A recent discussion with a client led to these questions—“How do we keep design of experiments (DOE) training “alive” so that long-term benefits can be seen? How do we ensure our employees will apply their new-found skills to positively impact the business?” In my 20+ years as a DOE consultant and trainer, I have seen many companies who invested in on-site training, only to have it die a quick death mere days after the instructor leaves. On the other hand, we have long-term relationships with clients who have fully integrated design of experiments into the very culture of their research and development, and wouldn’t consider doing it any other way. What are the keys that lead the latter to success?
Key #1: Top-Down Management Support
Management must focus on long-term results versus short-term fixes. Design of experiments is a key tool to gain a fundamental understanding of processes. When combined with basic scientific and engineering knowledge, it helps technical professionals discover the critical interactions that drive the process. It’s not free, experimentation costs time and money. But forward-thinking companies understand that the long-term gains are worth the short-term expense. Management needs to buy-in to the use of DOE as a strategic initiative for future success.
Key #2: Data-Driven Decisions
Long-term success is achieved when management insists on using data to make decisions. My first engineering role was in a company that told us “All decisions are made based on data.” Engineers were expected to collect data and bring it to the table. DOE was one of the preferred methods to collect and analyze data to make those decisions. Key #2 is ingraining the expectation into the business that data-driven results will benefit the company longer than gut-feel decisions.
Key #3: Peer-to-Peer Learning
People like to learn from each other. Training can be sustained by learning from DOE’s done by peers. One way to support this is to plan monthly “lunch and learn” sessions. Everyone brings their own lunch (or order pizza!) and have 2-3 people do informal presentations of either an experiment recently completed, or their proposed plan for a future experiment. If the experiment is completed, review the data analysis, lessons learned, and future plans. If it is a proposed DOE plan, discuss potential barriers and roadblocks, and then brainstorm options for solving them. The entire session should be run in an open and educational atmosphere, with the focus on learning from each other. This key demonstrates the practical application of DOE and inherently encourages others to try it.
Key #4: Practice, Practice, Practice
Company management should plan that the output of on-site training is a specific project to apply DOE. Teams should plan an experiment that can be run as soon as possible to reinforce the concepts learned. As DOE’s are completed, the data can be shared with classmates simply to provide everyone with some practice datasets. The mantra “use it or lose it” is very true with data analysis skills and setting aside some time to get together and review company data will go a long way towards reinforcing the skills recently learned. Schedule a follow-up webinar with the instructor if more guidance is needed.
Key #5: Local Champions
There are always a couple of people who gravitate naturally towards data analysis. These people just seem to “get it”. Invest in those people by providing them with additional training so that they can become in-house mentors for others. This builds their professional reputation and creates a positive, driving force within the company for sustainability.
The investment in on-site training should include a company plan to sustain the education long-term. Good management support is an essential start, establishing expectations on using design of experiments and other statistical tools. Employees should then be connected with champions, followed by opportunities to apply DOE’s and share practical learning experiences with their peers.
Design-Expert® software provides powerful features to add confidence, prediction, or tolerance intervals to its graphical optimization plots. All users can benefit by seeing how this provides a more conservative ‘sweet spot’. However, this innovative enhancement is of particular value for those in the pharmaceutical industry who hope to satisfy the US FDA’s QbD (quality by design) requirements.
Here are the definitions:
Confidence Interval (CI): an interval that covers a population parameter (like a mean) with a pre-determined confidence level (such as 95%.)
Prediction Interval (PI): an interval that covers a future outcome from the same population with a pre-determined confidence level.
Tolerance Interval (TI): an interval that covers a fixed proportion of outcomes from the population with a pre-determined confidence level for estimating the population mean and standard deviation. (For example, 99% of the product will be in spec with 95% confidence.)
Note that a confidence interval contains a parameter (σ, μ, ρ, etc.) with “1-alpha” confidence, while a tolerance interval contains a fixed proportion of a population with “1-alpha” confidence.
These intervals are displayed numerically under Point Prediction as shown in Figure 1. They can be added as interval bands in graphical optimization, as shown in Figure 2. (Data is taken from our microwave popcorn DOE case, available upon request.) This pictorial representation is great for QbD purposes because it helps focus the experimenter on the region where they are most likely to get consistent production results. The confidence levels (alpha value) and population proportion can be changed under the Edit Preferences option.
Ever wonder what the difference is between the various response surface method (RSM) optimization design options? To help you choose the best design for your experiment, I’ve put together a list of things you should know about each of the three primary response surface designs—Central Composite, Box-Behnken, and Optimal.
Central Composite Design (CCD)
Box-Behnken Design (BBD)
For an in-depth exploration of both factorial and response surface methods, attend Stat-Ease’s Modern DOE for Process Optimization workshop.
We’ve designed Design-Expert® software to be flexible and user-friendly. For those of you who haven’t had a chance to fully explore its capabilities, here are some tips to help you navigate the software and find options that are useful for you:
A central composite design (CCD) is a type of response surface design that will give you very good predictions in the middle of the design space. Many people ask how many center points (CPs) they need to put into a CCD. The number of CPs chosen (typically 5 or 6) influences how the design functions.
Two things need to be considered when choosing the number of CPs in a central composite design:
1) Replicated center points are used to estimate pure error for the lack of fit test. Lack of fit indicates how well the model you have chosen fits the data. With fewer than five or six replicates, the lack of fit test has very low power. You can compare the critical F-values (with a 5% risk level) for a three-factor CCD with 6 center points, versus a design with 3 center points. The 6 center point design will require a critical F-value for lack of fit of 5.05, while the 3 center point design uses a critical F-value of 19.30. This means that the design with only 3 center points is less likely to show a significant lack of fit, even if it is there, making the test almost meaningless.
TIP: True “replicates” are runs that are performed at random intervals during the experiment. It is very important that they capture the true normal process variation! Do not run all the center points grouped together as then most likely their variation will underestimate the real process variation.0
2) The default number of center points provides near uniform precision designs. This means that the prediction error inside a sphere that has a radius equal to the +/- 1 levels is nearly uniform. Thus, your predictions in this region (+/- 1) are equally good. Too few center points inflate the error in the region you are most interested in. This effect (a “bump” in the middle of the graph) can be seen by viewing the standard error plot, as shown in Figures 1 & 2 below. (To see this graph, click on Design Evaluation, Graph and then View, 3D Surface after setting up a design.)
Ask yourself this—where do you want the best predictions? Most likely at the middle of the design space. Reducing the number of center points away from the default will substantially damage the prediction capability here! Although it can seem tedious to run all of these replicates, the number of center points does ensure that the analysis of the design can be done well, and that the design is statistically sound.