Technical Training

Design of Experiments

Overview

Unlike 'One Factor At A Time' (1FAT), DESIGN OF EXPERIMENTS (DOE)  is a powerful tool that enables you to investigate and manipulate multiple key process input variables concurrently in order to optimise a specific output or response variable. 

This course will expose learners to key knowledge required to design and analyse statistical experiments using Minitab. 

Pre-requisite: Basic Statistics with Minitab will be a distinct advantage. However, the first day of the course will be a review of basic statistical concepts relevant to DOE.

Content includes

Course Outline:

DAY 1: DOE STATISTICS
1 Understand key statistical concepts and definitions
· Population, sample, types of data etc
· Measures of process variation and central tendency
· Descriptive and inferential statistics.

2 Understand the distribution of your data
· Distribution parameters
· Difference between PDF and CDF
· Probability Distributions – normal, t, binomial, Poisson, F, Chi-square.

3 Review of Parameter Estimation and Statistical Inferences with Minitab
· Confidence Intervals.
· Hypothesis testing.
· Analysis of Variance.
· Goodness-of-fit test.
· Individual Distribution Identification.

4 Review of Regression Analysis with Minitab.

DAY 2: DOE FUNDAMENTALS
1 Understanding DOE terms and concepts
· independent and dependent variables,
· factors and levels,
· treatment, error, replication,
· full and fractional designs,
· screening experiments,
· confounding, etc

2 Experimental Planning
· Measurement systems analysis,
· Identifying your objectives,
· Identifying factors and responses of interest,
· Design type selection,

3 Creating a Design In Minitab
· Create a Full Factorial Design,
· Understand Design Table,
· Modify your design to Fractional Factorial,
· Understand Aliasing and Alias Structure.

4 Manually Analyse A Full Factorial Design
· Understand Main Effects,
· Understand Interaction Effects

DAY 3: DESIGN AND ANALYSIS OF EXPERIMENTS
1 Create and Analyse A Screening Experiment
· Definitive Screening Design.
· Plackett-Burman Design
- Analyse Design Summary,
- Analyse Pareto Effects,
- Analyse Effects Plot,
- Analyse Main Effects,

2 Create and Analyse A Two-level Full Factorial Design
· Create and store your design.
· Analyse your design using available tools in Minitab
· Four-in-One plot, probability plot,
· ANOVA, Pareto, main and interaction plots etc.
· Reduce model by screening out factors that are not statistically significant.
· Optimise your design.
- Identify optimum settings using Contour plot, Surface Plot, Response Optimiser

3 Create A Response Surface Design.
· Central Composite Design.
· Response Surface with Categorical a factor.

4 Create and analyse a Split-Plot Design.

Who should attend

Manufacturing, Quality, R&D, Product Design Engineers, Inspection & Test personnel, Process Validation Team etc.

Upcoming dates:

No training dates available at the moment.