Webinars
Stat-Ease is proud to offer free webinars to those interested in design of experiments (DOE) and Design-Expert® software. Presented on a regular basis, topics range from beginner to advanced levels. Material may be new or drawn from our ever-popular DOE workshops.
If there is a particular subject you are interested in and don't see below, send us an email and let us know. If there is enough interest, we may present your topic in a future webinar.
"Thank you very much for the continuing education opportunities through Stat-Ease's webinar offerings. The instructors do an excellent job of explaining DOE. Thank you, Stat-Ease!"
—Jeff Reimer, Principal Scientist, Sigma-Aldrich Corporation
Upcoming Live Webinars:
In this presentation, I will speak from the heart on my lifelong involvement with design of experiments. Starting with my early years as a new engineer using DOE, I will then focus on providing DOE for 35 years at Stat-Ease. In wrapping up, I will give a sneak peek at version 13 of Design-Expert® software.
June 8 at 10:00am CT
In today’s Industry 4.0, industrial processes are becoming increasingly complex, presenting significant challenges to the industrial experimenter. In particular, modern experimental design practice can often lead to non-standard situations. In this talk I will discuss some examples of the non-standard experimental design situations I’ve encountered in modern practice, with the common denominator in all these situations being a split-plot treatment structure. I will illustrate the potential consequences of incorrectly analyzing experimental data collected as split-plots, as well as the capabilities of Design Expert to correctly design and analyze experiments of this type.
June 9 at 10:00am
A nice new addition to Design Expert is the KCV designs (Kowalski, Cornell, and Vining 2000 and 2002) for experiments that involve both mixture components and process variables. This talk presents an overview on these designs. It begins with a brief history of their origin. It then motivates the basic approach for the construction of these designs and contrasts this approach to other approaches popular at that time. It then discusses some of the subtleties involved in analyzing these designs. An example illustrates their use.
June 10 at 10:00 CT
Design of experiments is typically presented as a “one shot” approach. However, it may be more efficient to divide the experiment into smaller pieces, thus expending resources in a smarter, more adaptive manner. This sequential approach becomes especially suitable when experimenters begin with very little information about the process, for example, when scaling up a new product. It allows for better definition of the design space, adaption to unexpected results, estimation of variability, reduction in waste, and validation of the results.
The statistical literature primarily focuses on sequential experimentation in the context of screening, which in our experience is only the beginning of an overall strategy for experimentation. This tutorial begins with screening and then goes well beyond this first step for more complete coverage of this important topic:
- Screening before experimentation
- Adding and removing factors during the experiment
- Expanding, shrinking, or repairing the experimental design space during the experiment
- Validation experiments
- And more!
Several real-world examples will be provided and demonstrated using software, thus providing attendees a solid briefing with very helpful aspects for practical application of sequential experimentation.
June 11 at 10:00am CT
The control of ingredients quantities in pharmaceutical formulations is critical to a drug product quality. Often, the Active Pharmaceutical Ingredient (API), and sometimes specific excipients, concentrations testing in the final product is required to release a batch.
On an existing marketed product, 2 liquid chromatography methods have been developed and are currently used to test the API and an antioxidant concentration in an aqueous injectable solution. These methods are time-consuming and labor-intensive. The lead time to receive the results is in the range of days.
By applying Design of Experiments (DoE) and Multi-Variate Data Analysis (MVDA), a UV-Vis spectroscopy chemometric model has been developed which is capable of measuring simultaneously both ingredients contents in the formulation. The test can easily be applied to finished or bulk product and the results are immediately available.
Such method can be used in Quality Control lab to reduce the workload and batch release lead time. More interestingly, it can be implemented on-line to test the bulk product, opening the door to real-time process control and product release.
June 15 at 9:00am Central Time
Stratasys is the world’s leading company in the field of 3D printing for over 30 years. It is one of the only companies producing printers capable of simultaneously printing in 7 colours, enabling also colour-mixing on the tray.
Stratasys’ PolyJet technology is similar to 2D inkjet printing in that it jets a thin layer of resin materials onto the surface, which is then polymerized on the surface using UV light. The process is then repeated multiple times, each layer adding thickness to the model. Using this method layers of 14-28 micrometers are printed giving high accuracy.
As in most formulated products, inkjet ink is a complex mixture, with many elements often having a contradictory effect. The inks must match the substrate, the hardware, and each other. This difficult endeavor is exacerbated in the case of 3D printing. Any small mismatch, defect, or problem will be enhanced as each layer is deposited on top of the previous one.
When jetting ink for 3D printing, a support material must be also printed beneath model material overhangs. This material must be strong enough to support the weight of the model above it, but also be easily removable after printing. In addition, due to the printing sequence and the internal environment inside the printer, the support material must have a certain resilience and thermal stability, and provide adequate surface quality of the model post support removal.
One major drawback of our printing technology is that it is virtually impossible to anticipate the performance of the ink in the lab, and we therefore have to print each and every formulation. This process is both lengthy and cumbersome.
The first incentive to use DOE for this particular product was that no matter what changes were made to the formulation, the polymerized support material would melt as the ambient temperature increased during the printing process. The second incentive was to arrive at a lead formulation with a minimal number of trials.
June 16 at 8:00am Central Time
Non-ionic surfactants are an important component of industrial metalworking cleaners. For beverage can manufacturing these cleaners need to deliver good cleaning performance with low foaming properties. These attributes can be strongly influenced by different types of non-ionic surfactants.
Each type of non-ionic surfactant will provide its own property to the cleaning product. Low HLB (Hydrophilic–Lipophilic Balance) surfactants will reduce foaming to a certain degree. Medium HLB surfactants will improve the fluid’s wetting properties. High HLB surfactants will increase cleaning performance together but can also cause an unwanted increased foaming behavior.
With the help of Design-Expert® software an experimental design was set up in the form of an augmented simplex lattice to find an optimal balance between three selected surfactants. The experimental formulations were tested for foaming properties and cleaning performance. All experimental data was analyzed by the software to calculate models with good fit.
The calculated models confirmed that the higher HLB surfactant was responsible for the cleaning performance, but also gave significant foam formation. It also showed that the low HLB surfactant was not needed to reduce the foam, but a certain amount of medium HLB surfactant already gave significant de-foaming properties.
The optimization module within Design-Expert® was used to find the optimal surfactant combination according the set requirements. This optimal product was tested again and showed good results as expected. The use of experimental design and statistical software led to the development of a well-balanced and marketable product within a short period of time.
June 17 at 9:00am Central Time
Design of Experiments (DoE) facilitates the screening for relevant factors and the optimization of the latter in terms of one or several responses. The number of factors that can be included in screening designs is substantial, but may still be a bottleneck if complex processes with interdependent steps have to be optimized. For example, cryopreservation protocols for plant cells consist of more than 5 steps, each of which can depend on more than 10 factors. The way in which the steps affect each other can be complex (i.e. effects may only appear in the second next step or so) and difficult to predict. However, improving such protocols is important in the context of biopharmaceutical manufacturing because they affect the quality of cryo-stocks necessary to ensure a consistent batch-to-batch quality of the biological starting material as well as the product.
Here we present an iterative approach to structure the multi-parameter problem and illustrate how challenges regarding the experimental implementation of a design can be handled. We highlight how the resulting models can support a quality by design approach as recommended by the regulatory authorities for biopharmaceutical manufacturing. Therefore, we think that our data will be of interest to colleagues working in this research area as well as for all those developing complex procedures that require optimized working conditions and detailed documentation.
June 18 at 9:00am Central Time
Richard Williams, Stat-Ease Consultant, details what design of experiments (DOE) can do for you when catalyzed with DX12’s world-class statistical tools. Learn about factorial design, the core tool for DOE, followed by a peek at response surface methods (RSM) for process optimization and last, but not least, a look into mixture design for optimal formulation.
Recorded Webinars:
By way of example, this presentation lays out a strategy for design of experiments (DOE) that provides maximum efficiency and effectiveness for development of a robust process. It provides a sure path for converging on the ‘sweet spot’—the most desirable combination of process parameters and product attributes. Whether you are new or experienced at doing DOE, this talk is for you (and your organization's bottom line!).
Richard Williams, Stat-Ease Consultant, details what design of experiments (DOE) can do for you when catalyzed with DX12’s world-class statistical tools. Learn about factorial design, the core tool for DOE, followed by a peek at response surface methods (RSM) for process optimization and last, but not least, a look into mixture design for optimal formulation.
Pat Whitcomb details the cost-saving mixture-process models developed by Scott Kowalski, John Cornell and Geoff Vining (KCV). Design-Expert® software, version 12, drops this modeling tool right into the user's hands. See how it reduces the number of model terms and thereby reduces the number of runs required to estimate the complex relationship between mixture and process variables. Estimated length: 45 min.
Mark Anderson, Stat-Ease Consultant and lead author of the DOE/RSM/Formulation Simplified trilogy, will detail what design of experiments (DOE) can do for you when catalyzed with DX12’s world-class statistical tools.
Martin Bezener, Stat-Ease Consultant, introduces Design-Expert v12’s new tools for logistic regression.
Learn about the highlights for innovative new statistical tools and graphical presentations in version 12 of Design-Expert software. Enjoy an enlightening demonstration of DX12’s fabulous new capabilities presented at a beginner’s level.
Discover how to optimize your process while avoiding impossible factor combinations.
Learn about factorial design, the core tool for DOE, followed by a peek at response surface methods (RSM).
This case-study driven webinar is a must for all who experiment on APIs. Learn how to apply statistically valid, multifactor and multicomponent testing strategies that catalyze your development work.
Discover new features in Design-Expert® software, version 11, with a demonstration of the overhauled interface.
Use mixture design tools for multi-component product development and optimization, perfect for formulators. Learn why factorial designs won't work.
Pat Whitcomb reveals some tricks for making the most of your DOE.
Use graphical tools (half-normal & Pareto plots) to select effects quickly and accurately.
How to use automatic model selection tools to build on appropriate models. Pros and cons of the methods are discussed.
Learn how multivariate analysis (MVA) methods can be used in combination with design of experiments (DOE) tools. Presenters demonstrate how to build an optimal design from principal components and use the DOE results to find an optimal compound.
Learn how to use power and precision to properly size factorial and RSM designs.
Discover how split-plot designs are the best choice when you have hard-to-change (HTC) factors.
Review strategies for running confirmation to verify the results of an experiment.
Topics include foldovers, semifoldovers, building a CCD from a one-factor-at-a-time (OFAT) study, and optimal augmentation for RSM designs.
We have an answer for your question: "How many runs do I really need?"
A briefing on QbD, along with state-of-the-art response surface methods (RSMs) for developing a robust design space.
Discover how propagation of error and tolerance analysis can account for variation in your system.
Topics include using optimal designs for constrained design spaces, adding categoric factors, higher-order models, and design augmentation.
Review central composite designs and multiple response optimization.
Learn why, when, and how to use algorithmic (optimal) designs for your experiments.
An introduction to factorial design. What is DOE and how can you use it?
Historical data can be difficult to analyze - look out for these common issues.
Verification is an essential final step to improving a product or process. Learn how DOE makes this a breeze.
A dual response approach to response surface methods (RSM) provides a statistically sound way to make your process more robust.
Learn the differences between replicates, duplicates, and repeats, as well as the reasons for using each. Cost-based decision selection of one versus another are discussed.
Learn how Design-Expert's optimization tools help you find the "sweet spot" for your product or process.