Applied DoE

A worked-solution walkthrough of Kevin Dunn’s 12-week Design of Experiments short course, written for practitioners in industry. Each module is an answer key for the corresponding weekly worksheet at yint.org, and uses process_improve code end-to-end.

The series is organized into four phases. Each module page weaves in facilitator notes as inline Tip, Guidance, Check yourself, and Solution callouts so you can pause, reflect, and test your own understanding before reading the worked answer.

Before you start

A short planning template to fill in before any of the modules below.

• Plan your own experiment

Phase A. Foundations

The two-factor mindset; how factors, levels, and responses connect to a linear model.

• Why experiment? The 2x2 mindset
• Coding, scaling, and the linear model

Phase B. Full factorial designs

Scaling up to three and more factors, reading effects and interactions, and diagnosing the model that comes out.

• Full factorials: replicates, center points, and three factors
• A bigger factorial: four factors and model reduction

Phase C. Doing less, learning more

Screening designs when you cannot afford a full factorial, and judging whether a design is worth running before you spend the budget.

• Doing less: fractional factorials
• Trade-off table, blocking, and the first move toward optimization

Phase D. Optimization

Moving from a screening study to an optimum, including response surfaces, steepest ascent, and balancing multiple responses.

• Response surfaces and steepest ascent in two factors
• Course wrap-up: vocabulary, API map, and what comes next

Sources and attribution

Each module is a worked solution to the corresponding weekly worksheet from Kevin Dunn’s Design of Experiments short course. The original worksheets live at https://yint.org/resources. Worksheet questions are paraphrased here; the canonical wording remains in the source documents. The companion textbook is Process Improvement using Data (Chapter 5 covers DoE).