dcsimg
 

Four More Tips for Making Sure Your DOE isn't D.O.A.

In my last post, I shared some helpful advice for performing DOE from Minitab trainers Lou Johnson and Eduardo Santiago.

Read on for four more tips for making sure your design of experiments project isn't dead on arrival:

1. Fractionate to save runs focusing on Resolution V designs.

In many cases, it's beneficial to choose a design with ½ or ¼ of the runs of a full factorial. Even though effects could be confounded or confused with each other, Resolution V designs minimize the impact of this confounding which allows you to estimate all main effects and two-way interactions. Conducting fewer runs can save money and keep experiment costs low.
Choosing the right fractional factorial helps reduce the size of your experiment while minimizing the level of confounding of effects.
 

2. Replicate to improve the power of your experiment.

Power is the probability of detecting an effect on the response, if that effect exists. Power is also a function of the number of replicates. Increasing the power of an experiment via replication increases the chance that you will be successful in determining which input variables affect your response.

Power is a function of the number of replicates.

3. Improve power by quantifying a binary response.

Reducing defects is the main goal of most experiments, so there is a tendency to use defect count as the response. However, defect counts are a very expensive and unresponsive output to measure. Measuring a quantitative indicator related to your defect level can dramatically decrease your sample size and improve the power of your experiment.

Measuring a quantitative indicator of the reason you are producing defects can dramatically decrease your sample size and improve the power of your experiment.
 

4. Study all variables of interest and all key responses.

Factorial designs allow you to take a comprehensive approach to studying all potential input variables. Don’t let fear of complexity cause you to leave out potentially important input variables. Removing a factor from the experiment reduces the power of determining its importance (power) to zero.
Factorial designs take a comprehensive approach, but fear of complexity often causes experimenters to avoid many variables

Do you have any DOE tips or tricks to add to our list?

Minitab features Factorial, Response Surface, Mixture, and Taguchi designs to help you find the settings to optimize your processes. You can try Minitab Statistical Software free for 30 days by clicking here.

Of possible interest:

A DOE  in a Manufacturing Environment (Part 1)

A DOE in a Manufacturing Environment (Part 2)

Comments

blog comments powered by Disqus