How to Make the Best Use of DoE (Design of Experiments)

Six Sigma represents a vast collection of tools that can be used for improving the way your company does business. One of the most popular and powerful is a DOE. Let’s take a look at how to get the most from this incredible tool.

Begin with Your Factors in Mind:
You can’t do a successful DOE without having factors to test. This is probably a no-brainer for you if you’re even remotely familiar with this concept. Still, people often forget to double-check and make sure that all of the relevant factors are included before moving forward.
For example, let’s say you were doing a DOE for baking a cake.

You’d want to make sure you included the following important factors:
• Shape of the pan: rectangle (low) vs. round (high)
• Ingredients: four cups vs. six cups of flour
• Oven temperature: 350 degrees vs. 400 degrees
• Cooking time: 30 vs. 40 minutes

You would then figure out how you’re going to rank the results to figure out which factors produced the best possible cake. Now, of course, you probably won’t be using DOE for baking. Aside from the obvious, though not impossible to work with, cakes are subjective.
If you’re using Six Sigma at work, you’ll most likely use DOE for something far more objective. It will be clear which factors are ideal.

Establish the Intended Goals of Your Current Processes:
Again, this probably seems so self-evident it hardly needs to be mentioned, but before you move forward with a DOE, get clear about what the results are you’re looking for from a process.
Going back to our cake example, the goal would probably be to create the most delicious dessert possible. However, it could also be to bake the best looking one possible. A combination of the two could also be the goal for a DOE.

Register For a Course Near Me

Understand How You’ll Collect Data:
It doesn’t matter how well-designed your DOE is or even how well you pull it off. Without a plan in place for how you’ll collect data related to the DOE, you won’t be making the most out of this important practice.

Make Sure Your DOE Can Be Reproduced:
The vast majority of the time, you need a DOE that can be leveraged again and again. The DOE itself should involve numerous tests. This is how you get rid of any results that were outliers because of some random element involved.
Depending on the nature of your company’s work, this ability to use the same protocol over and over again is a good way to save time while carrying out important tests. You may need to make minor changes, but the template will be there to work from.

Document Everything:
Some of you will have no choice but to document your DOEs. The practices that you’re applying them to are so complicated that there would be no other way to plan them without lots of documentation.
Others may feel like this is an unnecessary chore. As we mentioned above, though, you never want to reinvent the wheel. Make sure you record the specifics about your DOE so it’s easier to use whenever required.
Of course, having the specifics of your DOE on paper is also how you can go back and scrutinize it for signs of what may have gone wrong or what you could do better next time.
There is no doubt that DOEs are an extremely powerful element in the Six Sigma arsenal. No matter what industry you work in, it’s been proven that DOEs are almost always a potential tool for finding how you can improve. Now that you’ve read the above, you also have a much better idea about how to get the most of your DOE.

Sources:
https://www.statease.com/pubs/sixsigma&DOE.pdf
https://www.qimacros.com/lean-six-sigma-articles/how-to-perform-a-design-of-experiments/

Stop by 6Sigma.us and learn how we can make a difference in your business and bring clarity to your current challenges.