Design and Analysis of Experiments (DoE) (Accredited)

Explore design of experiments (DoE) to understand how inputs affect outputs, using multiple factors at once, with examples like car mileage and factor interactions.

Identify independent variables and dependent variables, or inputs and outputs, using car mileage as the example, and distinguish noise factors that affect the response.

Identify and narrow down factors that affect output through screening experiments, discarding unimportant ones. Understand significant factors and interactions in design of experiments to optimize the process.

Explore the five stages of design of experiments, from planning to screening, modeling, optimizing, and verification; learn to select factors and apply Central Composite Design and Box-Behnken designs.

Identify factors as inputs controlled by the experimenter and define levels for each factor. Combine factor levels into treatments to study their effects using full factorial designs and analyze responses.

Explore design of experiments with two factors at two levels, including coding, randomization, and analysis of main effects, interactions, and a regression equation using a coffee example.

Draws main effect, interaction, and contour plots from a two-factor, two-level experiment with sugar and milk, showing no interaction and guiding upcoming regression analysis.

Derive a linear regression equation y = 6 + 1.5 xs + 1.5 xm from a two-factor two-level experiment, using central point coding and plots to predict taste.

Watch a minitab demonstration of a two-factor, two-level full factorial design with milk and sugar, building a regression model and comparing main effect, interaction, and contour plots to manual calculations.

Examine two factors at two levels and their interaction using main effects, interaction and contour plots to show how one factor’s impact depends on the other.

Develop a regression equation for a 2x2 experiment with interaction, using sugar and milk effects, interaction terms, and interpretation via Minitab outputs.

Set up a two-factor, two-level factorial experiment in Minitab with sugar and milk, then analyze with regression, main effects, and their interaction. Use contour plots to interpret the rating.

Explore design of experiments with a two-factor catapult study, analyzing firing and release angles to predict distance using regression, main effects, interactions, factorial plots, and contour analysis.

Examine how noise factors affect experiments, focusing on known and unknown influences. Apply blocking, analysis of covariance, and randomization to mitigate their effects.

Blocking addresses known noise factors by dividing land into blocks with minimal within-block variation and randomizing experiments within each block, as shown in two-factor, two-level designs.

Apply analysis of covariance to address known but uncontrollable noise factors by including covariates such as ambient temperature or humidity in the factorial design analysis in Minitab.

Learn how to handle unknown noise factors through randomization and replication in a designed experiment, distinguishing replication from repetition and boosting reliability with multiple replicates in a two-level factorial design.

Explains noise handling via blocking, analysis of covariance, and replication, and demonstrates a 2x2 catapult experiment with replication to improve p values and pareto charts.

Demonstrates setting up a two-factor, two-level full factorial design in Minitab with two replications per corner. Analyzes results with regression and contour plots to approach the 350 millimeter target distance.

Add a third factor by introducing bean type (light vs dark roast), yielding eight treatments and a cube. Derive a regression equation with main effects and two-way interactions.

Explore a three-factor two-level design with milk, sugar, and coffee bean to build and interpret a regression equation, including main effects and two-way and three-way interactions.

Analyze a three-factor DoE, focusing on main effects and two-way interactions; show sugar as a significant factor for taste, while milk and bean are insignificant, via Pareto and cube plots.

Set up and analyze a three-factor, two-level full factorial in Minitab, with milk, sugar, and coffee (bean type) affecting taste, and explore main effects, interactions, and plots.

Learn three-factor design with center points to reduce runs, use randomization to control lurking noise, and center-point replication with milk, sugar, and bean to estimate error and curvature.

Explore a minitab two-level factorial design with three factors (milk, sugar, bean) plus two center points to estimate error and curvature; analyze results to identify significant factors and curvature.

Learn how partial factorial designs reduce runs from full factorial, using half or quarter factorial for two-level factors while preserving visibility of main effects for sugar, milk, and bean.

Learn how partial factorial designs cause confounding or aliasing, with main effects confounded with two-factor interactions in a four-run design, and compare to full factorial design and design resolution.

Learn how design resolution governs confounding in experiments, and how main effects and two-factor interactions are aliased in resolution 3, 4, and 5 designs using Minitab.