Operational Decision-making in High-hazard Organizations

Drawing a Line in the Sand

Jan Hayes

• Publisher: CRC Press
• Year: 2017
• eBook ISBN: 9781317085355
• Topic: Technology & Engineering
• Subtopic: Operations
• Index: Technology & Engineering

This book takes a fresh look at safety decision-making by documenting and examining stories told by front-line managers in three different high-hazard industries: a chemical plant, a nuclear power station and an air-navigation service provider. From Piper Alpha to Deepwater Horizon, accident analysis has stressed the importance of excellent decision-making by those in charge out in the field. Organizations rely critically on the judgement and experience of such senior operations personnel and yet these qualities are undervalued in a business environment that emphasises documentation and measurement. Whilst operational managers are guided by rules, they also draw on their own long experience and can formulate a situation-specific 'line in the sand' to apply the experience of the operating team to complex, real-world situations that rule writers may not have foreseen. This volume refocuses our attention on the people who make these important decisions and the organizational processes that support the best choices. Jan Hayes uses her multi-disciplinary experience to draw together an account of safety decision-making that is both technically robust and yet accessible to academics, practitioners and regulators alike. Readers will see that the stories retold in this book provide a way for operational managers to share their knowledge, experience and expertise - with each other and with us.

Chapter 1
Introduction

‘Safety decision-making by operational people? There’s nothing to study because it’s all written into our procedures.’ This statement from one oil refinery executive is typical of the view held by many managers. In fact, this book seeks to highlight that this is a significant simplification of how people in high performing organizations dealing with complex hazards work to maintain public safety. As we shall see, such organizations rely critically on the professional judgement and expertise of senior operational staff, often without understanding that this is the case. This book seeks to describe how key decisions are made by people working in three high-hazard organizations – a chemical plant, a nuclear power station and an air navigation service provider. In these days of cost cutting and business process management where only things that can be measured are valued, this has important implications for public and worker safety.

Research on decision-making in organizational settings has a long history and has been taken up by a range of academic disciplines including economics, management and cognitive psychology. Decision-making has been widely viewed as a key management process and normative theories have led to the development of a range of practical aids and training courses on improved decision-making. Many of these use the ‘classical decision method’ involving rational analysis of options in order to make an optimal choice. Since the 1980s, safety decision-making has followed this trend and company systems are now typically based on a specific form of rational analysis that we call risk management. Such methods are apparently so ubiquitous that there may seem to be no need for a book on operational decision-making in particular since there is nothing very interesting to say.

On the contrary, the fundamental proposition of this book is that, despite little understanding or acknowledgement, organizations rely critically on the experience and judgement of professionals such as senior operational staff to keep workers and the general public safe. Risk-based rules and procedures play an important role in ensuring high levels of safety performance but such systems of decision-making replace one large judgement with lots of smaller ones. This book shows that such judgements are critical in the safe operation of complex, hazardous systems and that an excessive focus on process, rather than content, has obscured the role that professional experience and judgement has always played, and will continue to play, in organizational safety.

One group of industrial and infrastructure organizations operates in an environment that makes some decisions especially important. These are high hazard organizations, where the technology or activities involved mean that, if things go wrong, many people could be injured or killed. Air traffic control, nuclear power generation, offshore oil and gas production and petrochemicals manufacturing are all examples of organizational activities where the intrinsic danger is high, even if the risk is low due to the low likelihood of a serious failure. Much safety research in these types of organizations is based on field observations. The literature groans with accounts by social scientists of their experiences in control rooms, flight decks, operating theatres and emergency rooms. Another genre of decision-making research focuses on management decisions made in offices and meeting rooms well removed from the coal face (or control room, flight deck etc). The work described here does not adopt either of these perspectives, but aims to cover the space in between these two views. Our focus is on those decisions made by operational managers. These individuals, who supervise field personnel directly, are the most senior people on shift and provide the link to more senior management remote from day-to-day operations. The following story describes one such individual at work in his job at a nuclear power station.

STORY 1: COULD THE SAME FAULT BE PRESENT IN THE RUNNING REACTOR?

During an outage of one nuclear reactor for routine maintenance and inspection (whilst the other reactor on the site remained operational), an internal weld failure was discovered. This fault was unexpected and the duty shift was heavily involved with engineering and maintenance personnel in deciding how the repairs were to be carried out on the offline reactor.

Shift Manager Interviewee 4 came on duty at a later stage and asked the technical specialists two questions: could the same fault be present in the running reactor; if so, what is the worst credible damage that could be present? The specialists could not be sure that the running reactor did not have the same fault and felt that they could not discount the possibility of a significant gas release as a result (although there was no question of loss of radiation containment).

Based on the advice he received, Interviewee 4 decided to shut down the running reactor. This reactor was found subsequently to have a similar fault, but engineering analysis showed that there was no potential for gas leakage.

Interviewee 4 describes the outcome: ‘In a nutshell if you just want to look at it on the balance sheet, I took the reactor off for no good reason … and we were off for a week, so I cost the company a lot of money there. Do you see what I mean? In hindsight that was a week’s worth of generation lost because I decided to come off because I had this gut feeling that I wasn’t happy with it … he [the site manager] openly congratulated me on making the right decision.’

This single story illustrates several key issues. In this case, the operational manager made a very conservative choice. The traditional cautionary tale in the safety literature would have a different ending with the overall conclusion that he averted a near-disaster by his actions. In this case, the actual result was more banal and yet this story is perhaps more interesting because it does not take the expected line. Nuclear power station operations are highly proceduralized, but in identifying the potential problem and choosing this course of action, this operational manager had no procedure to follow. His actions were a direct result of his professional and organizational environment. We will explore the factors behind decisions such as these – the perspective on system operations that led to this operational manager seeing what he thought was the potential for an accident, his confidence in interrupting operations based on that potential and the response of the organization to the outcome. We will also consider what constitutes a good decision. In more everyday circumstances, we tend to judge decisions by their outcome and yet by that measure the operational manager in the story above made a poor choice. In seeking excellent safety performance, different measures are needed to judge the quality of actions taken.

Before looking further at decision-making practices, there are two further introductory issues to be addressed. The first is why we should be interested in looking at decision-making in cases where there has been no major problem. People in industry love hearing stories about disasters (provided that they happen to other people and the focus is on technical, rather than personal, details). Putting aside the question of schadenfreude (or taking pleasure in the misfortune of others), such examples can provide important, concrete lessons about what not to do. This book could have focused, for example, on problems with operational decision-making in the cases of two recent blowouts in the offshore oil and gas industry: Deepwater Horizon (Hopkins 2012) and Montara (Hayes 2012). Instead, in the tradition of high reliability research, the focus is on lessons that can be drawn from cases where things go right.

High Reliability Theory (HRT) attempts to explain how successful high hazard organizations manage to operate in a way that is generally failure free. Rather than focusing on accident analysis, theorists in this field look at how organizations behave in order to minimise both the number and severity of incidents. HRT focuses on the organizational qualities that are required to achieve ‘mindfulness’, which is seen as the key to a high level of safety performance (Weick and Sutcliffe 2001). Organizations that achieve this are known in this field as High Reliability Organizations or HROs. This book is in that tradition of ‘normal operations studies’ (Bourrier 2002, 2011). More detail on high reliability theory and its relevance to operational decision-making can be found in Chapter 2.

The final introductory point is to highlight the role played by stories in this book. This case study based research generally follows ethnographic methods of the kind widely used in sociological research (Silverman 2001), in sensemaking research (Weick 1995) and in safety research (Hopkins 2006). Most data collection was via semi-structured interviews, although workplace observations and document review also contributed. Interviews were recorded, transcribed and reported in the form of a series of stories detailing the experiences of operational managers with extensive direct quotations from the people involved. Reporting data in the form of stories was a deliberate choice. Dreyfus’s (1986) model of expertise and learning suggests that, whilst universal rules and generalized models are very important when one starts to learn a new skill, once an individual acquires some experience in any given field, his/her efforts move to using context-specific information to determine how best to achieve the desired outcome. Actions are based on experience and intuition, not conscious application of logic or rules. Flyvbjerg (2001) points out that case study research can provide a rich source of input to our development of context-specific experience and intuition – our mental models. Case study research is therefore an important way of increasing the store of human knowledge. The book aims to explore the individual and organizational context of safety decision-making in situations where production pressures are also ever-present; and to present the results of the work in a case study form that makes it easy for other experienced safety practitioners to add to their own understanding of these issues.

It is perhaps worth acknowledging at this point that this type of knowledge may be unfamiliar to some readers. In the physical sciences and in engineering, much use is made of relevant theories, and generally these can be expressed quantitatively, often in mathematical terms using equations. The situation in the social sciences is quite different in that ‘theory’ can also refer to a body of knowledge in qualitative terms. More detail on learning, stories and case studies can be found in Chapter 2.

This book is structured into 11 chapters.

Chapter 2 gives some theoretical background on the key ideas used in the analysis of decision-making.

The detailed content of the book is divided into two parts. In Part A, Chapters 3, 4 and 5 record the stories told by operational decision makers working in a nuclear power station, a chemical plant and an air navigation service provider.

Part B of the text examines the details and context of those stories. Chapter 6 discusses the dual organizational identities of the operational managers – as employees and as professionals. Chapter 7 addresses how rules are used by this group. Chapter 8 looks further at professionalism and how it impacts decision-making. The impact of relationships with peers, subordinates and managers is discussed in Chapter 9. Chapter 10 looks at the form of the experienced-based judgements made by the operational managers.

Chapter 11 includes a summary of the key arguments in the book and a discussion of the implications for organizations and for regulators, including some suggestions of how to support operational decision-makers in practical ways.

Chapter 2
Theoretical Perspectives on Making Safe Decisions

This chapter summarizes the key theoretical perspectives that form the basis of the analysis of decision-making practices in the remainder of the book. The aim is to provide background for those less familiar with these ideas and/or those who are particularly interested in the theoretical foundations of the work. The analysis in this book draws on each of these in making sense of the stories told by the operational managers.

As described in the Introduction, the main theoretical context to this research is High Reliability Theory (HRT). Work in this field has highlighted several key qualities of high performing organizations that are relevant to decision-making and these are discussed below in Section 2.1. Section 2.2 briefly describes the rather similar approach taken by Resilience Engineering (RE) researchers and the implications for operational decision-making. In contrast, many companies use the language of risk management to describe their decision-making and this is also discussed, within the context of classical decision theory, in Section 2.3. Theories of accident causation are another important aspect of safety research and what they have to say about decision-making is summarized in Section 2.4. The most relevant is James Reason’s Swiss cheese model.

Our decisions are based on what we know and hence are intertwined with how we learn. Theoretical perspectives on how experts learn are described in Section 2.5. Finally, the perspective of sensemaking is reviewed in Section 2.6. This way of thinking about decisions emphasizes the role of past experience, not from the perspective of rational deliberation but as a way of describing what experienced people notice in complex situations and how that translates to action.

2.1 High Reliability Theory

This body of work has its origins in analysis of the 1979 Three Mile Island nuclear power station incident in the US. The original ideas (La Porte 1981) were published in a set of essays (Sills et al. 1982) that also includes Perrow’s early work on Normal Accident Theory, which is described later in this chapter. Researchers studied organizations that are required to operate at very high levels of safety. La Porte (1991) also acknowledged a second operational challenge of the group of organizations defined as High Reliability Organizations or HROs: ‘to maintain the capacity for meeting intermittent, somewhat unpredictable, periods of very high peak demand and production’. (La Porte 1996: 60) Rochlin (1993) reviewed the participating organizations and established a set of six criteria that provides a working definition of an HRO:

1. The organization is required to maintain a high level of safety performance if it is to be allowed to continue to operate.

2. The organization must also maintain high levels of capability, performance and service to meet public and/or economic expectations and requirements.

3. Because of the consequences of error or failure, the organization cannot easily make marginal trade-offs between capacity and safety. Safety is not fungible.¹

4. As a result, primary task-related learning cannot proceed by trial and error since the first error may be the last trial.

5. The technology and primary task are both so complex that safety and capacity issues must be actively and dynamically managed.

6. The organization will be judged to have failed and will be criticized almost immediately if either the safety performance or service/product delivery degrades.

More recently Karlene Roberts (another member of the original Berkeley team who still works in this field) described an HRO as being ‘an organization in which errors can have catastrophic outcomes, but which conducts relatively error free operations over a long period of time making consistently good decisions resulting in high quality and reliable operations’. (Bourrier 2005: 94) It is argued by high reliability researchers that organizations such as nuclear power stations already operate at extraordinary levels of safety performance (La Porte 1996). The aim of their research is to identify those facets of organizations that lead to this high reliability, that is, a demonstrated greatly reduced potential for serious accidents.

Early work on high reliability organizations (La Porte and Consolini 1991) focused on decision-making as one of three areas where organizations with the potential for catastrophic failure were likely to differ from other (low reliability) organizations. (The other areas were structural responses to hazards and peak loads, and the tightly coupled, interdependent nature of operations.) LaPorte et al. (1991) describe the challenges of decision-making for HROs as:

• Extending the rational decision-making process as far as possible within the constraints of the data and time available for operational decision-making and mandatory adherence to formal documented operating procedures.

• Being sensitive to areas where incremental decision-making based on judgement m...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Contents
• List of Figures
• List of Tables
• List of Stories
• List of Abbreviations
• Foreword
• Preface
• Acknowledgements
• 1 Introduction
• 2 Theoretical Perspectives on Making Safe Decisions
• PART A DECISION-MAKING IN THREE HIGH RELIABILITY ORGANIZATIONS
• PART B ACTING BOTH AS EMPLOYEES AND AS PROFESSIONALS
• References
• Index