In this course we learn how to use the methods of statistical design of experiments (DOE) to design efficient experiments, analyze the results correctly and present them in a clear fashion. Statistical DOE is used widely in both industry and academia. Graduate and undergraduate students from any field of science or engineering can use the methods learned in the course in their projects and research. The course includes use of the statistical software JMP, planning and running real experiments as projects and classroom active learning exercises based on real case studies.
Introduction to Statistics or a similar basic statistics course
At least 10 of the 13 lectures and tutorials
During the course exercises and projects will be assigned. Submission is required! At the end of the course a final exam will be given (open-book and open-note, no laptops). The weight of the exam in the final grade is 40-50%. Projects and the final exam can be written in Hebrew or English.
1) Introduction and overview of principles of DOE 2) Two level factorial designs(2k, 2k-p) 3) Choice of factors, levels of factors, responses 4) Adding center points to two level factorial designs to test adequacy of linear model, response surface designs (to fit quadratic models) , optimization using JMP statistical software 5) Models for data with a special structure: fixed and random, nested and crossed effects e.g. for analyzing process data with a batch structure, and for investigating systematic vs. random variation 6) GRR (Gage Repeatability & Reproducibility) – evaluation of measurement variation 7) Principles of Taguchi (robust design) experiments – design factors and noise factors 8) Randomization test for random block experiments 9) Latin square experiments 10) Design and analysis of real experiments (projects)
All the relevant material will be provided in the lectures and tutorials. Lecture slides and tutorial material are on the Moodle and should be downloaded by the students. Two supplementary books that include more than required are: 1) George E. P. Box, William G. Hunter, J. Stuart Hunter, Statistics for Experimenters: An Introduction to Design, Data Analysis, and Model Building, Second Edition, John Wiley & Sons, New York, 2005. 2) Douglas C. Montgomery, Design and Analysis of Experiments, Sixth Edition, John Wiley & Sons, New York, 2004.