failed defences etc., as well as identifying the beginning and end of the accident sequence Staff of the National Transportation Safety Board pioneered the use of sequence diagrams or charts as analytical tools This lead to the Causal Factors (CF) diagram, which depicts in logical sequence the necessary and sufficient events and conditions for accident occurrence Accidents are rarely simple and almost never result from a single cause Rather, they are usually multifactorial and develop from clearly designed defined sequences of events which involve performance errors, changes, oversights and omissions The accident investigator (or investigating board or committee) needs to identify and document not only the events themselves, but also the relevant conditions affecting each event in the accident sequence Construction of the causal factor chart should begin as soon as the accident investigator begins to gather pertinent information The events and causal factors will usually not be discovered in the sequence in which they occurred, so the initial causal factors chart will only be a skeleton of the final product and will need to be upgraded as additional facts are gathered It should be started as early as possible as it helps to: Organize the accident data Guide the investigation Validate and confirm the true accident sequence Identify and validate factual findings, probable causes and contributing factors Simplify the organization of the investigation report Illustrate the accident sequence in the investigation report The following guidelines are suggested for construction of the chart Events should be arranged chronologically from left to right Events shown in rectangles and conditions in ovals Events are connected by solid arrows Conditions are connected to each other and events by dashed arrows Each event and condition should be enclosed in solid line (rectangles and ovals) if the evidence is factual or by dotted lines if presumptive The primary sequence of events should be depicted in a straight horizontal line (or lines in confluent or branching primary chains) Secondary event sequences, contributing factors and systemic factors should be depicted on horizontal lines at different levels above or below the primary sequence In reconstructing individual activities it is often useful to construct a line for each individual or unit (say in the case of collision) They can be integrated later Events should track in logical progression from the beginning to the end of the accident sequence The following are the criteria for Event and Condition descriptions: Each event should describe an occurrence or happening (tank lid ruptured) and not a condition (tank lid had leak in it) Each event description should be a short sentence (ship altered course to port) Events should be precisely described (mate turned valve clockwise) not (mate shut tank) Each event should be a single discrete occurrence (ship grounded) Each event should be quantified where possible (tank lost 3,000 tonnes of oil) Each event block should contain date and time if possible Each event should follow sequentially from the event and conditions preceding it Hypothetical case of dragging anchor 10.7 Six tests of safe operation Although it is not the purpose of an investigation to attribute liability or blame, judgement of individual and collective acts of commission and omission play an important part of any accident analysis Investigators are blessed with perfect hindsight, a hindsight, which must be used to make an accurate and professional analysis of the casualty to prevent such an accident occurring again It is equally important for the credibility of the investigation that the analysis (and the hindsight) should be realistic It is useful to remember the "substitution test" (see page 14) and it is also useful to apply the following six tests of safe operation: Was the casualty foreseen or foreseeable? Was the equipment in use fit for purpose? Were the systems and procedures effective to maintain safe operation? Were the staff members fit, competent and effective? Were emergency procedures and defences effective? Was there a management system to monitor and improve performance? 10.8 Analysis of physical evidence Documents Analysis of documents may involve two major aspects, cross checking documents from different sources that contain the same information or scientific analysis Analysis could include cross checking the bridge movement or "bell" book with the engine room records It cannot be emphasised enough that contemporaneous records, those made at the time, are of value, fair copies of log books, e.g the scrap log copied out in a fair hand are of limited value Of greater value is the cross checking of ship's records with external sources such as VTS tapes, harbour control tapes or log books, cargo terminal records, police records, customs records, or even TV or radio recordings Investigators must keep an open mind and think latterly asking "who else may have similar information" International Ship Management Code The ISM Code documentation should be inspected as a matter of routine It is important to ensure that the procedures in the code are adhered to The ship operator's "Documentation of Compliance" is valid for years, subject to annual verification The ship's "Safety Management Certificate" is valid for years subject to periodical verification by the administration All aspects of the code are important to an investigator and include but are not confined to the following • Training (ISM Code 6.3) • Passage planning and procedures with pilot embarked (ISM Code 6.4) • Information and language of ISM Code (ISM Code 6.6) • Plans, instructions, check lists for the safety of the ship and pollution prevention (ISM Code 7.0) • Emergency preparedness (ISM Code 8.0) • Reporting non-conforming incidents (ISM Code 9.1) • Corrective action (ISM Code 9.2) • Maintenance (ISM Code 10.1) • Critical equipment (ISM Code 10.3) • Documentation (ISM Code 11.1) • Record of internal audits (ISM Code 12.3) It is important to remember that safety investigators should review the ISM Documentation with a view to safety, not as an exercise to audit the scheme and to focus on issues not related to the casualty However, an investigator must bare in mind that any poorly maintained system may be symptomatic of a wider malaise If there was a departure from the code it is important to identify the non-conformity to establish whether the departure was consistent with reasonable decision making (see Course 1.3.4) Depending upon the incident it may also be necessary to check the ship's reporting of "non-conforming incidents"(ISM Code 9.1) and the management receipt of such records and subsequent action, which may include a record of corrective action (ISM Code 9.3) Scientific/forensic examination of documents Scientific analysis of evidence in general will be dealt with in 4.5 and the scientific analysis (if necessary) of documents will follow the same principles There are some tips for investigators, such as counting the pages of a log book, checking sequential page numbers, holding documents to the light to see if there has been any other writing or erasures Such things may be necessary if you suspect that somebody is not telling the truth Access to original documents may be difficult on both legal and practical grounds Administrations should ensure that the investigation procedures are covered by the necessary legislation and that the legislation gives the investigator access to documents and equipment On a practical level, when a ship is continuing in operation it may not be possible to keep original documents, therefore an investigator must complete his/her examination on the spot and take copies, facsimile copies where possible Photographs are important in analysis It is useful to scan photographs into a suitable computer program where they can be enhanced and small areas viewed in more details However, it is important that any changes made to the photograph is logged as it is possible to doctor photographs and it is important that the credibility of evidence is not challenged Scientific analysis It is important to provide the scientist with any material in a state that is as near as possible the same as when taken from its source (see module 2) Proper collecting bags, tins etc are important Samples of materials such as paint, oil, water, etc., may be subject to a battery of tests for forensic comparison to establish whether they came from a common source: a) Microscopic comparison (in paint for layer sequences) b) Microspectrophotometer, measuring the colour spectra of the samples c) Fourier Transform Infra Red spectroscopy Differences in absorption of infra red light gives off different wave lengths for different chemical bonds which in paint samples identifies the polymer holding the paint together d) X-ray fluorescence spectroposcopy X-ray directed at an object/sample, is absorbed and reflected at different wavelengths by different trace elements, providing a "signature" for the elements present Wires, chain or metallic samples will need to be sent to a laboratory with the appropriate skills and, where available, some national accreditation should be sought Universities may also be able to provide scientific analysis MARINE ACCIDENT AND INCIDENT INVESTIGATION: TRAINING MANUAL When testing cargo samples, the actual specification should be compared to the documentation (if any) carried on board, IMO Code requirements and industry standards Course recorders Many ships carry course recorders There is no compulsion to so and even when fitted there is no requirement for a ship to have a course recorder operating When available the course recorder provides objective data of the ship's heading at any given time It is not a record of the track made good In the absence of any black box, or GPS record or radar recording from the ship or shore, it is the most useful objective record available of the ship's head Careful scrutiny and analysis of the course recorder can help pin point the time of collision or grounding as well as the ship's heading leading up to the accident Course recorders should be checked for any time difference between the investigation's base time Many recorders are hard to synchronise accurately and alignment with the ship's head is sometimes problematic These should be checked as accurately as possible before the trace is removed There are a variety of models of course recorders Some show the course over quadrants, with a separate quadrant pen Some also include a helm indicator, which can be invaluable Other types show a 360° range across the trace paper Another problem is the "point" of the trace pen This can often be very thick There are several techniques or combinations of techniques which are of practical use An enlarged photocopy is a useful tool, though it should be realised that photocopying will often distort the page slightly, although more modern copiers may distort less The photocopy technique may not be suitable for a formal court process, but for a safety investigation it is a useful way to achieve a suitable scale for measuring and analysis Scanning the trace on a computer scanner is a very successful way of analysing course recorder traces The computer can be used to zoom in and enlarge areas of the trace Also, using Adobe Illustrator or Corel Draw or similar graphics packages, the investigator can overlay his/her own grid scale A ruler and high power magnifying glass remain reliable If using a course recorder trace the investigator must obtain the best quality trace possible Quality of trace is an important factor in lifting off information and a trace can be one of the most important sources of information In principle, analysis of the heading record is simple, but in practice it is difficult and requires the use of manual methods over a small area of interest DETERMINING OCCURRENCE SEQUENCE, ANALYSING EVIDENCE The following has been extracted from a 1982 paper by Ewan C B Corlett MA, PhD, F Eng, Hon Vice President Royal Institution of Naval Architects: By selecting the area of interest changes in heading caused by an external force may be detectable by careful analysis By selecting a fairly narrow time base, say three minutes and taking headings against time and plotting them on graph paper a number of analyses may be possible These include: • headings • angular velocity curve • angular accelerations • rate of change of heading Analyses of a trace can show the exact (recorder time) of the moment of impact If two ships are in collision, depending upon the relative sizes of the vessel, both ships will experience a change in their angular velocity by reason of the contact One may experience an acceleration force and the other an arresting force If a ship contacts a solid object it will experience an arresting force These can be quite rapid phenomena producing a sharp change in the curve If the contact is through hydrodynamic effect, say with the side of a channel, the effect may be less rapid, producing a smoother curve Assuming an accuracy in time to the nearest second from the course recorder trace, and to a tenth of a degree in heading, it is possible for purposes of illustrating a method for analysis, to draw a graph of ship's heading in degrees against time in seconds for a period from seconds to 180 seconds, as in Figure 4a, the heading record plot Examination of the figure reveals that from secs to just over 40 secs, there was no change of heading and the course was steady on degrees, or 003 degrees Thereafter the course changed to 007 degrees, with a relatively slow change of heading between 42 seconds till around 74 seconds At that point, the heading changed rapidly by about degrees in just over a second, after which there was a reduction in the rate of change of heading till the course was 007 There was no change of heading for the short period that the ship was steady on 007 Then, the heading changed back to 003, the change of heading taking place slowly at first, reaching a maximum rate between 100 seconds to around 102 seconds, reducing thereafter to zero from about 125 seconds onward Using the heading record plot and the slope at various points of the curve, we can obtain and plot angular velocities against time For the first 42 seconds, the angular velocity is zero, increasing gradually till at around 75 seconds the angular velocity increases to about 1.2 degrees per second, corresponding to the near vertical trace on the heading plot at the same time The angular velocity then reduces gradually in line with reducing rates of change of heading becoming zero just after 85 seconds After this point the angular velocity increases as the ship's head is altered to 003, with a maximum angular velocity being obtained soon after 100 seconds Once the ship's head is steady on 003 or thereabouts the angular velocity reduces to zero once more MARINE ACCIDENT AND INCIDENT INVESTIGATION: TRAINING MANUAL Similarly, we can plot a curve of angular acceleration against time, so we now have three curves, the heading record plot, and its first and second derivatives, angular velocities and accelerations The angular acceleration curves reveal high acceleration and deceleration, indicating the possibility of a collision, and subsequent water cushioning effect (Part of a paper presented at the IMO/China Seminar, 1982) Other records There is no standardisation of ship recording equipment Investigators should always establish what is available on any specific ship Data loggers are becoming more common DETERMINING OCCURRENCE SEQUENCE, ANALYSING EVIDENCE Voyage Data Recorders (VDR) the equivalent to the aircraft "black box" are carried on certain ships, "reading" the information on the VDR may require cooperation with other flag States Some pieces of ship's equipment allow information to be recovered Some fire alarms have a memory and record the times of alarms and the location of the sensor In ECDIS Units it is a design requirement that tracks should be able to be recovered and replayed Other than ECDIS there are not any standard requirements but investigators should be alert to the possibility of being able to down-load GPS data, or recovering radar displays information on a video drive facility from certain sets 10.9 Analysis of interviews This is a process of assessing those interviews or parts of the interview process where the same or similar information was provided and those areas where contradictory evidence was given It is important that all information is checked against the real and documentary evidence It can be the case that consistent, or similar accounts can be wrong and one single contradictory account correct Where ever possible all accounts should be checked against all other evidence and corroborating real or independent evidence normally preferred over evidence from the same interest group 10.10 Analysing human involvement Comparing the events and condition chart(s) with the witness interviews should help in analysing the human involvement The human involvement will often (more often than not) extend beyond the ship Remember the Reason and SHEL models 10.11 Identifying potential safety problems and safety actions Careful analysis, using event and condition charting can identify the relevant safety problems The developing of safety actions fall into three basic categories: Immediate (local) action Flag/coastal State (unilateral) action International (IMO) action It is important to ensure that any recommended safety action is practical and will address any safety deficiency without causing others (see notes under 4.3) Safety action may involve a local ship board or port solution Remember here the ISM Code procedures and any ISO 9002 procedures Safety actions may require an amendment to national procedures or requirements or may be such that an international response may be required through IMO 10.12 Revisiting witnesses Although it may not always be possible, investigators should never rule out revisiting or otherwise contacting witnesses to try and resolve ambiguities Even the most experienced investigator will find it very difficult to cover all aspects of all factors in a single interview Remember, if procedures not rely on sworn testimony, most people can be reached by fax, phone, letter or e-mail 11 Reporting 11.1 Natural justice - the circulation of the draft report Natural justice, being fair to all parties is an essential element in establishing a system that has the acceptance of the industry Investigation procedures should allow for the draft of a report, or relevant part of the report, to be circulated to any individual or organisation that may be affected by the report's conclusions Any such person or organisation should be given a reasonable period of time to provide other evidence or information or make a submission relating to the report The investigator must then take such submissions into account and, if necessary, amend the report or in some way fairly reflect the views of any person making a submission 11.2 IMO reporting requirements Do not forget the requirements of MSC/Circ 827 (MEPC/Circ 333) "Harmonised Reporting procedures - Reports required under SOlAS regulations 1/21 and MARPOl 73/78 articles and 12" 11.3 Report format Reports are compiled from the evidence gained, which allows us to tell the story, analyse the factors and draw sensible conclusions based on the human factor models The Code at section 14 provides guidance on the content of the report Reports should include, wherever possible: a summary outlining the basic facts of the casualty and stating whether any deaths, injuries or pollution occurred as a result; the identity of the flag State, owners, managers, company and classification society; details of the dimensions and engines of any ship involved, together with a description of the crew, work routine and other relevant matters, such as time served on the ship; a narrative detailing the circumstances of the casualty; analysis and comment which should enable the report to reach logical conclusions, or findings, establishing all the factors that contributed to the casualty; a section, or sections, analysing and commenting on the causal elements, including both mechanical and human factors, meeting the requirements of the IMO casualty data base; and where appropriate, recommendations with a view to preventing similar casualties 11.4 Narrative The narrative should describe the circumstances and events leading to the investigation in a neutral way, without making judgements A simple factual story 11.5 Analysis The relevant causal factors should be reviewed and analysed Any causal factor analysed should be identifiable in the narrative section The chart of events and conditions is a useful guide Nor should the six questions of safety operation be ignored It helps to keep the REPORTING analysis focused and in the realms of practicality.There is nothing wrong in the investigator providing expert opinion, providing that: • it is clear that he/she is giving an opinion; • it is very carefully thought through; and wherever possible • it is canvassed with other suitably qualified persons (there is nothing worse that an unrealistic or unsubstantiated opinion) 11.6 Conclusions The conclusions must be drawn from the analysis section They are often the part of a report read initially by interested parties and other professionals Care must be exercised to ensure they are accurate Remember, after sending out the draft report your conclusions will be difficult to amend in terms of being more severe 11.7 Recommendations Recommendations must follow logically from the text of the report Again wherever possible the recommendations should be tested with suitably qualified colleagues and all aspects of their impact thought through Consider the human factors involved A common response is to legislate Legislation to regulate human behaviour, in itself, is seldom the answer if the situational factors are ignored 11.8 The completed report and distribution The Code at section 12 deals with the issuing of casualty and incident reports and submissions to IMO and makes provision for a dissenting report to be issued where a "substantially interested State" disagrees with the whole or part of report How widely the report is distributed is a matter for your administration However, unless people read about accidents and are presented with accurate facts and expert analysis lessons will not be learnt and the nature of accidents will not be understood 11.9 Media Section 5.15 offered advice on contact with the media When releasing a report public attention can be drawn to the publication by media release Depending upon the trustworthiness of the individual media outlets and the importance of the incident it may be possible to provide a report under embargo so that a journalist can read the report and provide a full and comprehensive report But never forget, the creed of the majority of journalists is to get a story that will sell newspapers, truth and accuracy are often, but not always, the first casualties 11.10 Reopening an inquiry In addition to introducing a more uniform approach to marine casualty and incident investigations and promoting cooperation between substantially interested States, the stated aim of the Code is to prevent similar casualties in the future This can only be achieved by undertaking high quality investigations and reporting based on the best evidence available If after completing a casualty or incident investigation (including reporting) significant new evidence is provided to the State(s) involved, provision should be made to reopen the inquiry There are a number of precedents for this, perhaps the most famous include the Titanic and the Derbyshire ... not normally be covered by normal marine casualty or incident investigation procedures 1Adapted from Lloyds World Casualty Statistics MARINE ACCIDENT AND INCIDENT INVESTIGATION: TRAINING MANUAL... ro-ro safety 2.3 Accidents as an iceberg Action ratio study" What can we learn from accident investigations and analysis? Which incidents should be investigated? Quite obviously no marine authority... determine the causes of such an accident This course is designed to introduce potential flag State investigators with an introduction to accident investigations and accident investigation methodology