QLTY09032 2013 Advanced Experimental Design
The student will be learn how to design conduct and analyse mixed level experiments, and interpret the data from these experiments.
Learning Outcomes
On completion of this module the learner will/should be able to;
Conduct two and three level fractional factorial experiments and analyse the resulting data.
Plan, conduct and analyse experiments using Response Surface Methodology (RSM).
Design and analyse mixed level experiments.
Analyse multiple response experiments and interpret the results.
Analyse and interpret data from experiments involving random effects models.
Formulate the expected means square rules to develop appropriate statistical models.
Use Minitab to design an experiment, analyse, interpret and evaluate the resulting data.
Module Assessment Strategies
Continuous Assessment
Project work - Design and analyse own experiment 20%.
Final Examination
One written paper of 2.5 hours duration on experimental desgign theory and analysis 40%
One computer based exam of 2.5 hours duration involving data analysis and interpretation 40%
Indicative Syllabus
1. Two Level Factorial Designs : full factorial Designs. fractional factorial designs. Analysis of Residuals. Blocking in two level designs. Fold-over designs. repeat experiments vs. replicate experiments. Building the regression model and verification of the model.
2. More complex Two Level Factorial designs : Analysis of single replicate designs using probability plots and Lenth's method. Data transformations in a factorial design, variance stabilisation and Bob-Cox transformations. Analysis of multiple response experiments e.g. mean response and variability of response. Addition of centre points to a design. Robust methods of experimental design such as Taguchi methods. Solution of static and dynamic problems using Taguchi methods.
3. Three Level Factorial Designs: Full Factorial Designs, fractional factorial designs, Blocking in and designs, Analysis strategies for multiple responses.
4. Design and Analysis of Mixed Level Designs: Constructing mixed level designs using method of replacement. Analysis strategies for mixed level designs.
5. Response Surface Methods: Method of Steepest Ascent, verifying adequacy of first order model. Analysis of second order response surface. Locating the stationary point. Characterising the response surface. Ridge systems. Multiple response problem. Selecting designs for fitting response surfaces. Central composite designs. Box-Behnken designs. Face centred cube design. Blocking in Response surface designs. Introduction to mixture experiments. Evolutionary Operation.
6. Experiments with Random factors: The random effects model. Rules for Expected means squares. Repeatability and Reproducibility (R&R) studies using the random effects model. Estimation of variance components. Mixed models. Approximate F tests and Satterthwaite's method.
7. Nested and Split Plot Designs: Crossed vs. Nested designs. Two stage nested design. Diagnostic checking and estimation of variance components. Staggered nested design. The general m-stage nested design. Analysis of Split-plot designs.
Coursework & Assessment Breakdown
Coursework Assessment
Title | Type | Form | Percent | Week | Learning Outcomes Assessed | |
---|---|---|---|---|---|---|
1 | Project Design and analyse own experiment | Coursework Assessment | UNKNOWN | 20 % | OnGoing | 1,2,7 |
End of Semester / Year Assessment
Title | Type | Form | Percent | Week | Learning Outcomes Assessed | |
---|---|---|---|---|---|---|
1 | Final Exam One final exam 2.5 hour theory paper. One computer based 2.5 hour exam involving data analysis and interpretation | Final Exam | UNKNOWN | 80 % | End of Term | 1,2,3,4,5,6,7 |
Full Time Mode Workload
Type | Location | Description | Hours | Frequency | Avg Workload |
---|---|---|---|---|---|
Lecture | Not Specified | Lecture | 3 | Weekly | 3.00 |
Independent Learning | UNKNOWN | Independent Learning | 4 | Weekly | 4.00 |
Part Time Mode Workload
Type | Location | Description | Hours | Frequency | Avg Workload |
---|---|---|---|---|---|
Lecture | Distance Learning Suite | Lecture | 3 | Weekly | 3.00 |
Independent Learning | UNKNOWN | Independent Learning | 4 | Weekly | 4.00 |
Module Resources
Authors |
Title |
Publishers |
Year |
Montgomery, Douglas |
Design and Analysis of Experiments, 8th Edition |
John Wiley & Sons |
2013 |
Box, Hunter & Hunter |
Statistics for Experimenters, |
John Wiley & Sons. |
2005 |
Draper, N. & Smith, H., |
Applied Regression Analysis, |
John Wiley & Sons. |
1998 |
Montgomery, DC & Myers, R. |
Response Surface Methodology: Process and Product Optimization using Design Experiments |
John Wiley & Sons. |
2009 |
Montgomery, D.C, Peck, E.A. & Vining, G.G. |
Introduction to Linear Regression Analysis. 5th Edition |
John Wiley & Sons. |
2012 |
Wu, J. & Hamada, M., |
Experiments: Planning, Analysis and Parameter Design Optimization |
John Wiley & Sons. |
2009 |
None
None