随机-认知混合不确定性下稳健性评价新指标
Establishing New and Robust Evaluation Indexes under Aleatory and Epistemic Uncertainties

在随机不确定性和认知不确定性信息混合并存的产品稳健设计中，现有的基于概率理论、模糊集理论、区间理论等建立的质量稳健性度量指标适用性受到限制。针对此问题，将各类不确定性在证据理论的统一框架下表征和处理，构建了产品质量指标的证据理论模型和不确定性量化传递方法，从随机不确定性和认知不确定性两个维度提出了产品质量稳健性的评价指标和评价方法。实例仿真表明，该方法可克服现有方法的不足，对产品质量稳健性的评价更加全面、合理。
In designing robust products that have aleatory uncertainty and epistemic uncertainty, the existing theories of probability, fuzzy set, interval and so on are not enough to establish new and robust evaluation indexes. To solve this problem, various types of uncertainty were described and processed according to the evidence theory. Then the evidence model for product quality evaluation indexes was established; the uncertainty quantification and propagation methods were developed. Under the aleatory and epistemic uncertainties, we proposed the robust indexes and methods for evaluating product quality. The simulation results show the effectiveness of the proposed indexes and methods