Risk and vulnerability analyses can potentially play important roles in both the mitigation and preparedness phase of an emergency (see Figure 3-1). Conducting risk and vulnerability analyses, then, is one component (of many) in a proactive approach towards emergency management and safety. The purpose of the analysis has large influences on how it will and should be carried out. Available resources, the competence of the analysts and much more will and should affect the choice of method, the comprehensiveness of the study and so on. In the present thesis, a distinction is made between two overriding purposes, or roles, that risk and vulnerability analyses can have: a decision-oriented and a process-oriented role (see Figure 3-2).
Analysis process System information
Methodology
Analysis output – e.g. risk measures
Decision-oriented Process-oriented
Figure 3-2. Illustration of the analysis process and whether the focus of an analysis that has been carried out is on the process itself or the output of the process, adapted from Rouvroye and van den Bliek (2002).
A purely decision-oriented approach is only concerned with what the analysis is able to produce, for example in terms of quantification of risk by use of some risk
measure. The actions performed during the analysis process are only important to ensure the quality and validity of the analysis output. Many contemporary applications of risk analysis (especially in engineering) are directed toward providing input to decision-making, such as investigating whether a certain activity is acceptable from a risk perspective or which actions are most cost-effective in reducing a risk. An example of a decision-oriented approach is put forward by Cox when describing traditional Quantitative Risk Assessment (QRA) as “providing a logical framework and a systematic procedure for organizing and applying scientific and engineering knowledge to improve “rational” (consequence-driven) decision making when the consequences of alternative decisions are uncertain” (Cox, 2007).
Thus, ensuring that the best available knowledge is used as input to decision-making will help make the decisions as “rational” and good as possible.
A purely process-oriented approach is not concerned with what the analysis will produce, for example in terms of risk estimations; instead the focus is on the activities that are performed during the analysis process per se. Most often a process-oriented analysis is conducted by people that themselves are part of the
“system” being analysed. Conducting risk analyses are then believed to enhance people’s risk awareness and encourage them to reflect upon their actions and change those actions that may contribute to the overall risk. Furthermore, in many cases the people that participate in the analysis process play an essential role in shaping the risk in the system, for example by acting as responders in the acute phase of an emergency or as operators in the control room of a chemical plant.
Conducting risk analyses with these people can function as a catalyst for building trust and friendship ties among the participants leading to better preconditions for adequate communication and coordination prior to and during an emergency. In addition, common mental models about risks and emergencies can be created, which also can lead to positive effects.
Process-oriented approaches can for example be found in emergency preparedness and planning. Many researchers argue that the value of emergency planning does not lie in the generation of an outcome in terms of a written plan, but in the processes of producing the plan (Quarantelli, 1997; Perry and Lindell, 2003).
According to McLoughlin this is because “the officials who are responsible for emergency operations have spent time determining which official will do what and how operations will be coordinated” (McLoughlin, 1985). Thus, the emergency planning in itself is a means of reducing the overall risk for the system in question, by increasing the capabilities to respond to perturbations. The same can be said about conducting risk analyses with a process-oriented approach – the analysis process is believed to reduce the level of risk by increasing the risk awareness and the response capabilities of the people “in the system”.
In conducting interviews with people participating in risk analysis processes, Busby and Hughes (2006) found that many persons were actually sceptical about the ability of risk assessment methods to produce valid, objective risk measures; instead they argued that the value of the risk analysis “lay in helping people who generate the risk, or find themselves responsible for risk, reflect on their practices”.
Increased focus on the analysis processes per se is likely to be preferred as systems become more complex. A parallel can be drawn to Peter Checkland’s distinction between hard and soft systems (Checkland, 1993). In hard systems problems are well-defined and there is a general agreement on which objectives to pursue. The problem then dwells down to choosing the best possible future strategy. In softer systems, on the other hand, problems are ill-defined and it is not even clear which objectives that should be pursued. Checkland argues that in such circumstances the purpose of making use of methods to perform various types of analyses rather is to structure a discussion about the problems (in the context of this thesis these problems concern risky activities) than to generate any concrete input to decision-making. The line of reasoning can be applied to risk analyses as well; as systems become “softer” and more complex it is increasingly difficult to generate valid and objective measures of risk as input to concrete decisions. Instead, the analyses in themselves are seen as risk reducing activities.
Risk and vulnerability analyses that are carried out in practice can rarely be classified as either a purely decision-oriented or a purely process-oriented approach.
Instead, they often involve a mix of the two types of purposes, although the purpose is often leaning towards the one or the other. For example, in the guidelines published by the Swedish Emergency Management Agency (SEMA) regarding risk and vulnerability analyses that Swedish authorities, municipalities and county councils are obliged to conduct, both purposes are stressed (SEMA 2006a; 2006b). On the one hand, the risk and vulnerability analysis is the tool for knowledge attainment that subsequently can be used for deciding which risk and vulnerability reducing measures are most effective, or at least effective “enough” to implement. On the other hand it is stressed that in conducting the analyses the aim is to increase the risk awareness of the participants and to create and maintain a safety culture in the organisations. A type of risk analysis where the decision-oriented purpose is much accentuated is the risk analyses conducted in the context of the Swedish land-use planning process. In the Swedish Plan and Building Act one (of many) requirements is that when someone is to exploit land for housing, offices or other activities, planning must be carried out with regards to people’s health (SFS 1987:10). Often this requirement is interpreted in the sense that a risk analysis has to show that the risk for the plan area, exposed by surrounding hazards, is acceptably low. To show risk acceptance, external risk experts are usually
consulted who conduct quantitative risk analyses that can be compared to established acceptance criteria and a decision regarding the suitability of the proposed land exploitation can be taken. When a decision has been taken the risk analytic effort ends, i.e. no continuing process is initiated with the purpose of for example increasing the safety culture of the involved actors.
The demands one can put on an analysis and the method used for the analysis are dependent on where on the continuum, shaped by the two extreme roles, a specific analysis can be positioned. The validity of the method, in terms of the correspondence between any estimations and reality, is not equally crucial if the analysis is very process-oriented rather than decision-oriented. A process-oriented analysis must instead ensure that discussion and deliberations throughout the analysis promote the purpose of the analysis, e.g. if the purpose is to increase trust relations among people having a role in the emergency response, it is important that the “right” people participate in the analysis.
In the context of emergency management both roles of risk and vulnerability analysis are often very important. In Swedish municipalities, for example, it is common, but not always the case, that the ones conducting the analyses are also acting in the response phase of an emergency. Therefore, analyses can constitute a way for these to interact and create trust relations among themselves. At the same time there is a clear need to make decision regarding which resources and skills are needed, whether the municipality should start cooperating with neighbouring municipalities, whether structural mitigation measures should be implemented and so on. Of course, such a decision can be made without formal analysis; however, the possibility to make a rational decision is definitely increased if a properly made analysis is used as input to the decision.
The methods being developed in the present thesis primarily aims to provide concrete input to decision-making. It is thus important that the analysis method ensures that the outcome, given that the analysis is performed appropriately, is of high quality and validity. However, this as a primary aim does not exclude the methods from being useful from a process-oriented perspective as well.