Why risk analysis is difficult, and some thoughts on how to proceed

A Conceptual Framework to Classify and Manage Risk, Uncertainty and Ambiguity: An Application to Energy Policy

Global issues are such that we should assess and manage a variety of risks and uncertainties. Due to increasing world complexity, the development of an adequate and innovative conceptual framework, anchored in the literature, is required. This article contributes to this effort with an approach particularly relevant to decision-makers dealing with threats of different natures, limited heterogeneous information, and experts’ assessments tainted by doubts. Our approach is based on two pillars: 1) An “acuity scale”, based on the probability of the occurrence of an event, its impact and the experts’ degree of doubt; 2) A taxonomy focused on the concepts of risk, uncertainty, gamble and butterfly ambiguity. Accordingly, we present in a second step the major management implications of such approach. Global policy trends (e.g., sustainability transition) put energy sector decision-makers at the forefront of risk and uncertainty management. Consequently, we carry out a case study focused on Swiss energy policy since the 1980s, including its inception, the turnaround provoked by the Fukushima accident, and the government’s 2050 energy strategy. Our investigation shows that the proposed conceptual framework allows for the development of an original analysis of the main drivers that influence governmental policies and stakeholder strategies.

Setting the most robust effluent level under severe uncertainty: application of information-gap decision theory to chemical management

Decisions in ecological risk management for chemical substances must be made based on incomplete information due to uncertainties. To protect the ecosystems from the adverse effect of chemicals, a precautionary approach is often taken. The precautionary approach, which is based on conservative assumptions about the risks of chemical substances, can be applied selecting management models and data. This approach can lead to an adequate margin of safety for ecosystems by reducing exposure to harmful substances, either by reducing the use of target chemicals or putting in place strict water quality criteria. However, the reduction of chemical use or effluent concentrations typically entails a financial burden. The cost effectiveness of the precautionary approach may be small. Hence, we need to develop a formulaic methodology in chemical risk management that can sufficiently protect ecosystems in a cost-effective way, even when we do not have sufficient information for chemical management. Information-gap decision theory can provide the formulaic methodology. Information-gap decision theory determines which action is the most robust to uncertainty by guaranteeing an acceptable outcome under the largest degree of uncertainty without requiring information about the extent of parameter uncertainty at the outset. In this paper, we illustrate the application of information-gap decision theory to derive a framework for setting effluent limits of pollutants for point sources under uncertainty. Our application incorporates a cost for reduction in pollutant emission and a cost to wildlife species affected by the pollutant. Our framework enables us to settle upon actions to deal with severe uncertainty in ecological risk management of chemicals.

Integrating risk and resilience approaches to catastrophe management in engineering systems

Recent natural and man-made catastrophes, such as the Fukushima nuclear power plant, flooding caused by Hurricane Katrina, the Deepwater Horizon oil spill, the Haiti earthquake, and the mortgage derivatives crisis, have renewed interest in the concept of resilience, especially as it relates to complex systems vulnerable to multiple or cascading failures. Although the meaning of resilience is contested in different contexts, in general resilience is understood to mean the capacity to adapt to changing conditions without catastrophic loss of form or function. In the context of engineering systems, this has sometimes been interpreted as the probability that system conditions might exceed an irrevocable tipping point. However, we argue that this approach improperly conflates resilience and risk perspectives by expressing resilience exclusively in risk terms. In contrast, we describe resilience as an emergent property of what an engineering system does, rather than a static property the system has. Therefore, resilience cannot be measured at the systems scale solely from examination of component parts. Instead, resilience is better understood as the outcome of a recursive process that includes: sensing, anticipation, learning, and adaptation. In this approach, resilience analysis can be understood as differentiable from, but complementary to, risk analysis, with important implications for the adaptive management of complex, coupled engineering systems. Management of the 2011 flooding in the Mississippi River Basin is discussed as an example of the successes and challenges of resilience-based management of complex natural systems that have been extensively altered by engineered structures.