Risk Assessment: The Human Dimension - Softcover

Hurst, N

Softcover

ISBN 10: 0854045546 ISBN 13: 9780854045549

Verlag: Royal Society of Chemistry, 1998

Inhaltsangabe

Risk Assessment: The Human Dimension begins by looking at quantified risk assessment and considers, by using case studies, how accident causation can be considered from the three main perspectives of hardware failures, human error and failures of systems and cultures.

Über diesen Titel

Über die Autorin bzw. den Autor

Charlotte y Peter Fiell son dos autoridades en historia, teoría y crítica del diseño y han escrito más de sesenta libros sobre la materia, muchos de los cuales se han convertido en éxitos de ventas. También han impartido conferencias y cursos como profesores invitados, han comisariado exposiciones y asesorado a fabricantes, museos, salas de subastas y grandes coleccionistas privados de todo el mundo. Los Fiell han escrito numerosos libros para TASCHEN, entre los que se incluyen 1000 Chairs, Diseño del siglo XX, El diseño industrial de la A a la Z, Scandinavian Design y Diseño del siglo XXI.

Von der hinteren Coverseite

Risk Assessment: The Human Dimension begins by looking at quantified risk assessment and considers, by using case studies, how accident causation can be considered from the three main perspectives of hardware failures, human error and failures of systems and cultures. The book then goes on to place risk assessment firmly within the broader context of the current, controversial debate concerning risk issues and the nature of risk. It addresses these issues mainly from the perspective of the chemical and process industries by looking at the process of risk assessment, its strengths and weaknesses and attempts to reconcile the human dimensions of risk assessment with the need for science and objectivity in risk-based decision making. Designed to be accessible to a wide range of disciplines, and enjoyable to the reader, Risk Assessment: The Human Dimension is broadly based and rooted in the author's practical experience of both risk assessment and organizations and how they function. With diagrams, summary and discussion sections in each chapter, this book will prove invaluable for the insights given in this increasingly important area.

Auszug. © Genehmigter Nachdruck. Alle Rechte vorbehalten.

Risk Assessment

The Human Dimension

By Nick W. Hurst

The Royal Society of Chemistry

Copyright © 1998 The Royal Society of Chemistry
All rights reserved.

ISBN: 978-0-85404-554-9

Contents

Author's Notes, vii, 
Preface, viii, 
Glossary, xii, 
An Overview of the Book, xiv, 
1 Different Perspectives on Accident Causation: Some Accident Case Studies, 1, 
1.1 Hardware (Case Studies 1–7), 2, 
1.2 People (Case Studies 8–9), 7, 
1.3 Systems and Cultures (Case Studies 10-12), 9, 
1.4 Summary and Discussion, 10, 
2 Models of Accident Causation and Theoretical Approaches, 14, 
2.1 Some Theoretical Approaches, 14, 
2.2 An Analysis of Different Approaches, 32, 
2.3 Summary and Discussion, 37, 
3 The Assessment of Risk – Quantification, 42, 
3.1 Engineering Approaches to Risk Assessment, 43, 
3.2 Human Reliability Assessment, 48, 
3.3 Safety Management Standards and Quantified Risk Assessment, 51, 
3.4 Safety Culture and Quantified Risk Assessment, 56, 
3.5 Summary and Discussion, 57, 
4 Risk and Decision Making, 65, 
4.1 Risk-based Decisions, 66, 
4.2 Measuring Risk Performance Between Sites – Issues of Completeness, 73, 
4.3 Summary and Discussion, 77, 
5 Discussion and Conclusions – Where Does All This Leave Risk Assessment?, 80, 
5.1 Conclusions from the Previous Chapters, 80, 
5.2 Risk Assessment – The Human Dimension, 83, 
5.3 The Subjective/Objective Debate, 84, 
5.4 Implications for Risk Estimation, 87, 
5.5 Implications for Risk Evaluation, 91, 
5.6 Final Discussion, 94, 
Subject Index, 100, 

CHAPTER 1

Different Perspectives on Accident Causation: Some Accident Case Studies

Table 1 lists some well known case studies. These are events or accidents which have been in the public eye at some time and have been the subject of investigations, enquiries, newspaper reports and so on. They are drawn from various areas: aircrashes, failures of chemical plant, a football stadium disaster and others. Generally, there are detailed published accounts and analyses of these events which describe how they happened. These are listed in the references. Table 1 also lists some different perspectives that have been taken, at various times, by others in considering the causes of the events. These are illustrated in Figure 1.1. My purpose here is to look briefly at these perspectives for the events described. At this stage I am not trying to show that some may be right and others wrong, but simply to show that different individuals and organisations tend to take different views. This raises the question as to whether a search for more complete descriptions might not suggest practical consequences for risks and safety.

Although the case studies are drawn from various areas, I am assuming that the lessons which can be learnt are independent of the area and can be applied to other areas, i.e. I assume that the underlying causes of accidents, events and incidents will cover similar, general types in the various areas considered. Consequently I have not chosen chemical plant incidents, exclusively, but events or incidents which I consider best illustrate the different perspectives. I assume that the conclusions drawn will be applicable to chemical plant incidents in a general sense, e.g. that weak management will lead to problems irrespective of the area which is managed weakly.

1.1 HARDWARE

1.1.1 Case Study 1

A warehouse fire involving reactive chemicals is described in the Health and Safety Executive's (HSE) investigation into the fire at Allied Colloids Ltd, Bradford on 21 July 1992. In Table 1 this incident is listed under the perspectives Technical issues highlighted – How did it happen? and In what sequence?

The HSE investigation report includes lessons learnt and considers the management of health and safety, but emphasises both the possible sequences of events and the scientific basis used to explain how the fire happened. The following points from the report give an indication of the nature of the investigation.

• Numerous employees who had been directly involved with the incident were identified and interviewed.

• Efforts were concentrated on gaining an appreciation of the range and quantities of chemicals involved and how and where they were stored.

• The scientific investigation examined the properties of and interactions between the materials stored, the nature of the products of combustion and the general spread of the fire. Extensive studies were carried out on a number of chemical samples from the site.

• The investigation also examined potential sources of heat capable of raising the temperature of the contents of reactive chemical kegs to the point where sufficient material would decompose to cause that package to fail.

• The investigation looked at the failure characteristics of the kegs and the spread of the fire through the warehouse. A technique known as computational fluid dynamics was used to help quantify the heat sources.

HSE investigations soon established that the stores contained a self-reactive substance which was thermally unstable and capable of undergoing violent decomposition at relatively low temperatures. The incident started when two or three kegs of this substance ruptured. These were stored on the top shelf of the racking in the warehouse, close to the steam condensate return line and a roof light panel. The sun would not have been shining directly on the kegs, and it was concluded that a malfunction of the steam heating system or operator error caused the condensate pipe to be hot.

1.1.2 Case Study 2

Other incident investigations, such as the catastrophic failure of an ammonia/air mixer described by Verduijn follow a similar form. There is an emphasis on establishing the sequence of events and the direct technical causes, which include the chemical-reactions involved and any failures of chemical plant items and also on describing lessons learnt.

1.1.3 Case Study 3

The investigation of the Hillsborough Football stadium disaster by the Health and Safety Executive focused on technical aspects of the disaster, including metallurgical examination of crash barriers, collapse load calculations and development of a model to predict crowd pressures, among other issues. Ninety-five people died from crush asphyxia because of severe overcrowding. The investigation provided important evidence in determining what happened at Hillsborough and enabled the elimination of some of the theories put forward in the aftermath of the disaster.

1.1.4 Case Study 4

Clearly, after a tragic incident there is a desire to find out what happened. This can sometimes lead to alternative technical explanations of the cause of the incident (even after the court of enquiry into the event). On 11 July 1978 a disaster occurred at a campsite in Spain in which over 200 people lost their lives due to fire and explosions involving a road tanker carrying liquefied petroleum gas which burst open and lost its contents. The exact events which occurred were the subject of a debate in the literature which centred around whether or not the tanker exploded due to over-loading followed by sunshine raising the tank temperature; whether a vehicle crash occurred; or, finally, whether the tanker was first engulfed in a fire which subsequently caused the explosion. Such debates and investigations are important because the outcome can affect subsequent design and operation of hazardous technologies.

1.1.5 Case Study 5

Alternative explanations also arose in the case of an ammonia tank failure in Lithuania.6 On 20 March 1989, a large storage tank containing re-frigerated liquid ammonia failed suddenly without any warning and moved sideways, demolishing a reinforced concrete bund wall and releasing 7000 tonnes of liquid ammonia. The ammonia vapour ignited and the resultant fire spread to involve 35,000 tonnes of compound fertiliser in nearby warehouses. Acid fumes from the burning solid continued for three or four days and the plume was seen from some 45 km distance, the clouds being discoloured with nitrous fumes many kilometres downwind. There were seven fatalities reported at the time of the incident and 57 reported injuries on site, most being gassed by the ammonia which formed a pool of liquid up to 70 cm in depth over a wide area. There were no reported fatalities outside the plant. Subsequent debate centred around the question of whether the event had occurred due to 'rollover', in which warm liquid at the bottom of the tank rises while cold liquid settles, the process causing the rupture of the tank, or whether the tank failed due to overpressure because of poor vent design.

1.1.6 Case Study 6

As mentioned above, it is important to establish the technical and operational detail of accident causation when the lessons learnt affect the subsequent design and operation of hazardous plant. The propane fire at Feyzin in France in 1966 is a case in point. A cloud of propane vapour spread 160 m until it was ignited by a motor car on the adjoining motorway. The liquid spillage burnt underneath the tank and over-heated it until it burst. A wave of burning propane then engulfed the firemen while debris broke the legs of the next tank and spread the fire. Major weak-nesses in the design of such installations were not appreciated until after the Feyzin disaster. In addition, the method of operation to be used with the improved design could be clearly specified. The Feyzin fire is listed in Table 1 under the perspectives of underlying hardware failures, where the cause of the accident is considered to be inadequate design of the plant, i.e. a hardware failure which is underlying rather than direct.

1.1.7 Case Study 7

Case study 6 can be contrasted with the ammonia tank failure at Potchefstroom, South Africa in 1973 when a 50 tonne storage tank for liquid ammonia failed. The fertilizer plant had four 50 tonne capacity horizontal bullet type ammonia pressure storage tanks. One of these tanks failed without warning, with a section approximately 25% of the cross sectional area coming out from one dished end of the 2.9 m diameter x 14.3 m long vessel as a result of brittle fracture. An estimated 30 tonnes escaped from the tank plus another 8 tonnes from a tank car. The gas cloud that was immediately formed was about 150 m in diameter and nearly 20 m in depth. The ammonia cloud caused the deaths of eighteen people including six outside the plant fence. Approximately 65 people required medical treatment in hospital and an unknown number were treated by private doctors. The end of the tank which failed had been repaired by welding and the welded area had not been stress relieved. Therefore, the most likely cause of failure was brittle fracture. In Table 1 this is called a direct hardware failure, as opposed to an underlying design problem.

The case studies on hardware failures therefore emphasise the technical and scientific causes of accidents. The sequences of events are established and plausible explanations are explored and set one against the other. Often samples are removed from the site and tested for their chemical and metallurgical properties. Often important lessons for the future are learnt in terms of design, material specification and scientific mechanisms. Scale models may be constructed and tested so that the best explanations for the incident become apparent.

1.2 PEOPLE

1.2.1 Case Study 8

The investigation into the Clapham Junction Railway Accident (12 December 1988, the Hidden report) is listed, in Table 1 under the perspectives of 'Complete Investigations' and 'Human Error'. This is because, to quote the report, "The direct cause of the accident was un-doubtedly the wiring errors made by a specified individual in his work in the Clapham Junction 'A' relay room (human error is a direct cause) ... a welter of criticism was justifiably laid at the door of this individual". But the report is also very broad in its approach. To quote Sir Bob Reid, "The immediate cause was some faulty wiring work. It was not particularly hard to pinpoint one or two individuals who were at fault. But lying behind the accident and lying behind individual errors was a whole chain of circumstances that has everything to do with management responsibility." The Hidden report covers, for example, training, supervision, working practices, hours of work and even pay and reward structures. At the time of the accident, MPs questioned whether there was a link between the accident and financial cutbacks at British Rail. Thus the incident high-lighted human error as a direct cause, but later investigations introduced the concepts of weak safety culture, underlying causes of failure and failures of management.

1.2.2 Case Study 9

Similarly, the Kegworth air disaster on 8 January 1989 implicated human error as a direct cause. It became clear that the engine on the port (No 1 or Left) side of the aircraft had caught fire but the starboard engine (No 2 or Right) had been shut down, in flight, by the pilots rather than having failed. A spokesman for the company which made the engines said that a pilot shutting down an engine would have to carry out two or three separate actions. A shutdown would have to be a series of deliberate acts.

Much later (February 1996) the Daily Mail ran a headline "Pilot blamed for jet disaster gets £100,000". The story ran:

A pilot partly blamed for the Kegworth air disaster which killed 47 people has won more than £100,000 compensation, it emerged yesterday. The Captain was awarded the cash by British Midland's insurers after he claimed the airline had not trained him properly on the twin-engined Boeing 737 he was flying. An enquiry into the crash showed he and his co-pilot shut down the wrong engine moments before the Heathrow to Belfast plane plunged on to a motorway in Leicestershire, UK. The Captain was paralysed in the accident seven years ago and is confined to a wheelchair. His solicitor refused to disclose the exact amount of the award, saying only that it was a 'substantial' six-figure sum. He added: 'We are reasonably satisfied.' The crash inquiry heard that the 737 was preparing for an emergency landing at East Midlands airport after one engine caught fire. But instead of shutting down the left-hand one which was engulfed in flames, The Captain and First Officer stopped the working right-hand engine. The aircraft, with 125 people on board, belly-flopped into a field, then ploughed across the motorway and was embedded in an embankment. The Captain claimed compensation for personal injury on the grounds that the plane developed a fault which led to the crash and he had not been given simulator training on how to deal with the problem. A Transport Department report on the disaster criticised The Captain and First Officer for acting against their training when they shut down the wrong engine. It said: "Their misdiagnosis must be attributed to their too-rapid reaction and not to any failure of the engine instrument system." But the reasons behind the error were never fully established. The Captain told the crash inquest: "I never doubted that we had got the correct engine." And the First Officer was convinced the new aircraft's electronic instrumentation panel wrongly indicated that the right-hand engine was on fire.

The Kegworth air crash is analysed by Denis Smith who says that while the pilots of the aircraft have shouldered the bulk of the blame to date ( 1992) it can be argued that the roots of the crash are much more complex than simply pilot error. Bill Richardson makes this point well. There are a range of views on the issue of what causes disasters. These range from 'simple-causal' explanations to 'complex systems' explanations. The 'simple' view holds that 'It is somebody's fault. A simple activity would have avoided this'. This view sees the problem as being essentially one of human failing, i.e. people (behaving negligently) do things or fail to do things and so create disasters.

However, there is an altogether more 'messy' view of disaster causation. In this 'messier' view, complex systems of beliefs, power, economics, social relationships, technologies, management systems and nature, for example, in both the organisation and its environment, interact with one another to create a complex, interactive system which is naturally prone to what Perrow has termed 'normal accidents'.

This complexity is clearly illustrated by the Daily Mail article quoted above in which the pilot 'blamed' for the accident receives compensation for his injuries. Thus the Kegworth incident implicated human error as a direct cause of the incident but also clearly shows the complexity of these issues and how underlying causes contribute to failure.

1.3 SYSTEMS AND CULTURES

1.3.1 Case Study 10

The escape of three prisoners from Parkhurst Prison on the Isle of Wight on 3 January 1995 is analysed by the Learmont report. The report concludes that Parkhurst should never have been retained as a high security prison. The procedures employed were lax and unprofessional, made worse by the disastrous indecision and apathy which pervaded so many working practices. The escape revealed a chapter of errors at every level and naivety that defies belief. The many shortcomings at Parkhurst illustrate the failure of the prison service to learn the lessons of the past. Thus, the escape was not due to one person's folly, because many of its ingredients can be traced along the lines of communication to prison service head-quarters. The numerous failings identified in this report indicate that there were many hands on the tiller on this voyage to disaster. The Guardian summarised this with the headline "Damning verdict on prison chiefs", which clearly implies a failure of management. In addition the idea of different levels of responsibility and failures of communication between them is clearly shown by the report.

(Continues...)

Excerpted from Risk Assessment by Nick W. Hurst. Copyright © 1998 The Royal Society of Chemistry. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

Details zur Ausgabe

• Verlag: Royal Society of Chemistry
• Erscheinungsdatum: 1998
• ISBN 10: 0854045546
• ISBN 13: 9780854045549
• Einband: Tapa blanda
• Sprache: Englisch
• Anzahl der Seiten: 124