Failure mode and effect analysis (FMEA) is a prospective risk analysis instrument applied in various fields for eliminating known or potential failures in products. However, conventional FMEA has suffered from plenty of criticisms that limit its effectiveness. Due to fuzzy and imprecise information often exist in risk assessments, numerous modified FMEA methods based on fuzzy theory are employed to improve the classical FMEA. However, the previous fuzzy-based FMEA approaches still have some drawbacks, e.g., requisition of extra information, no ability in handling hesitation of individual and subjectivity of group simultaneously, consideration of only three risk elements, and deeming the experts’ judgments are rational among most of them. Hence, this paper develops a novel concept of interval-valued intuitionistic fuzzy rough number (IVIFRN) by utilizing the merit of interval-valued intuitionistic fuzzy set in handling fuzziness and hesitation of individuals’ judgments and the benefit of rough number in manipulating imprecision and subjectivity of group’s assessments and presents a new FMEA model based on IVIFRN. Then, maintenance is regarded as a new risk aspect and to further establish a risk evaluation structure containing eight risk elements. Moreover, the experts’ bounded rationality and risk elements’ importance are determined by two synthetic weighting methods considering both subjective and objective information. Finally, a real case is performed to illustrate the effectiveness of the developed model.

故障模式和影响分析（FMEA）是应用于各种领域的前瞻性风险分析工具，用于消除产品中已知或潜在的故障。但是，传统的FMEA遭受了许多批评，限制了其有效性。由于风险评估中经常存在模糊和不精确的信息，因此采用了许多基于模糊理论的改进FMEA方法来改进经典FMEA。但是，以前基于模糊的FMEA方法仍然存在一些缺点，例如，需要额外的信息，无法同时处理个人的犹豫和群体的主观性，仅考虑三个风险因素以及认为专家的判断是合理的。大多数。因此，本文利用区间直觉模糊集的优点来处理个体判断的模糊性和犹豫性，以及利用粗糙数在处理群体不精确性和主观性方面的优势，提出了区间直觉模糊粗糙数（IVIFRN）的新概念。评估并提出基于IVIFRN的新FMEA模型。然后，维护被视为新的风险方面，并进一步建立包含八个风险要素的风险评估结构。此外，专家的有限理性和风险要素的重要性是通过考虑主观和客观信息的两种综合加权方法确定的。最后，通过一个实际案例来说明所开发模型的有效性。