Compara mètodes

	Anàlisi de sensibilitat amb gràfic de control	Gràfic de control robust
Camp	Disseny experimental	Disseny experimental
Família	Process / pipeline	Process / pipeline
Any d'origen	Integration practice documented from the 1990s onward	1989–1997 (foundational period)
Autor original	Rooted in Shewhart (control charts, 1920s) and Saltelli et al. (global sensitivity analysis, 1990s–2000s); integration practice developed in quality engineering literature	David M. Rocke; L. G. Tatum (key contributors)
Tipus	Hybrid analytical framework	Statistical process monitoring technique
Font seminal	Saltelli, A., Ratto, M., Andres, T., Campolongo, F., Cariboni, J., Gatelli, D., Saisana, M., & Tarantola, S. (2008). Global Sensitivity Analysis: The Primer. Wiley. ISBN: 978-0470059975	Tatum, L. G. (1997). Robust estimation of the process standard deviation for control charts. Technometrics, 39(2), 127–141.
Àlies	SA-SPC integration, control chart sensitivity analysis, SPC sensitivity assessment, sensitivity-enhanced control charting	robust Shewhart chart, outlier-resistant control chart, robust SPC chart, distribution-free control chart
Relacionats	6	6
Resum	Sensitivity analysis integrated with control charting evaluates how uncertain or varying inputs — such as sample size, subgroup frequency, distribution assumptions, or measurement error — affect the detection performance of a statistical process control chart. By quantifying which parameters most strongly influence chart metrics such as the average run length (ARL) or false alarm rate, engineers can design more robust monitoring schemes and understand where control chart conclusions are fragile.	A robust control chart replaces the classical mean and standard deviation estimators in a Shewhart-style chart with resistant alternatives — such as the median and median absolute deviation (MAD) — so that a small fraction of outliers or non-normal process data cannot distort the control limits. The approach preserves the real-time monitoring logic of standard control charts while protecting against inflated or deflated limits caused by contaminated Phase I reference data.