ASSESSING EXTERNAL THREATS TO STRUCTURES USING LIMITED STATISTICAL DATA: AN APPROACH BASED ON DATA RESAMPLING
提出用自助重采样法从有限数据中计算结构受外部威胁的损伤概率置信区间，帮助工程师在数据稀缺时评估风险。
The paper deals with an estimation of risk to structures posed by extreme, dangerous phenomena called in brief the external threats. It is considered how to calculate risk values when a limited amount of data on actions imposed by these phenomena is available. The key methodology suggested in the paper for estimating the risk is the so‐called bootstrap resampling, known also as statistical or Efron's resampling. The paper presents a procedure allowing to apply the limited data to calculating bootstrap confidence intervals for probabilities of damage which can be caused by the actions. The application of the procedure is based on the assumption that the limited data can be expressed in the form of statistical sample which possesses the property of representativeness. It is discussed how to incorporate the confidence intervals in an expression of the risk induced by external threats. The proposed procedure can be viewed as a way of utilising limited and often very expensive data gained from experiments imitating occurrences of external threats. Findings described in this paper can be applied to design the structures for the so‐called accidental situations.