2. Risk perspective
An important consideration in risk analyses concerns the conceptual understanding of risk and the corresponding risk perspective. In this article, risk is understood as uncertainty about and severity of consequences with respect to something that humans value (Aven, 2010). This implies adopting a constructivist basis for risk analysis, based on an interpretation of the views of experts and risk assessors about the possible occurrence of events of interest, in light of available evidence (Goerlandt and Montewka, 2015a). One fundamental issue in this approach is that the analysis should not be seen as a revelation of truth, but rather as a reflection of the most reliable knowledge available at a given time. The aim is to put forward an argument about the occurrence of events and consequences (Watson, 1994).
In line with this, the following risk perspective, i.e., the systematic approach to measure and describe risk, is adopted:
In the above, risk R is described by (“∼”) as a set of events A, consequences C, occurrence uncertainties OU and strength of evidence assessment SE. The risk analysis is performed based on the available background knowledge BK, which can consist of data, models, judgments and assumptions.
Uncertainties about the possible occurrence of the events A and consequences C are described. In the current analysis, subjective probabilities are applied to measure the occurrence of events A. Such probabilities are defined as an assessor’s degree of belief in light of the available evidence BK (Watson, 1994). For the consequences C, a categorical measurement scale is applied, in line with the accident severity classification adopted by IMO (IMO, 2013).
The assessment of strength of evidence of BK, i.e., the available background knowledge on which the assessment is based, is performed by means of an ordinal measurement scale suggested in Abrahamsen et al. (2014). The purpose of this assessment is to qualitatively convey how warranted the measurements of A and C are.