Failure Modes and Effects Analysis (FMEA) is a method that has long proven its worth in various fields. It consists of evaluating and prioritizing risks to better control their potential impacts. Since no expert can proclaim a perfect knowledge of all the systems assessed, the evaluation process considers several experts’ opinions. Hence, how a consensus can be drawn? In addition, the traditionally used crisp ordinal numbers may be hardly assignable to describe faithfully failure modes in classical FMEA. This work combines FMEA and a new aggregation operator for Interval-Valued Intuitionistic Fuzzy Numbers (IVIFN) to overcome both shortcomings.