Can we understand complex systems in terms of risk analysis?

The concept of integrated operations (IO) introduces new ways of operations in the offshore petroleum industry. IO is often characterized by virtual decision arenas where many safety critical decisions are supported and made by distributed actors with different rationalities and responsibilities. This will challenge more traditional decision processes in several ways and it has been questioned whether the risk analysis framework can handle what some authors denote as emerging and escaping risks. Complexity is often considered as a source of such risks. In this paper risk is defined as uncertainty regarding occurrence and severity of undesired events. Next a variety of techniques for structuring and quantifying uncertainty are listed. To approach complexity it is proposed to identify a set of complexity characteristics in relation to the accidental scenarios to be undertaken in the analysis. This enables uncertainty due to complexity to be approached within the same framework as that used to cope with other sources of uncertainty. The important steps in such an integrated risk and complexity analysis are listed, and some of these steps are discussed in the light of examples relevant to IO.