In steel industries a need exists for the reduction of variation in the ele
ment concentrations. This need is triggered by an increasing demand for hig
h quality steel products by clients. The results of a large-scale experimen
t concerning the identification and quantification of sources of variation
in the production, sampling and analysis of steel are presented. The result
s are obtained by means of a strategic approach that consists of six steps.
By means of this strategic approach, insight is obtained in the build-up o
f the total variation. This knowledge can be used to reduce the influence o
f these factors that have a major impact on the total variation. Attention
is focused on estimating the magnitude of sources of variation apparent in
sampling of certain stages of the steel making process and in analysis of s
teel samples by means of spark emission spectrometry.