Comparer des méthodes

	Bayesian Six Sigma DMAIC	Conception bayésienne d'expériences
Domaine	Plans d'expériences	Plans d'expériences
Famille	Process / pipeline	Process / pipeline
Année d'origine	1986 (DMAIC); Bayesian integration circa 1995–2010	1956 (foundational); formalized 1970s–1990s
Auteur d'origine	Six Sigma: Bill Smith / Mikel Harry at Motorola (1986); Bayesian integration developed in quality literature through 1990s–2000s	Lindley (1956); Chaloner & Verdinelli (1995) landmark review
Type	Hybrid quality-improvement framework	Bayesian optimal experimental design
Source fondatrice	Pan, J.-N. (2007). Bayesian approach to estimation of process capability indices in process quality assurance. Quality and Reliability Engineering International, 23(1), 3–14.	Chaloner, K., & Verdinelli, I. (1995). Bayesian Experimental Design: A Review. Statistical Science, 10(3), 273–304.
Alias	Bayesian DMAIC, Bayesian Six Sigma, B-DMAIC, Probabilistic Six Sigma DMAIC	Bayesian DOE, Bayesian optimal design, Bayesian experimental design, BDE
Apparentées	6	3
Résumé	Bayesian Six Sigma DMAIC integrates Bayesian statistical inference into the classical Define-Measure-Analyze-Improve-Control quality-improvement framework. Rather than relying solely on frequentist hypothesis tests and point estimates, it incorporates prior knowledge — from expert judgment, historical production data, or pilot studies — and updates beliefs about process parameters as new data arrive. The result is a more adaptive, uncertainty-aware approach to reducing defects and improving process capability, particularly valuable when sample sizes are small or prior domain knowledge is rich.	Bayesian design of experiments selects experimental runs by maximising a utility function — typically the expected information gain — computed over prior beliefs about model parameters. Unlike classical design, which optimizes algebraic criteria such as D-optimality under fixed assumptions, Bayesian DOE incorporates prior knowledge and uncertainty about the system, yielding designs that are optimal in expectation across all plausible parameter values.