Open problems and issues in optimal design

Abstract

In recent years, there has been an increasing emphasis in the design of experiments literature on optimal design. Many authors have recommended that optimal design of experiments should be used in lieu of classical designs. We believe the optimal design approach is widely applicable and encourage its increasing adoption in industrial applications. However, for some industrial concerns the adoption of optimal designs is not widespread. This article provides some reasons why practitioners may be reluctant to adopt optimal designs for problems. We also provide open questions that have not been adequately addressed in the literature on optimal design of experiments.

Keywords:

• classical design of experiments
• factorial experiments
• replication
• response surface methodology
• robustness
• sequential experimentation

Acknowledgments

We are grateful for the feedback and input from several referees and the editor, even if they did not necessarily agree with everything that we have written. We believe their probing questions and insights have helped improve this article.