A statistical analysis of failure mode and effects analysis validated by experts for heating ventilation and air conditioning systems in positron emission tomography radiopharmaceutical centers

The reliability of heating, ventilation and air conditioning (HVAC) systems is critical for Good Manufacturing Practice (GMP) compliance and radiation protection in PET radiopharmaceutical production centers. This study applies a Failure Mode and Effects Analysis (FMEA) combined with a Delphi expert panel to identify and prioritize HVAC failure modes in a PET facility, and provides a statistical validation of the resulting risk model. A panel of 10 experts from 8 Latin American countries (Mexico, Uruguay, Argentina, Cuba, Dominican Republic, Costa Rica, Peru and Chile) evaluated 15 HVAC failure modes using 1–10 scales for occurrence (O), severity (S) and detection (D), and a risk priority number (RPN) was calculated as $$RPN = O \times S \times D$$. Descriptive statistics, coefficients of variation (CV), Shapiro–Wilk tests, Pearson correlations, independent t-tests and interquartile ranges were used to assess the stability, construct validity, discriminant validity and robustness of the method. Results show low relative variability of expert ratings, strong and significant correlations between RPN and O and S, significant differences in O and S between critical (RPN $$\ge 150$$) and non-critical modes, and a balanced distribution of RPN across four risk categories. These findings support the Delphi-based FMEA as a statistically robust tool to prioritize HVAC-related risks in PET radiopharmaceutical centers.