Hazard Event Frequency, Distribution of Objects and Success Frequency of Avoiding Hazards
暂无分享,去创建一个
The chapter comprises a set of approaches for computing overall hazard event scenario probabilities by providing methods to generate event frequency rates (basic event rates), exposure distributions and rates of successful avoidance of hazard events in case of events. The last two factors can be understood as sample conditional probabilities that lead to event scenarios in case the scenario occurs. However, when using historical event data for a first estimate, all past influences on overall probabilities of hazard scenarios are already taken account of. Therefore it is difficult to generate event data from historical data that does not already incorporate some type of prevention, protection and even mitigation. The chapter reviews a step-wise approach to determine event frequencies from empirical-historical records, which lays more emphasis on control and plausibility checks when compared with the database-driven empirical-historical risk analysis approaches. As an alternative approach, the application of fault tree analysis is mentioned. As understood in the present case, this is only an alternative, if the event is due to a technical (sub)system failure that can be quantified using basic failure rates within the deductive fault tree scheme. In the same way, the present text does not discuss attack trees in detail. The chapter discusses a real word scenario, which allows to introduce discrete objects along with local discrete time-dependent exposure quantities for individuals and groups. This approach is a more hands-on intuitive approach when compared to the more flexible use of object densities. Local absolute numbers can be first choice, if the number of objects at risks is known and if they are strongly localized. The use of discrete occupation numbers allows to compute local individual and group risks as well as individual and collective non-local risks, e.g. risks of single persons or groups at a single place or all places visited within one week. Also total average risks are computed. The approach is formalized showing that average total risks somewhat oversimplify the results when compared to a set of collective risk numbers. This is also a motivation for the F-N curve assessment. In the case of technical safety, when assessing the need for further technical efforts to control risks, in particular when employing safety functions, the options for the reduction of effects of hazards contribute to the overall assessment. This is an example of a risk-driven approach that takes systematically account of the successes frequency of avoiding hazards. To further clarify the definition of the success (conditional) probability of avoiding hazards, the chapter introduces four examples and discusses their validity.