Abstract
‘Bad’ data resulting from ill‐designed experiments or in‐production/after‐production monitoring require a careful use of proper statistical techniques for their analysis. Prior exploration of the available data is of paramout importance especially if we wish to apply off‐line quality control techniques such as the Taguchi method. A stepwise approach is proposed involving data analysis and straightforward significance tests, which can ensure statistically valid and useful conclusions.

This publication has 9 references indexed in Scilit: