Is design of experiments really used? A survey of Basque industries

Abstract

A new survey is presented concerning the knowledge and use of the design of experiments technique (DoE) within industry in the Basque Country, a region recognised throughout Europe for its quality management. The survey was carried out within manufacturing companies, yielding a response rate of 18%. Results show that 94% of companies undertake experimentation; most of them use one-factor-at-a-time strategies, and only 20% of those follow a pre-established statistical methodology. Survey results show that research and development and manufacturing make up 85% of DoE use. Furthermore, results show that lack of knowledge about general statistics is commonplace and only 31% of companies claim to be knowledgeable about DoE. In addition, although Taguchi methods are well known among companies, only 7% apply this method. Despite every effort by specialists in quality and statistics, DoE has yet to be applied as widely as it could and should be.

Keywords:

• design of experiments
• empirical study
• experimental methodology
• industry
• questionnaire

Acknowledgements

We would like to thank the anonymous referee for helpful comments, which improved the paper. We are also grateful to Sonia Romero for the construction of the web page.

Notes

See: www.tecnun.es/personal/mtanco/encuesta.asp

With p=q=0.5 and a significance level of 95%.

We did not send the questionnaire to industries with fewer than 50 employees.

Weight for small industries is 0.71, for medium industries is 0.20, and for large industries is 0.09.

Using the test for big samples with a significance level of 95% and assuming complete independence of samples.

An analysis for comparing two big samples states with a confidence level of 99% that no significant difference exists. We must appreciate that slight differences exist between the populations considered.

Six Sigma is an organised and systematic method for strategic process improvement and new product and service development that relies on statistical methods and the scientific method to make dramatic reductions in customer defined defect rates (Brady and Allen 2006).