How the Human Factor Theory Correlates with the Piper Alpha Disaster - Term Paper Example

 Contents Abstract 3 Introduction 4 Role of Human Factor in organisational safety management system 4 Overview of accident causation methods 5 Piper Alpha organisational accident 7 Description of event 7 ECFS (Events and causal factors charting) analysis 7 Environmental Swiss cheese model 11 Findings 12 Industry practices and regulations 13 Limitations of the research 15 Conclusion 16 References 16 Appendix A. Types of human errors 19 Appendix B. Levels of causation 20 Appendix C. Accident Causation Methods 21 Abstract This report gives an overview of the investigative project aimed to explore how the Human Factor theory correlates with the Piper Alpha Disaster. Prior to the analysis of the accident data a brief summary of existing accident causation methods is presented, including a brief description of major elements of the Human Factor theory. Further, the Piper Alpha Disaster accident is investigated so that: (1) to determine a sequence of events; (2) to identify various causes that entailed accidental consequences (facts); (3) to identify accident causation methods, to which accidental facts are correlated; (4) to analyse accident by using chosen accident causation methods. Each of accident methods can be applied for different purposes, so to provide a comprehensive investigation of the Piper Alpha Disaster two methods - ECFE (Events and causal factors charting) and Swiss Cheese model have been explored for this accident. The report also presents a summary of corrective actions, industry practices and safety policy regulations, which have been taken after the Piper Alpha Disaster, and which currently make a significant contribution to industrial safety in UK and worldwide. Introduction Role of Human Factor in organisational safety management system During the last two decades organisational safety management has experienced an important shift - from procedures that focus only on physical environment and equipment to more complex system that includes two other aspects – systems and people (see Fig.1). Today’s researchers and practitioners view human factors as an integral part of a good safety management system and examined them like any other risk control system. Figure 1. Complex organisational safety system. Source: Health and Safety Executive, 2008, p.7 Health and Safety Executive (2005) refers human factors to “environmental, organisational and job factors, and human and individual characteristics, which influence behaviour at work in a way which can affect health and safety” (p.1). In relation to safety management human factors reveal as human failures (unsafe acts) that may cause serious accidents. Health and Safety Executive (2005) distinguishes three types of human failures (p.2): Unintentional errors: Errors (slips / lapses) – actions that were not as planned, occurring during a familiar task; Mistakes – errors of judgement or decision-making, where people do the wrong thing believing it ti be right. Intentional errors: Violations – intentional, usually well-meaning failures (e.g. deliberate deviations from the rules), but sometimes wilful (e.g. sabotage). A comprehensive taxonomy of human failures is represented in Appendix A. In order to manage human failures effectively, organisations should take measures on their detection, analysis and prevention. Using accident causation methods in the safety management practices is one of the important steps to design of effective organisational safe system. Overview of accident causation methods Kjellen (2000) defines an accident as “a sequence of logically and chronologically related deviating events involving an incident that results in injury to personnel or damage to the environment or material assets (cited in Sklet, 2002, p.9). Sklet (2002) introduces organisational accidents as “comparatively rare, but often catastrophic, events that occur within complex, modern technologies such as nuclear power plants, commercial aviation, petrochemical industry, etc. Organisational accidents have multiple causes involving many people operating at different levels of their respective companies.” (p.6). Kjellen (2000) asserts that “each accident model has its own characteristics as to the types of “causal factors” that it highlights” (p.31). So, before reviewing the accident causation methods, it is useful to understand a different nature of causal factors. DOE (1999) describes three types of causal factors (cited in Sklet, 2002, p.21): Direct cause - an immediate event or condition that caused the accident; Contributing cause - an event or condition that together with other causes increase the likelihood of an accident but which individually did not cause the accident; Root causes - the causal factor(s) that, if corrected, would prevent recurrence of the accident. Accident causation methods are divided onto three major groups (Qureshi, 2007, p.1): Sequential (or event-based or root causes) accident models that work for losses caused by failures of physical components or human errors in relatively simple systems. The most famous of them are Heinrich’ Domino theory (see Appendix C, Fig. C1), ILCI model, Events and Causal Factor Charting (ECFC) (see Appendix C, Fig. C2), Fault Tree Analysis, and many others. Epidemiological accident models that aim to explain accident causation in complex systems, they view events causing the accidents as analogous to the spreading of a disease. The most famous model among epidemiological is the Reason’s Swiss cheese model of defences (see Appendix C, Fig. C3); it has made an important contribution to understanding of accidents, revealing relationship between latent and immediate causes of accidents. Systemic accident models are based on systems theory; they describe the characteristic performance on the level of the system as a whole. The most notable of them are Rasmussen’s hierarchical socio-technical framework (see Appendix C, Fig. C4), Leveson’s STAMP (Systems-Theoretic Accident Model and Processes) model. Nowadays researchers number more 30 different causation methods, including those widely used in other management areas (e.g. Ishikawa Fishbone model in Project Management). Piper Alpha organisational accident Description of event “On 6 July 1988 the fixed steel jacket oil production platform Piper Alpha, located in the UK Sector of the North Sea, was destroyed completely by an explosion and fire. It was initiated by a condensate leak in the gas compression module, which exploded, causing a fire that burned through the gas risers, enveloping the platform in a jet fire. The survivors jumped into the sea. 165 of the 226 people on board died together with 2 rescuers” (Smith, 1995. p. 513). The major severe consequences of accident: 167 deaths and loss of about £1.7 billion. The largest and oldest oil platform in the North Sea has ended its existence. Besides about 670 tonnes of oil were spilled, a slick was 3.6 km long and 100 m wide. ECFS (Events and causal factors charting) analysis It is a simple graphical chronology of events that is used for organising events so that to portray a whole picture of the accident. Sklet (2002) considers: “Events and causal factors charting is useful in identifying multiple causes and graphically depicting the triggering conditions and events necessary and sufficient for an accident to occur” (p.31). General model of the Events and Causal Factor Charting (ECFC) you can see in Appendix C, Fig. C2. Timeline of the accident is presented in the Table 1. Table 1. Timeline of the Piper Alpha Disaster1 Time Event 12:00 a.m. Pressure safety valve (PSV) on Pump A was removed for routine maintenance. Open condensate pipe was closed by a blind flange that was not fully tightened. 18:00 p.m. Shift changed, but the operators were unaware about the PCV, because permit that was written by the on-duty engineer in the morning was not found. 21:45 p.m. Process upset occurred - Pump B stopped. Operator had to act quickly; otherwise the power supply could be broken, but he did not know what to do exactly. 21:52 p.m. Operator found in the documentation that he can start Pump A, not knowing about its actual condition. Blind flange was not visible, because it was locate above the Pump. 21:55 p.m. Operator switched Pump A on. Under overpressure gas flowed into the pump, leaking through the blind flange. Leak was noted both people (by sound) and system (gas alarms were activated). Then gas ignited and exploded, destroying the firewall. 22:04 p.m. The organisation of evacuation was disintegrated, because control room was abandoned, and no communications were between rooms. Fire and smoke were huge, but Claymore and Tartan continued pumping until , because managers were not able to make quick decision. 22:33 p.m. Final radio message. So, following a guidance of Sklet (2002, p.33), we’ve encompasses the main events of the Piper Alpha Disaster as a primary events sequence. As a secondary event sequence we’ve encompasses the event that were contributing events and those that form the secondary line of the chart. Our purpose was to represent the order of only main events of the accident, which we include in the analysis (see Fig. 2). Figure 2. Events and causal factors charting analysis Environmental Swiss cheese model The model was coined by Reason, who “emphasises the concept of organisational safety and how defences (protection barriers such as physical, human and procedures) may fail. In this approach …latent conditions (arising from management decision practices or cultural influences) combine adversely with local triggering events (weather, location, etc.) and which active failures (errors and / or procedural violation) committed by individuals or teams at the sharp end of an organisation, to produce the accident” (Qureshi, 2007, p.3). See the general model of Swiss cheese methods, which we used in the research, in Appendix C, Fig. C3. There are four defence layers (slices of cheese) in the model, for each of them we define a number of barriers, which were absent or non-working at the Piper Alpha platform (see Table 2). Lack of a barrier means “a hole” in a slice of cheese. We assume that lack of these barriers at the Piper Alpha platform at a whole is the major latent cause that has led to the terrible tragedy. Table 2. Swiss cheese analysis Design safety features Engineering, safeguard, interlocks, alarms Safety management, systems, procedures Human factors, competences Safe platform design Reliable fire walls Protective clothing Fire protection systems Escape systems Reliable communications Safety and warning devices Reserve and backup equipment Regular reliable audit The proper permit to work system Effective decision-making system Well-documented emergency shutdown procedures Regular emergency training Regular risk and hazard analysis Proper allocation of resources (physical and human) Competency assurance system (incl. managers) Adherence to procedures Decision-making style Leadership and responsibility Supervision and line management oversight Knowledge management and learning lessons Regular planning and scheduling of work activity Findings Anderson (2004) says that “historically, many accidents in major hazard industries were considered as the “fault” of someone at the sharp end – the last person who touched the equipment” (p.2). After the Piper Alpha Disaster, Lord Cullen did not blame any person, possibly, at the first time. He made a comprehensive analysis of gathered investigation data and announced 106 recommendations for improving control of major hazards offshore. We consider, however, that in this particular case, it is possible to find a party in fault; and this “fault” party is management of the organisation. All mentioned-above barriers are connected with particular activity or inactivity of managers of all levels. The consequence of terrible events can be broken if the company would apply necessary procedures and adhere to right policies. We completely agree with Collins and Keeley (2003) arguing that “although the immediate causes of major incidents frequently involve “human error” of operators or maintenance personnel, the reasons that these errors occurred in the first place were the responsibility of those more senior in the organisation” (cited in Anderson, 2004, p.2). In the research we used one of sequential or cause-related methods and one of the environmental ones. We see that facts of the Piper Alpha Disaster are perfectly correlated with various accident causation methods, because there are: Clear consequence of events in timeline; Close cause-effect relationships between events; Both immediate and latent causes for investigation. We consider it would be useful to analyse the Piper Alpha accident by one of the systemic analysis method also, but currently we don’t possess information enough for comprehensive STAMP or AcciMap analysis. In order to be able to derive more important outcomes from the systemic analyses, it is necessary to know policies and practices of the organisation at a high-level. While we don't know responsibilities and permissions of managers on all levels - from linear to general ones, we unlikely can get results more interesting than what we have today. Industry practices and regulations As we said above the first official response of the UK government was provided by Lord Cullen, who announced 106 recommendations regarding offshore’ safety management. The next big step was undertaken by Government in establishing of a special The Health and Safety Executive Offshore Division (HSE). Recently the Division celebrated its 20th anniversary, this is an organisation, primarly responsible for “the regulation of the offshore industry” (Burgess, 2008, p.1). The great progress is observed, according to Anderson (2004) in standardisation and legislations of big accidents in the oil, gas and chemical industries (as in a case of the Piper Alpha), where the new standard COMAH is actively used today. Major hazard performance indicators are defined in the standard to measure controls of risk management, including indicators related to such issues as (Anderson, 2004, p.5): Risk assessments and improvements; Plant and process change; Functioning of safety critical equipment; Competency and training; Resources (financial, staff, equipment); Procedures; Plant inspection and maintenance; Releases and near misses; Monitoring, audit and review. But not only government were concerned about policies and active steps toward safety, after the Piper Alpha disaster several important strategic alliances were emerged, e.g. UK and Norwegian regulators (UKHSE and NPD), as well as a number of international research projects were organised, e.g. Project on Fire and Explosion. Bull (2004) informs that: “The aim of the project, which ran from May 1990 to Feb 1991, was to identify the best available knowledge worldwide and to identify significant gaps in that knowledge base leading to uncertainty in design, as well as to devise future programmes aimed at reducing the uncertainty level” (p.4). There were also taken into account such society aspects as: Environmental concerns, where demands for removal of redundant installations, has been satisfied through development of strategies to cope with new risks (Burges, 2008); Health and society in the offshore industry, where governmental and non-governmental structures unit together to rise standards of health and safety in the offshore industry (Health and Safety Executive, 2008; Burges, 2008). Safety critical or Structural elements and systems, “which are used for design and provision of control and barriers for explosions” (Birkinshaw, 2008, p.4). Pertinent technical issues (Birkinshaw, 2008, p.6): robustness criteria for setting integrity levels review of current and evolving guidance overview of  methods of assessment structural components and/or system strength key issues related to FPSO's retrofit of topsides for explosion risk reduction Limitations of the research In evaluation of limitations of our research we, first of all, agree with Grabowski et al (2009) who suppose that: “In complex systems, event analyses are constrained by the quality of the data gathered, the maturity of the associated reporting system, and the training and background of the investigator and reporter. Such constraints place limits on the adequacy and strength of analyses conducted with the data.”(p.1185). We acknowledge that may have weaknesses in one or more listed above aspects. Our second concern of research limitation is connected with chosen causation methods, in particular, with epidemiological Swiss cheese model. It gives a static “frozen” view of the organisation, whilst a real socio-technical organisational system is dynamic, so barriers (or holes in the cheese) are constantly moving. Conclusion In our project we investigated the Piper Alpha Disaster accident and explored how the Human Factor theory correlates with the Piper Alpha Disaster. We’ve made certain that sequential and epidemiological models help significantly in understanding of the accidents. But they are not able to deal with complexity and dynamics of complex systems. In the case of complex, multifaceted accident, systemic analysis models are needed. References Anderson, M. (2004) Behavioural Safety and Major Accident Hazards: Magic Bullet or Shot in the Dark? Hazards XVIII Symposium Proceedings, November, UMIST, Manchester. Health and Safety Executives. [Online] Available from: http://www.hse.gov.uk/humanfactors/topics/magicbullet.pdf [Accessed 26April 2010] Birkinshaw, M. (2009) Technical policy relating to structural behaviour under explosion hazards. Health and Safety Executive. [Online] Available from: http://www.hse.gov.uk/offshore/explosionhazards.htm [Accessed 26April 2010] Bull, D. (2004) A critical review of post Piper-Alpha developments in explosion science for the Offshore Industry. Research Report 89. Norwich, HSE Books. Burgess, P. (2008) Health and Safety Executive statement on 20th anniversary of Piper Alpha. Health and Safety Executive. [Online] Available from: http://www.hse.gov.uk/press/2008/e08035.htm [Accessed 26April 2010] Collins, A. & Keeley, D. (2003) Analysis of onshore dangerous occurrence and injury data leading to a loss of containment. HSL seminar paper, May 2003. DOE (1999). Conducting Accident Investigations DOE Workbook, Revision 2. Washington, U.S. Department of Energy. Grabowski, M., Zhou, Z.Y.Z., Song, H., Steward, M., and Steward, B. (2009) Human and organizational error data challenges in complex, large-scale systems. Safety Science, 47 (8), 1185-1194. Health and Safety Executive (2005) Human factors in the management of major accident hazards. Health and Safety Executive. [Online] Available from: www.hse.gov.uk/humanfactors/topics/toolkitintro.pdf [Accessed 26April 2010] Health and Safety Executive (2008) Play your part! How offshore workers can help improve health and safety. Suffolk, HSE Books. [Online] Available from: http://www.hse.gov.uk/pubns/indg421.pdf [Accessed 26April 2010] Kjellen, U. (2000) Prevention of Accidents Through Experience Feedback. London, Taylor & Francis. Livingston, A. D., Jackson, G. and Priestley, K. (2001) Root causes analysis: Literature review. Contract Research Report 325/2001. Norwich, HSE Books. Qureshi, Z.H. (2007) A Review of Accident Modelling Approaches for Complex Socio-Technical Systems. In: Cant, T. (Ed.) 12th Australian Workshop on Safety Related Programmable Systems (SCS’07), Conferences in Research and Practice in Information Technology, Vol. 86. Adelaide, Australian Computer Society. Rasmussen J. (1997) Risk Management in a Dynamic Society: A Modelling Problem. Safety Science, 27 (2/3), 183 – 213. Sklet, S. (2002) Methods for accident investigation. Report No. ROSS (NTNU) 200208, Norwegian University of Science and Technology. [Online] Available from: www.ntnu.no/ross/reports/accident.pdf [Accessed 26April 2010] Smith, E. J. (1995) Risk management in the North Sea offshore industry: History, status and challenges. Acta Astronautica, 37, 513-523. Appendix A. Types of human errors Source: Health and Safety Executive, 2005, p.3 Appendix B. Levels of causation Source: Livingston et al., 2001, p.2. Appendix C. Accident Causation Methods Figure C1. Domino model (Source: Qureshi, 2007, p.2) Figure C2. Events and causal factors charting (Source: Sklet, 2002, p.33) Figure C3. Swiss cheese model. (Source: Health and Safety Executive, 2008, p.6) Figure C4. Rasmussen’s hierarchical socio-technical framework. (Source: Qureshi, 2007, p.5) Read More