Resilience Engineering in Practice, Volume 2

Becoming Resilient

Christopher P. Nemeth, Erik Hollnagel

• Publisher: CRC Press
• Year: 2016
• ISBN:
  9781317065227
• Edition: 1
• Topic: Technologie et ingénierie
• Subtopic: Santé et sécurité au travail

This is the fifth book published within the Ashgate Studies in Resilience Engineering series. The first volume introduced resilience engineering broadly. The second and third volumes established the research foundation for the real-world applications that then were described in the fourth volume: Resilience Engineering in Practice. The current volume continues this development by focusing on the role of resilience in the development of solutions. Since its inception, the development of resilience engineering as a concept and a field of practice has insisted on expanding the scope from a preoccupation with failure to include also the acceptable everyday functioning of a system or an organisation. The preoccupation with failures and adverse outcomes focuses on situations where something goes wrong and the tries to keep the number of such events and their (adverse) outcomes as low as possible. The aim of resilience engineering and of this volume is to describe how safety can change from being protective to become productive and increase the number of things that go right by improving the resilience of the system.

Chapter 1

An Emergent Means to Assurgent Ends: Societal Resilience for Safety and Sustainability

Per Becker, Marcus Abrahamsson and Henrik Tehler

Societal safety and sustainability are key challenges in our complex and dynamic world, causing growth in interest of applying the concept of resilience in broader societal contexts. This chapter presents a concept of societal resilience that builds on established theory of Resilience Engineering and operationalises the concept by presenting its purpose, required functions and a way to identify and analyse the complex network of actual forms that together achieve these functions in society. The framework for analysing societal resilience is then tested in practice with interesting results. Although the framework has challenges and limitations, the Resilience Engineering approach to societal resilience seems to be both a conceptually and pragmatically fruitful path to follow.

Introduction

Contemporary society seems preoccupied with the notion of risk and recent examples of calamity have given rise to growing public discontent with the performance of present risk management institutions (Renn, 2008, p. 1). The safety and sustainability of society is thus increasingly becoming the centre of attention of policymakers from various administrative levels and countries around the world (for example, OECD, 2003; Raco, 2007). Advancing safety and sustainability is challenging in this context, as there are multiple stakeholders to involve (Haimes, 1998, p. 104; Renn, 2008, pp. 8–9), values to consider (Belton and Stewart, 2002), and stresses to include (Kates et al., 2001, p. 641). On top of this lies the multitude of factors and processes contributing to the susceptibility of what stakeholders’ value to be the impact of each stress (Wisner et al., 2004, pp. 49–84; Coppola, 2007, pp. 146–61).

The real challenge of societal safety and sustainability is however not the number of elements to include, but the complexity and non-linearity of relations between these elements (Yates, 1978, p. R201), separating cause and effect in both space and time (Senge, 2006, p. 71). Unfortunately, in efforts to promote safety and sustainable development, stakeholders often reduce problems into parts that fit functional sectors, organisational mandates and academic disciplines (Fordham, 2007). This is likely to be a major weakness as it clouds the bigger picture of risk (Hale and Heijer, 2006, p. 139) and is further complicated by various processes of change increasing the dynamic nature of our world, for example, globalisation (Beck, 1999), demographic and socio-economic processes (Wisner et al., 2004), environmental degradation (Geist and Lambin, 2004), the increasing complexity of modern society (Perrow, 1999b) and climate change (Elsner et al., 2008). It is in this context that Resilience Engineering may offer a conceptual framework to build on for meeting the challenges of societal safety and sustainability in the 21st century and beyond.

The purpose of this chapter is to present a framework for addressing challenges to the safety and sustainability of societies, by defining and operationalising a concept of societal resilience. The chapter also presents examples from applications of the framework in different contexts.

A Concept of Societal Resilience

Resilience Engineering has been paramount in demonstrating that the main challenge for safety is to recognise dynamic complexity and non-linear interdependencies in the system in question (for example, Hollnagel, 2006, pp. 14–17). Similarly, Sustainability Science has been equally paramount in demonstrating the same challenge for sustainability (for example, Kates et al., 2001). While Resilience Engineering appears to have generally been focusing on socio-technical systems (for example, Cook and Nemeth, 2006, p. 206; Leveson et al., 2006, p. 96), Sustainability Science has often approached our world as a complex human-environment system (for example, Turner et al., 2003; Haque and Etkin, 2007). Regardless of type of system, destructive courses of events that threaten safety and sustainability are, in both views, not results of linear chains of events, like dominos falling on each other (Hollnagel, 2006, pp. 10–12), but are instead non-linear phenomena that emerge within the complex systems themselves (Perrow 1999a; Hollnagel, 2006, p. 12). Such destructive courses of events are thus not discrete, unfortunate and detached from ordinary societal processes, but intrinsic products of everyday human-environment relations over time (Hewitt, 1983, p. 25; Oliver-Smith, 1999), and rooted in the same complex system that supplies human beings with opportunities (Haque and Etkin, 2007).

The concept of resilience has a wide range of definitions, developed in scientific disciplines spanning from engineering to psychology. Many of these definitions describe resilience as ability to ‘bounce back’ to a single equilibrium (for example, Cohen et al., 2011), as a measure of robustness or buffering capacity before a disturbance forces a system from one stable equilibrium to another (Berkes and Folke, 1998) or as ability to adapt in reaction to a disturbance (Pendall et al., 2010). Although many of these definitions are useful for their intended purposes, human-environment systems are adaptive and entail human beings with the ability not only to react to disturbances but also to anticipate and learn from them. For instance, a bicycle lane with a curb crossing would not be considered a particularly resilient system, even if it involved a qualified nurse and bike repairman ready to aid the unfortunate passerby to get back into the saddle again. It would be more resilient if the curb is removed, either before the foreseeable accident happens (anticipation), or after the first incident (learning).

To advance societal safety and sustainability, it is crucial to approach society as a complex human-environment system, and its level of safety and sustainability is determined by internal attributes. Societal resilience is in this sense an emergent property of such a system in the same way as Pariès’ (2006) organisational resilience of complex organisations. To better grasp this emergent property, Rasmussen (1985) suggests to structure systems in a functional hierarchy from purpose, through increasingly concrete levels of function, to the observable physical forms of the system contributing in the real world to fulfil the functions and meet its purpose. However, it is important to note that resilience is not a linear outcome of these functions, but an emergent property of the system that is connected in complex ways to the ability of the system to perform.

In the context of societal resilience, the overarching purpose of the human-environment system under study is to protect what human beings value, now and in the future. Hollnagel’s (2009) four cornerstones of resilience form a comprehensive foundation for the functions fulfilling that purpose. Although his framework is compelling, with its focus on anticipation, monitoring, responding and learning (Ibid.), it needs some minor alterations to suit the broader societal context. More specifically, to increase the usefulness of it in practice we suggest the introduction of a set of general functions on a lower level of abstraction (see Rasmussen, 1985) that will facilitate the assessment of societal resilience. It is important to note that although Hollnagel’s four pillars, and the associated abstract functions in our approach, are complete, the more concrete generalised functions used in this chapter are not. There may in other words be other generalised functions that are useful in other contexts, and the ones presented may be divided in other ways.

Moreover, we also suggest that these generalised functions could be divided into proactive and reactive ones. A function is here defined as proactive if it has an ex ante focus, that is, it focuses on something that has not yet happened. A reactive function, on the other hand, has an ex post focus, that is, it focuses on an actual event that has already taken place. This division into groups of reactive and proactive functions is useful in the present context since it emphasises that societal resilience is not only about ‘bouncing back’ from a disturbance, but also about adapting the system beforehand and learning from previous events. Furthermore, the division into two sets of functions also allows us to study the interaction between the two types, for example between the preparedness and response functions (see below), and it allows us to specifically address the different contextual factors that influence the performance of the proactive functions compared to the reactive ones. Examples of contextual factors that usually exist to a greater extent for reactive functions than proactive ones include high time pressure, large stakes and rapidly changing conditions.

Starting with the first cornerstone of Hollnagel’s framework, the function of anticipation, we suggest that a more concrete generalised function in a societal context is the function of Risk assessment. We agree with Hollnagel when stating that methods for risk assessment that focus on linear combinations of discrete events may fail to sufficiently represent risk as they fail to take into account the complexity of our world (Ibid., pp. 125–7). However, this is not a general attribute of risk assessment per se, and there are methods that to a greater extent incorporate such complexity (Haimes, 2004; Petersen and Johansson, 2008). There is obviously no such thing as a perfect method for risk assessment, but, as Hollnagel admits, ‘a truly resilient organization realizes the need at least to do something’ (Hollnagel, 2009, p. 127). A related but perhaps less contentious way of anticipation is Forecasting, for example, weather forecasts, river flow as a result of potential rainfall, ocean waves if a storm grows stronger and so on. Both Risk Assessment and Forecasting are proactive functions.

The second cornerstone emphasises the need to monitor specific predefined indicators of potential problems (Ibid., pp. 124–5), for example actual river flow, number of cholera cases in the area and so on. Hollnagel’s concept of monitoring covers in other words what ‘is or could be a threat in the near term’ (Ibid., p. 120), but not functions that are vital when the system is already in a specific disastrous event. In such a situation the system needs a function to recognise what impact that event has on the system. It is thus suggested that the second cornerstone of resilience is modified and called recognising, covering the generalised functions of Monitoring and Impact assessment. The latter clearly being reactive, while the former being potentially both proactive and reactive depending on how the set value for the indicator that is being monitored is defined in relation to what constitutes a real crisis (Figure 1.1). Impact assessment is in other words not an abstract function corresponding to Hollnagel’s cornerstones in itself, but an addition to his framework. The name of the associated abstract function in our approach is thus changed from Monitoring to Recognising to indicate that.

The third cornerstone accentuates the importance to be able to adapt the system in different ways based on what is anticipated to have a potential to become a problem in the future, what is recognised as critical or soon to be critical in the current situation or what is learnt to be a problem from experience. Hollnagel (2009) calls this responding, but includes adaptations to respond to and recover from specific events, as well as different ways to prevent/mitigate or prepare for an adverse event. As responding in the broader societal context connotes only the reactive response to a disaster situation, the name of the cornerstone is altered to Adapting. Generalised functions in a societal context corresponding to Adapting are Prevention and mitigation, Preparedness, Response and Recovery, where the two former are proactive and the two latter are reactive functions.

Hollnagel’s fourth cornerstone is Learning, as he clearly states that a ‘resilient system must be able to learn from experience’ (Ibid., p. 127). What failed in a specific disastrous event, as well as who is to blame for it, is not the focus here. Learning should instead be a continuous planned process focused on how the system functions, links between causes and effects, its interdependencies and so on (Ibid., pp. 129–30). In a societal context, learning from disturbances is often associated with evaluations of what happened and how various actors responded to the event in question. We call the generalised function Evaluation, which can be both proactive and reactive as it is not only possible to learn from what has happened in reality but also from counterfactual scenarios (Abrahamsson et al., 2010). It should however be noted that the actual learning in a system is very much dependent on the feedback loops from Evaluation to the other generalised functions (Figure 1.1), requiring that changes are made based on this input. There are many other aspects of learning that are not captured by the generalised function of Evaluation, but, as stated earlier, the generalised functions presented in this chapter are a selection for a specific societal context.

We argue that societal resilience is an emergent property determined by society’s ability to anticipate, recognise, adapt to and learn from variations, changes, disturbances, disruptions and disasters that may cause harm to what human beings value. Above, we have suggested how these four abstract functions can be transformed into generalised functions that are more concrete and provides more guidance on how to identify them in a societal context.

Figure 1.1 The abstract and generalised functions of societal resilience

Operationalising Societal Resilience

To meet the stated purpose of societal resilience, the system under study must have sufficient capacities for the abstract functions of anticipation, recognising, adapting and learning, which can be further specified by the generalised functions of risk assessment, forecasting, monitoring, impact assessment, prevention/mitigation, preparedness, response, recovery and evaluation (Figure 1.1). It is important to note that just as there may be many ways to describe any system (Ulrich, 2000, pp. 251–3), there may be different ways of concretising the four overall abstract functions for societal resilience. However, it is also important to note that there are dependencies between these functions making the functioning of one dependent on the output of the functioning of others, for example, for the public to be able to undertake the preparedness measure to take shelter for a coming cyclone necessitates warning information from forecasting or monitoring the weather (Figure 1.1).

To analyse societal resilience in a particular context, it is not enough to establish what functions that are needed to meet th...