Comparer des méthodes

	Déploiement de la Fonction Qualité Bayésien	Déploiement Robuste de la Qualité (Robust Quality Function Deployment)
Domaine	Plans d'expériences	Plans d'expériences
Famille	Process / pipeline	Process / pipeline
Année d'origine	QFD: 1966–1972; Bayesian QFD extensions: 2000s–present	2000s (robust extensions of QFD originating 1966)
Auteur d'origine	Yoji Akao (QFD); Bayesian extension developed by multiple researchers including Fung, Tang, and colleagues	Extension of Yoji Akao's QFD (1966); robust adaptation by Fung, Kwong and others (early 2000s)
Type	Probabilistic customer-driven design planning method	Hybrid quality-engineering planning method
Source fondatrice	Tang, J., Fung, R. Y. K., Xu, B., & Wang, D. (2002). A new approach to quality function deployment planning with financial consideration. Computers & Operations Research, 29(11), 1447–1463.	Fung, R. Y. K., Tang, J., & Tu, Y. (2002). Modeling of quality function deployment planning under resource allocation constraints. Computers & Industrial Engineering, 43(1–2), 313–328.
Alias	Bayesian QFD, Probabilistic QFD, Bayesian House of Quality, Bayesian Voice of the Customer Analysis	Robust QFD, Uncertainty-tolerant QFD, Fuzzy-robust QFD, Robust House of Quality
Apparentées	5	4
Résumé	Bayesian Quality Function Deployment (Bayesian QFD) integrates Bayesian probabilistic inference into the classical House of Quality framework to handle uncertainty in customer preference data and relationship matrices. By expressing relationship weights and importance ratings as probability distributions rather than point estimates, it propagates uncertainty through the planning process and yields more defensible engineering prioritization decisions under incomplete or conflicting customer information.	Robust Quality Function Deployment (Robust QFD) extends the classical House of Quality framework by explicitly modeling uncertainty and variability in customer requirements, perception ratings, and engineering correlation judgments. Instead of treating inputs as crisp single-point values, it applies fuzzy sets, interval analysis, or Taguchi-inspired robustness techniques to ensure that the resulting design targets remain stable and customer-satisfying even when inputs are imprecise or fluctuating.