Learning from experience - adopting a systems approach to analysis of marine incidents
V Pomeroy, from Lloyd's Register, and B Sherwood Jones, from Process Contracting Ltd, comment on the evaluation of evidence from marine incidents in such a way that the maximum volume of useful information is extracted to support the development of effective risk control measures through regulation and standards.*
T
HE authors recognise that ships and marine systems are becoming more complex and integrated, and that a 'system' includes its operators. Improvements in operational safety can be achieved by dealing with the 'relevant systems' that support operation. This view requires a different way of abstracting information from incident data. The methodology presented here provides a structured 'systems approach' which leads to the clear identification of the initiating points where a corrective action either in terms of regulation or operational procedure could reduce the associated risk, effectively and specifically.
In presenting some re-assessment of the information that has been presented in publicly- available formal and informal incident reports, there is no criticism intended of the investigators or their conclusions. However, by using the same base data, the authors demonstrate that other valuable information can be elicited, which could be important in preventing future different incidents.
The history of industrialised society is littered with examples where early warnings of hazards were ignored until sufficient hard evidence had been accumulated, often a very long time later. Some good illustrative examples are described by the European Environment Agency (EEA) (Ref 1) in a report which is based around the following four questions:
• when was the first credible scientific 'early warning' of potential harm?
• when and what were the main actions or inactions on risk reduction taken by regulatory authorities and others?
• what were the resulting costs and benefits of the actions and inactions, including their distribution between groups and across time?
• what lessons can be drawn that may help future decision-making?
The concept promoted by the EEA authors is one of precaution, and the evidence supports the argument that industrial society is not good at taking cognisance of early warnings, foreseeing future impact, and forestalling the inevitable consequences.
* Extracts from the paper 'Learning from experience - adopting a systems approach to analysis of marine incidents', presented at The Royal Institution of Naval Architects' international conference Learning from Marine Incidents 3, held in London on January 25-26, 2006.
60
Given the application of the 'precautionary principle' in marine legislation, the discussion by the EEA authors on the need to address risk, ignorance and uncertainty is relevant to the maritime community, as is the discussion on the 'level of proof' required to justify action.
Effectiveness and access to data It is often assumed that most lessons from incidents will necessarily come from major events, which is generally assumed to be from those incidents that demand formal investigation and reporting. These will usually involve loss of life or pollution, or the recognition that avoidance of either of these outcomes was fortuitous. Many similar incidents may not be reported in detail because the consequences were simply less dramatic. For instance, a loss of propulsion power at sea may require towing to a safe haven or repair at sea but a similar event close to shore or in the confines of a port can have far greater impact.
If lessons learned from marine incidents are to be of the highest value in terms of prevention, then it is imperative that information is gleaned from minor incidents as well as the major cases. This means access to formal investigations and informal, usually unpublished, work. In many cases, minor incidents are simply not recognised as significant, and any corrective action is dealt with by the operator and, maybe, the equipment supplier without any expectation of providing a learning opportunity for the marine community. To develop the capacity of the marine industry to learn from incidents, consideration has to be given to formal investigations by flag administrations and others, damage investigations supported by scientific and engineering analysis, and the collected wisdom of ships' staff, owners' superintendents, equipment suppliers, and surveyors. Careful and thorough data collection is essential for the effective analysis of incidents. This can be time-consuming and is dependent on eliciting basic facts from all individuals who might have relevant knowledge. Data is also extracted from recording systems, written records and from technical investigations. For well over 50 years, Lloyd's Register has benefited from the systematic investigations of failures which has been carried out by its own expert investigation team, known throughout the marine industry variously as the Engineering Investigation Department, Technical
Investigations
Department and now the Technical Investigations element of Consultancy Services.
The reports of investigations carried out for marine clients are based on an assessment of available data and, usually, field measurements and supporting engineering analysis. In most investigations of incidents this level of investigative rigour is not found, with greater reliance on the records from operations and the recall of those involved.
The most accurate data is collected shortly after an incident, which requires access to people and the ship so that the investigators can proceed
before data is lost or recollections become more distant. Investigators are faced with a number of inhibiting pressures which constrain the effectiveness of the all-important data collection stage. Incidents usually involve insurance claims and there is increasing likelihood of litigation, and, of course, individuals involved will face the prospect of disciplinary processes and adverse consequences on their livelihood.
Investigations may be necessary to support insurance claims or to satisfy legal or political demands. However, these purposes may, themselves, restrict the learning opportunities and reduce the overall effectiveness of incident analysis in terms of preventing the occurrence of similar incidents.
In some industries, the adoption of a 'blame free' and often confidential reporting scheme has resulted in the systematic collection of incident information, although this necessarily suffers from a lack of substantive analysis of the facts. Some efforts have been made to introduce similar arrangements into the marine industry and these could be used to provide better early warning data, since individuals can report events which could have resulted in an incident but where circumstances precluded the full event development.
Reconstruction and analysis Without doubt, the analysis of marine incidents provides an essential source of information to the regulators and operators. It follows that the incident reports must be credible, presenting well-analysed conclusions and recommendations. The scope of analysis and reporting is typically based around a few 'causes' and 'some contributing factors' - the 'causal field' is fairly narrowly drawn. There are some theoretical concerns with this, discussed later. In practical terms however, it is unlikely that the maximum value can be extracted from an incident investigation if the conclusions and recommendations are specific to that incident and do not draw wider implications. The wider value may not sit comfortably in an incident report, but the learning value would suggest that it is incumbent on the investigators to disseminate their findings, including facets which might not have been significant for the incident under investigation but nevertheless indicate the need for some form of corrective action. The phrase 'learning from incidents' may be needed to complement 'incident analysis' and is used here. 'Hindsight bias remains the primary obstacle to accident investigation, especially when expert human performance is involved' (Cook, Ref 2). Dekker (Ref 3) has highlighted the need to reconstruct people's unfolding mindset as central to the analysis process. For a valid understanding of how an incident came about, it is necessary to apply the approach to latent errors at the 'blunt end' as well as active errors at the 'sharp end'. It would normally be expected for human and technical factors to occur in concert. Whilst assigning causes to factors such as these may aid understanding, they may actually impede
THE NAVALARCHITECT FEBRUARY 2006
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100