Can Design of Experiments (DOE) Be Simplified?

Mark J. Anderson and Patrick J. Whitcomb just published the the third edition of their innovative book, DOE Simplified: Practical Tools for Effective Experimentation, and I had the chance to speak with Mark about its content. I asked him: "Nothing could be simpler than the traditional scientific method so why abandon this old-school approach for experimentation?" Here is Mark's complete answer:

It is true that that the scientific methods we all learned in school could not be any easier -- simply hold everything in your system constant other than changing only one factor at a time (OFAT). For example, It works very well for demonstrating scientific principles in a chemistry class. For the following two reasons, however, the OFAT method falls flat for real-world experiments:

1. It cannot detect interactions of factors (e.g., the combination of time and temperature burning the food you cook in your kitchen).
2. It is very inefficient.

I explain why in this very brief two-part primer -- Trimming the FAT out of Experimental Methods and No-FAT Multifactor Design of Experiments.

Now that you understand why OFAT must be abandoned, you are ready to take the next step (i.e., mastering the multifactor test methods of design of experiments -- DOE -- by reading our book). We start you off easy with several chapters with some basic statistics and simple comparative experiments. Then in Chapters 3 through 5, we explain how to use the primary tool for DOE: two-level factorials. These designs are excellent for screening many factors to identify the vital few. They often reveal interactions that would never be found through OFAT methods. Furthermore, two-level factorials are incredibly efficient, producing maximum information with a minimum of runs. Most important, these designs often produce breakthrough improvements in product quality and process efficiency.

Continuing on from there, the book fills in all the gaps remaining for getting an effective grasp on the powerful tools for multifactor testing. We cap it all off with a new chapter written especially for 3rd Edition -- It reveals a middle way, called “split plot,” between OFAT and DOE when experimenters are confronted with factors that are very hard to change.

By the way, this new edition vaults the book into a format amenable to electronic users as it's available in many eBook formats. Now, web-connected experimenters around the globe can read DOE Simplified. Furthermore, mainly from the hard work of my colleagues at Stat-Ease, we accomplished a re-launch of the “Launch Pad” -- a series of interactive web-based lectures that cover the first several chapters of the book. The Launch Pad e-learning class provides momentum that propels readers through the DOE Simplified text. Contact workshops@statease.com for more information about the Launch Pad. 

Have any readers here used the new edition of DOE Simplified: Practical Tools for Effective Experimentation? Has it been helpful? Do you agree with Mark Anderson's points? 

Combining Statistical Design and Statistical Control -- Can it Work?

Kieron Dey recently published a book titled Competitive Innovation and Improvement: Statistical Design and Control, which explains how to combine two widely known statistical methods — statistical design and statistical control — in a manner that can solve any business, government, or research problem quickly with sustained results.I spoke with Kieron last month and asked him: "Why did you decide to write this book and what makes it unique?" Here is is his complete answer:

I first got the idea about separating tiny signals from large amounts of noise from time spent in radar design, and I wondered why similar was not much used in industry to solve problems. Where statistical design was used, it tended to be on a small scale and not much in processes involving lots of people.

The idea to combine statistical design and control came from a book on survey sampling. This fusion was controversial for years among professionals, and for no reason. Everything used is in the literature.

“Intent-to-treat" is also used throughout (which means, roughly, allowing an element of laissez-faire to get real world results, not forced ones that don't hold longer term).

Simultaneous design (where more than one design runs at once, overlaid) was added in 2012. The simultaneous designs have been important in cross-channel optimization in retail and in complex healthcare improvements. This was the last addition as the theory was tricky and it finally fell in place in 2011. It found that what had seemed weaknesses (where interactions across designs might be a problem) in fact hid a large strength, which is in Chapter 8 with real cases. The method had to be simplified so that users could apply easily.

Finally, the scientific method is used throughout (which folds nicely into comparative effectiveness research, DMAIC or PDCA, etc.), and the book explains what (and how simple) this is. The scientific method allows the same method to be used for existing and new processes: hence the “improvement and innovation” in the title. Innovation becomes less elusive in this way – it can be designed rather than waiting for inspiration. In addition, getting back to pure, simple science means using right-brain (creativity) as well as left (analytic) so more people can contribute in a valuable manner for the enterprise (which can be business, industry, research or government).

There is no mathematical notation so that anyone can read and use these well-established methods. Scientists and researchers will find Chapter 8 challenging on scientific method and randomization, so there's something for everyone. Mathematics is used a lot behind the scenes of the book but the real world is used more: to understand how businesses work and make them work better. 

There are about 20 exercises peppered through the book, for the reader to accelerate what would be learned in field experience and get started on real business competitive problems. 

Surprisingly, it turns out to be a management tool, not one technical people alone can accomplish; it’s not top down though and the book explains why. 

If anyone has read Kieron Dey's book, please feel free to post your comments here.