Abstract
Commonly, the efficacy of new FMECA methods is conducted through
qualitative comparisons between rankings; for a small number of failure
modes, this approach is suitable but can become unpractical or lead to
misleading results for more extensive problems. This fact motivated us
to introduce an alternative approach to compare different FMECA methods
based on agreement metrics that allow the statistical comparison between
rankings generated by independent raters. Despite its relevance, the
application of agreement coefficients is limited in the FMECA context.
The proposed approach considers the agreement assessment between
different methodologies used in FMECA analysis (Risk Priority
Isosurfaces RPI, VIKOR, ITWH, and Type-II Fuzzy Inference System) when
applied to a study case regarding blood transfusion widely used in the
literature for benchmarking and consisting of eleven failure modes. We
selected the RPI methods as a reference to compare the other forenamed
methods. Results show that our agreement coefficient-based comparison
approach proves effective for the statistical comparison of different
FMECA methods instead of the rankings qualitative comparison.