Analysis of Optimization Experiments

• Autores: V. Roshan Joseph, James Dillon Delaney
• Localización: Journal of quality technology: A quarterly journal of methods applications and related topics, ISSN 0022-4065, Vol. 40, Nº. 3, 2008, págs. 282-298
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen

  • Industrial experiments are typically analyzed by identifying effects with a five percent level of significance and then optimizing the model containing only those effects. The risk of using this approach is demonstrated and a methodology is proposed using the easily-specified practical significance level. In addition, empirical Bayes estimation is used to provide shrinkage estimates of the effects. Connections are established between the proposed approach and Taguchi’s beta coefficient method. Advantages of the proposed method are demonstrated with real examples and simulation.