Probabilistic risk assessment (PRA) is a systematic and comprehensive methodology to evaluaterisks associated with a complex engineered technological entity (such as anairliner or anuclear power plant) or the effects of stressors on theenvironment (probabilistic environmental risk assessment, or PERA).[1]
Risk in a PRA is defined as a feasible detrimental outcome of an activity or action. In a PRA, risk is characterized by two quantities:
Consequences are expressed numerically (e.g., the number of people potentially hurt or killed) and their likelihoods of occurrence are expressed asprobabilities orfrequencies (i.e., the number of occurrences or the probability of occurrence per unit time). The total risk is theexpected loss: the sum of the products of the consequences multiplied by their probabilities.
The spectrum of risks across classes of events are also of concern, and are usually controlled in licensing processes – it would be of concern if rare but high consequence events were found to dominate the overall risk, particularly as these risk assessments are very sensitive to assumptions (how rare is a high consequence event?).
Probabilistic risk assessment usually answers three basic questions:
Two common methods of answering this last question areevent tree analysis andfault tree analysis – for explanations of these, seesafety engineering.
In addition to the above methods, PRA studies require special but often very important analysis tools likehuman reliability analysis (HRA) andcommon-cause-failure analysis (CCF). HRA deals with methods for modelinghuman error while CCF deals with methods for evaluating the effect of inter-system and intra-system dependencies which tend to cause simultaneous failures and thus significant increase in overall risk.
One point of possible objection interests the uncertainties associated with a PSA. The PSA (Probabilistic Safety Assessment) has often no associated uncertainty, though inmetrology anymeasure shall be related to a secondarymeasurement uncertainty, and in the same way any mean frequency number for arandom variable shall be examined with thedispersion inside the set of data.
For example, without specifying an uncertainty level, the Japanese regulatory body, the Nuclear Safety Commission issued restrictive safety goal in terms of qualitative health objectives in 2003, such that individual fatality risks should not exceed 10−6/year. Then it was translated in a safety goal for nuclear power plants:[2]
The second point is a possible lack of design in order to prevent and mitigate the catastrophic events, which has the lowest probability of the event and biggest magnitude of the impact,[2] and the lowest degree of uncertainty about their magnitude. Acost-effective of thefactor of safety, contribute to undervaluate or completely ignore this type of remote safety risk-factors. Designers choose if the system has to be dimensioned and positioned at the mean or for the minimum level of probability-risk (with related costs of safety measures), for beingresilient androbust in relation to the fixed value.
Such external events may benatural hazard, including earth quake and tsunami, fire, and terrorist attacks, and are treated as a probabilistic argument.[2] Changing historical context shallcondition the probability of those events, e.g. a nuclear program oreconomic sanctions.
