Once influential factors are identified, the next goal usually consists of optimising their settings.
This module covers the construction of experimental designs for optimisation applications. Several types of optimisation designs exist. Data modelling is carried out with response surface methodology – RSM.
- Objectives and Context of Use of Optimisation Designs
- Issues involved in Optimisation
- Relevant Statistical Concepts
- The Need for an Alternative to ANOVA
- Construction of Optimisation Designs
- Notion of Central Point and Repetitions
- Properties of Optimisation Designs
- Central Composite Designs (CCD) and Box-Behnken Designs
- Use of Blocks
- Methodology of Response Surfaces
- Use of Models for Prediction & Optimisation
- Advanced Methods
- Summary
This module is intended for researchers who conduct experiments or studies, who wish to optimise their model, process or formulation and who wish to determine optimal conditions using an appropriate optimisation design.
Participants must possess a working knowledge of the construction of general full factorial designs (main effects, interactions, etc.) and analysis of Variance (ANOVA) as well as linear regression modelling techniques, or, equivalently, must have attended the courses:Upon completion of this module, participants will be able to:
-
- Select the factor and the levels to be tested
- Construct an optimisation design
- Construct a response surface model and interpret the results
- Assess the goodness-of-fit of the model
- Use the model to determine the optimal factor settings
- Plot the model using response surfaces and contour plots
- Make predictions and optimise the process using response surfaces
Recommended Duration: 2 day(s)
Course Materials:
TBD
Case Studies:
N/A