Risk Analysis: Assessing Uncertainties Beyond Expected Values and Probabilities - Hardcover

Aven, Terje

Hardcover

ISBN 10: 0470517360 ISBN 13: 9780470517369

Publisher: Wiley, 2008

Synopsis

Everyday we face decisions that carry an element of risk and uncertainty. The ability to analyze, predict, and prepare for the level of risk entailed by these decisions is, therefore, one of the most constant and vital skills needed for analysts, scientists and managers.

Risk analysis can be defined as a systematic use of information to identify hazards, threats and opportunities, as well as their causes and consequences, and then express risk. In order to successfully develop such a systematic use of information, those analyzing the risk need to understand the fundamental concepts of risk analysis and be proficient in a variety of methods and techniques. Risk Analysis adopts a practical, predictive approach and guides the reader through a number of applications.

Risk Analysis:

• Provides an accessible and concise guide to performing risk analysis in a wide variety of fields, with minimal prior knowledge required.
• Adopts a broad perspective on risk, with focus on predictions and highlighting uncertainties beyond expected values and probabilities, allowing a more flexible approach than traditional statistical analysis.
• Acknowledges that expected values and probabilities could produce poor predictions -   surprises may occur.  
• Emphasizes the planning and use of risk analyses, rather than just the risk analysis methods and techniques, including the statistical analysis tools.
• Features many real-life case studies from a variety of applications and practical industry problems, including areas such as security, business and economy, transport, oil & gas and ICT (Information and Communication Technology).
• Forms an ideal companion volume to Aven’s previous Wiley text Foundations of Risk Analysis.

Professor Aven’s previous book Foundations of Risk Analysis presented and discussed several risk analysis approaches and recommended a predictive approach. This new text expands upon this predictive approach, exploring further the risk analysis principles, concepts, methods and models in an applied format. This book provides a useful and practical guide to decision-making, aimed at professionals within the risk analysis and risk management field.

About this title

About the Author

Terje Aven is Professor of risk analysis and risk management at the University of Stavanger, and a Principle Researcher at the International Research Institute of Stavanger (IRIS). Having worked in both industry and academia, he has participated in, and led, many safety and risk related projects, winning several awards for both research and teaching. He has published numerous papers in international journals, and has authored several books.

Risk analysis can be defined as a systematic use of information to identify hazards, threats and opportunities, as well as their causes and consequences, and then express risk. In order to successfully develop such a systematic use of information, those analyzing the risk need to understand the fundamental concepts of risk analysis and be proficient in a variety of methods and techniques. Risk Analysis adopts a practical, predictive approach and guides the reader through a number of applications.

• Provides an accessible and concise guide to performing risk analysis in a wide variety of fields, with minimal prior knowledge required.
• Adopts a broad perspective on risk, with focus on predictions and highlighting uncertainties beyond expected values and probabilities, allowing a more flexible approach than traditional statistical analysis.
• Acknowledges that expected values and probabilities could produce poor predictions - surprises may occur.
• Emphasizes the planning and use of risk analyses, rather than just the risk analysis methods and techniques, including the statistical analysis tools.
• Features many real-life case studies from a variety of applications and practical industry problems, including areas such as security, business and economy, transport, oil & gas and ICT (Information and Communication Technology).
• Forms an ideal companion volume to Aven’s previous Wiley text Foundations of Risk Analysis.

Professor Aven’s previous book Foundations of Risk Analysis presented and discussed several risk analysis approaches and recommended a predictive approach. This new text expands upon this predictive approach, exploring further the risk analysis principles, concepts, methods and models in an applied format. This book provides a useful and practical guide to decision-making, aimed at professionals within the risk analysis and risk management field.

Risk analysis can be defined as a systematic use of information to identify hazards, threats and opportunities, as well as their causes and consequences, and then express risk. In order to successfully develop such a systematic use of information, those analyzing the risk need to understand the fundamental concepts of risk analysis and be proficient in a variety of methods and techniques. Risk Analysis adopts a practical, predictive approach and guides the reader through a number of applications.

• Provides an accessible and concise guide to performing risk analysis in a wide variety of fields, with minimal prior knowledge required.
• Adopts a broad perspective on risk, with focus on predictions and highlighting uncertainties beyond expected values and probabilities, allowing a more flexible approach than traditional statistical analysis.
• Acknowledges that expected values and probabilities could produce poor predictions -   surprises may occur.  
• Emphasizes the planning and use of risk analyses, rather than just the risk analysis methods and techniques, including the statistical analysis tools.
• Features many real-life case studies from a variety of applications and practical industry problems, including areas such as security, business and economy, transport, oil & gas and ICT (Information and Communication Technology).
• Forms an ideal companion volume to Aven’s previous Wiley text Foundations of Risk Analysis.

Professor Aven’s previous book Foundations of Risk Analysis presented and discussed several risk analysis approaches and recommended a predictive approach. This new text expands upon this predictive approach, exploring further the risk analysis principles, concepts, methods and models in an applied format. This book provides a useful and practical guide to decision-making, aimed at professionals within the risk analysis and risk management field.

Excerpt. © Reprinted by permission. All rights reserved.

Risk Analysis

Assessing Uncertainties Beyond Expected Values and ProbabilitiesBy Terje Aven

John Wiley & Sons

Copyright © 2008 John Wiley & Sons, Ltd
All right reserved.
ISBN: 978-0-470-51736-9

Chapter One

Risk analysis methods

This chapter presents a selection of methods that can be used when carrying out a risk analysis. These methods are described in detail in the literature and in a number of textbooks within this professional field (Modarres 1993, Rausand and Hyland 2004, Bedford and Cooke 2001, Vose 2000, to mention a few). In this book, we present a short summary of the most fundamental methods, partly based on Aven (1992).

6.1 Coarse risk analysis

A coarse risk analysis (often also referred to as a preliminary risk analysis) is a common method for establishing a crude risk picture, with relatively modest effort. The analysis covers selected parts of, or the entire, bow-tie (see Figure 1.1), i.e. the initiating events, the cause analysis and the consequence analysis. The analysis team typically consists of 3-10 persons.

Often the coarse risk analysis is performed by dividing the analysis subject into sub-elements and then carrying out the risk analysis for each of these sub-elements in turn. This applies regardless of whether the analysis focuses on a section of a highway, a production system, an offshore installation or other analysis subjects. Checklists may be used as a tool for identifying and analysing hazards and threats for each sub-element to be analysed.

The form used to document the risk analysis is often standardised. An example of an analysis form for a risk analysis of a road tunnel is shown in Table 6.1. We see from the table that the risk is described by using categories. The categories cover possible undesirable events, along with the probability and expected consequence if such an event should occur. We see from the table that, in the case of a bus fire, we expect that there will be 10 people killed. The number can be 0, 1 or 30, but the expectation is 10.

In stating probabilities, terms such as often and seldom should be avoided as they are open to different interpretations. A better alternative is to say directly what we mean, for example, 10-50% probability that an event will occur within the period of 1 year. Some people will perhaps say that it is difficult for the analysis group to express that there is a 1-10% probability or a 10-50% probability, for example. The answer to this is "yes," it can be difficult to assign probabilities. However, it does not help "hiding" behind expressions such as "often" or "seldom" without explaining what one means by these terms. Also, the consequence categories should be precisely defined, rather than using terms such as high, low, etc.

A coarse risk analysis is often combined with other analysis methods. The coarse analysis identifies the most important risk contributors, and then the causal picture and/or the consequence picture can be assessed in detail using more detailed analyses.

Example: workplace accidents

The working environment committee at the Packing Factory Ltd has found that in the bag department, which has about 90 employees, the number of injuries is too high. The committee therefore decides to implement some injury preventive measures to reduce the injury rate in the department. It is, however, not clear where such measures should be directed and which measures would most effectively prevent injuries. Opinions in the working environment committee differ widely.

In the bag department, production is a series production of large, multiple-ply paper bags. Each production line comprises several machines. The raw material is paper rolls. The main operator task in the department is to monitor and adjust the machines. Some operators deal with manual handling of products. It is the operators who are operating the machines that are most exposed to injuries.

The working environment committee instructs the safety delegate of the company to work out a basis for decisions on preventing measures. The safety delegate is familiar with risk analyses and he uses such an analysis to establish the desired decision basis.

By means of the risk analysis, the safety delegate will identify possible injuries that might occur in the department, where they might occur, possible causes and the severity of the injuries. Table 6.2 presents a summary of the analysed hazards. With respect to probability/frequency, the following classification is used:

1. very unlikely: less than once per 1000 years (yearly probability 1:1000);

2. unlikely: once per 100 years (yearly probability 1:100);

3. quite likely: once per 10 years (yearly probability 1:10);

4. likely: once per year;

5. frequently: once per month or more frequently.

Also for the consequences five categories are used:

I does not result in injuries II minor injuries III major injuries

IV death or total disability

V death or total disability for several persons.

The starting point for the hazard identification was the injury reports for the whole department for the last 9 years. Data on near misses were also available. In addition, the system description was studied to identify other hazards. For the hazards based on the injury reports, the classification is based on this statistics. For example, two injuries caused by crushing in the cutting machinery are registered. This gives a classification "likely." Judgement has been used for the hazards that are not based on the injury statistics.

In Figure 6.1 the hazards have been placed in a consequence probability diagram. The consequence categories are marked along the horizontal axis with consequences increasing to the right. Similarly, the frequency/probability is marked along the vertical axis, with frequency increasing downwards. The following conclusions are drawn:

highest risk: hazard number 14, which equals "paper roll falls from tackle;" other contributors to high risk are: events that include crushing and catching in the machinery.

The events with the lowest risk are numbers 9 and 15. Thus, in general, for this type of consequence probability diagram, the events with the highest risk are those in the bottom right corner, whereas the events with the lowest risk are placed in the upper left corner. We should, however, be very careful when drawing conclusions from the matrix since it is based on a rough classification.

Note that hazard 15 (paper fire) has low risk in relation to personnel. If we focus on material assets or economic values, this hazard would contribute much more to risk.

Based on the risk analysis, the safety delegate can now identify and rank measures to prevent accidents, as a basis for the decision-making on which measures to implement.

6.2 Job safety analysis

A job safety analysis is a simple qualitative risk analysis methodology used to identify hazards that are associated with a work assignment that is to be executed. A job safety analysis is usually checklist-based. Normally, the persons planning/executing the work assignment are part of the analysis team. By carrying out a job safety analysis, we ensure:

It is clarified whether the work assignment is a "standard" operation that can be carried out according to procedures and normal practice or if it is a non-standard case that requires special measures or studies. The latter case may lead to postponement until more detailed studies are carried out. Possible conflicts between different jobs may be identified; for example, painting and welding jobs close to each other at the same time. The persons carrying out the work assignment will think through what they should do and consider each work assignment in a risk-related perspective. The mere act of thinking through and planning the work assignments can, in itself, be a risk-reducing measure. What can go wrong at the various steps of the job will be assessed. Through this process, those carrying out the job will become aware of the most risky aspects of the work assignment, and adequate risk-reducing measures can then be implemented.

A job safety analysis is carried out by dividing the job into a number of sub-jobs or tasks and then performing an analysis for each task. The division into tasks is illustrated by the following example:

Change of a car wheel

1. Set the hand brake. 2. Take out the spare wheel from the boot. 3. Check the air pressure. 4. Remove the hub cap. 5. Ensure that the jack fits and is stable. 6. Jack up the car, but not so much that the wheels leave the ground. 7. Loosen the wheel nuts. 8. Jack up the car further, but not more than is necessary. 9. Remove the wheel, and so on.

The identification of hazards includes a check of:

What type of injuries that may occur, for example crushing. Are special problems or deviations likely to occur? Is the task difficult or uncomfortable to carry out? Are there alternative ways of carrying out the task?

The identified hazards are assessed and the conclusions categorised for example in the following way:

0 insignificant risk 1 acceptable risk; actions unnecessary 2 the risk should be reduced 3 the risk must be reduced; there is a need for immediate actions.

When evaluating the risk and the need for actions/risk-reducing measures, considerations should be given to, for example:

violation of statutory requirements; violation of requirements set by the company; high risk documented by means of accident statistics; high energy concentration; unreasonable requirements with respect to attention and vigilance for the operator; low tolerance for human errors in the technical system; whether the solution of the problem is known and available.

Special sheets have been developed for job safety analysis. Such sheets will typically include the following main points:

description of the job accident experience (statistics) accident potential requirements job sequence (tasks) risk assessment actions/measures.

Often the sheets include a list of possible actions that are to be considered. The actions may for example be related to improved equipment and tools, better work instructions, improved education and training, and so on.

6.3 Failure modes and effects analysis

Failure Modes and Effects Analysis (FMEA) is a simple analysis method to reveal possible failures and to predict the failure effects on the system as a whole. The method is inductive; for each component of the system, we investigate what happens if this component fails. The method represents a systematic analysis of the components of the system to identify all significant failure modes and to see how important they are for the system's performance. Only one component is considered a time, and the other components are then assumed to function perfectly. FMEA is therefore not suitable for revealing critical combinations of component failures.

FMEA was developed in the 1950s and was one of the first systematic methods used to analyse failures in technical systems. The method has appeared under different names and with somewhat different content. If we describe or rank the criticality of the various failures in the FMEA, the analysis is often referred to as an FMECA (Failure Modes, Effects and Criticality Analysis). The criticality is a function of the failure effect and the frequency/probability as seen below. The difference between an FMEA and an FMECA is not distinct, and in this book we do not distinguish between these two methods. In the following we also use the term FMEA when the analysis includes a description/ranking of criticality.

In several enterprises, it is nowadays a requirement that an FMEA be included as part of the design process and that the results from the analysis be part of the system documentation.

To ensure a systematic study of the system, a specific FMEA form is used. The FMEA form may for example include the following columns:

Identification (column 1). Here the specific component is identified by a description and/or number. It is also common to refer to a system drawing or a functional diagram.

Function, operational state (column 2). The function of the component, i.e. its working tasks in the system, is briefly described.

The state of the component when the system is in normal operation, is described, e.g. whether it is in a continuous operation mode or in a stand-by mode.

Failure modes (column 3). All the possible ways the components can fail to perform its function are listed in this column. Only the failure modes that can be observed from "outside" are included. The internal failure modes are to be considered as failure causes. These causes can possibly be listed in a separate column. In some cases it will also be of interest to look at the basic physical and chemical processes that can lead to failure (failure mechanisms), such as corrosion.

Often we also state how the different failure modes of the component are detected and by whom.

Example: In a chemical process plant, a specific valve is considered as a component in the system. The function of the valve is to open and close on demand. "The valve does not open on a demand" and "the valve does not close on a demand" are relevant failure modes, as well as "the valve opens when not intended" and "the valve closes when not intended." However, "washer bursts" is an example of a cause of a specific failure mode.

Effect on other units in the system (column 4). In those cases where the specific failure mode affects other components in the system, this is stated in this column. Emphasis should be given to identification of failure propagation, which does not follow the functional chains of the functional diagrams. For example, increased load on the remaining pillars that are supporting a common load when a pillar collapses; vibration in a pumping house may induce failure of the driving unit of the pump, etc.

Effect on system (column 5). In this column, we describe how the system is influenced by the specific failure mode. The operational state of the system as a result of failure is to be expressed, for example, whether the system is in the operational state, changed to another operational mode, or not in an operational state.

Corrective measures (column 6). Here we describe what has been done or what can be done to correct the failure, or possibly to reduce the consequences of the failure. We may also list measures that are aimed at reducing the probability that the failure will occur.

Failure frequency (column 7). Under this column, we state the assigned frequency (probability) for the specific failure mode and consequence. Instead of presenting frequencies for all the different failure modes, we may give a total frequency and relative frequencies (in percentages) for the different failure modes.

Failure effect ranking (column 8). The failure is ranked according to its effect with respect to reliability and safety, the possibilities of mitigating the failure, the length of the repair time, the production loss, etc. We might for example use the following grouping of failure effects:

Small: A failure that does not reduce the functional ability of the system more than normally is accepted. Large: A failure that reduces the functional ability of the system beyond the acceptable level, but the consequences can be corrected and controlled.

Critical: A failure that reduces the functional ability of the system beyond the acceptable level and which creates an unacceptable condition, either operational or with respect to safety.

Remarks (column 9). Here we state, for example, assumptions and suppositions.

By combining the failure frequency (probability) and the failure effect (consequence), the criticality of the specific failure mode is determined.

(Continues...)

Excerpted from Risk Analysisby Terje Aven Copyright © 2008 by John Wiley & Sons, Ltd. Excerpted by permission.
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.

Edition details

• Publisher: Wiley
• Publication date: 2008
• Language: English
• ISBN 10: 0470517360
• ISBN 13: 9780470517369
• Binding: Hardcover
• Edition number: 1
• Number of pages: 204