private (national and international) organisations, leading to great challenges for analysing them from a holistic point of view. Several research initiatives exist with the purpose of modelling critical infrastructures from a holistic perspective, e.g.
Haimes (2001), Dudenhoeffer, Permann et al. (2002), Brown , Beyler et al.
(2004), Lee, Mitchell et al. (2004), Newman, Nkei et al. (2005), Min, Beyler et al.
(2007); however, overall the state-of-the-art is still in a quite rudimentary stage (Little, 2002; Rinaldi, 2004).
It is important to note that the model described and discussed above is a crude representation of a real emergency. A more detailed model would for example divide the response to an emergency into different temporal phases, such as detection, reaction, mobilization etc. However, as mentioned earlier, the crude representation is intentional since the purpose of the model was to represent emergencies generally and capture the most important characteristics of these.
3.6 Reflections on the model: implications for risk
picture of the risks and vulnerabilities in a community, all these factors and interactions therefore need to be studied.
The complexity of an emergency situation constitutes great challenges to risk and vulnerability analyses. There are two main reasons for this: the many dependencies and interdependencies, and the importance of human actions and responses. The fact that there are many dependencies and interdependencies implies that it is difficult to partition an analysis into sub-analyses. For example, analysing emergency response capabilities and assistance needs separately may lead to misleading results. This is because the two factors are not independent, since if the emergency response actors can satisfy the assistance needs in a timely manner, they may prevent future ones from arising. On the other hand, if the emergency response actors are not able to satisfy assistance needs, new needs may be created.
There are of course pragmatic reasons for separating analyses into several sub-analyses, such as separating analyses of individual infrastructures, analyses of social vulnerability and assistance needs, analyses of demands on emergency response actors, analyses of emergency response capabilities, and analyses of structural barriers, since analysing “everything at the same time” may be a too arduous task.
However, then the interactions between the different units of analyses must also be studied in order to integrate the findings into an overall picture of the risks and vulnerabilities.
The other reason for the great complexity is the fact that human actions and behaviour play such an important role in emergencies. In the traditional technical risk analyses, human behaviour basically did not have a role in the analyses; it was rather only the technical aspects of the systems that were studied in the analyses.
More recently, quantitative risk analyses started to incorporate human aspects and behaviours into the analyses, but only in terms of the behaviour of operators in for example nuclear power plants. The field of Human Reliability became a great contributor to risk analysis by first providing methods for estimating probabilities of human errors, and later to investigate how the performance conditions and the context affect human actions (Fujita and Hollnagel, 2004). However, still the main focus is on avoiding the occurrence of accidental events, not so much on studying how these events subsequently are responded to by affected systems and emergency response actors.
In taking a broader view of emergencies it is clear that human and organisational behaviour play important roles for emergency management, not primarily as sources of errors, but as sources of resilience and capability to minimise impacts, i.e. people and organisations act as attenuators during an emergency. Descriptive studies of emergencies and crises, such as case studies, can provide important
insight into which characteristics people and organisation should attain in order to initiate and maintain a good response to emergencies. Often these studies emphasize the importance of factors such as adaptability, flexibility, improvisation and creativity, since in many cases things do no not develop according to what was expected on beforehand. Findings from the emergency management field, therefore, can provide vital input to risk and vulnerability analyses by helping the analyst to understand how different characteristics of people and organisations affect performance during a response.
To conclude this chapter, it is clear that taking a broader view of emergencies and accounting for more factors in risk and vulnerability analyses will lead to a higher degree of complexity. Human actions and performance are often difficult to predict, meaning that it is difficult on beforehand to know how an emergency will evolve. Another feature that adds to the complexity is that in an emergency situation there are often several value dimensions that are relevant, e.g. number of fatalities is often not the only dimension of concern. In attaining a complete picture of risk, therefore, one must model how all relevant value dimensions are affected in an emergency, clearly requiring a greater modelling effort. Much research is definitely needed in order to satisfy these research demands.
4 Approaches to understand, model and analyse complex systems
The present thesis is essentially about the problem of analysing risks in and vulnerabilities of complex systems. In order to do that, we must explore different ways of making sense of such systems. In the research literature there are several approaches and suggestions available. In this section an overview of three general and broad approaches will be given, which have been influential to the research presented here. The three approaches are:
¾ Systems theory and the systems movement,
¾ Complexity theory and the theory of complex adaptive systems,
¾ Network theory and analysis.