CBE 9190B – ADVANCED STATISTICAL PROCESS ANALYSIS
COURSE OUTLINE 2014‐2015
Description This course is for engineers involved with experimental investigation and interpretation of data. Basic, applied statistical concepts are reviewed. Regression analysis techniques for fitting and discrimination of theoretical models are discussed. Methods for design and analysis of experiments are examined, with emphasis on factorial designs, and response surfaces. Prerequisites Graduate student status or permission of the department, Statistical Sciences 2141a/b or 2143a/b or equivalent. Note: It is the student's responsibility to ensure that all Prerequisite and Corequisite conditions are met or that special permission to waive these requirements has been granted by the Program. It is also the student's responsibility to ensure that they have not taken a course listed as an Antirequisite. The student may be dropped from the course or not given credit for the course towards their degree if they violate the Prerequisite, Corequisite or Antirequisite conditions. Corequisites None. Antirequisites CBE 4425, CBE 9420, or the former ES 509. Contact Hours 3 lecture hours per week, 1 tutorial hour per week, 0.5 course. Monday and Tuesday, 4:30 – 6:30, SEB 3102. Tutorials will be on Tuesday. If you wish to ask a question outside of class, please see me after class/tutorial or contact me by email to set up a mutually convenient time. Instructor
Dr. A. Hrymak, (SEB 2008) Telephone: 519‐661‐2128 email: [email protected] (email preferred)
Graduate Coordinator Ashley Jokhu (TEB 477) Telephone: 519‐661‐2111 ext: 88352 email: [email protected] Teaching Assistant Atieh Motaghi, email: [email protected] Required Text
You must have a copy of Minitab and the most convenient route is to rent or buy a licence on‐line at http://www.onthehub.com/minitab/minitab_english.htm Reference Text G.E.P. Box, J.S. Hunter, and W.G. Hunter, Statistics for Experimenters ‐ Design, Innovation and Discovery, 2nd edition, Wiley. ISBN: 978‐0471718130. Montgomery, D.C. and Runger, G.C. “Applied Statistics and Probability for Engineers 5th Ed” Wiley. Or most recent edition. Course Notes Course notes and lecture slides will be available on Sakai. “Process Improvement Using Data” by Kevin Dunn. Lab Notes None. Laboratory None. Units SI, British and US Engineering units will be used. General Learning Objectives The general objectives will require the student to be able to: • Understand and apply basic applied statistical concepts to engineering problems.
• Use Matrix methods to develop linear least squares analysis and analysis of variance for both single and multi‐response systems. • Apply Factorial and Response Surface Methods to experimental design. • Use of latent variable methods to analyze large sets of data. Specified Learning Objectives • To review fundamental statistical inference and uncertainty concepts. a) be familiar with normal, chi‐squared and F and t distributions b) confidence limits c) hypothesis tests • To be able to fit models to data and to then judge the appropriateness of the models; the model fitting will cover linear models, nonlinear models and multi‐response models. a) linear least squares b) ANOVA c) matrix methods d) joint confidence regions e) multi‐response estimation
• To provide an appreciation for the concepts of statistical process control. • To understand the concepts behind the design of experiments and to be able to set up appropriate designs in practice. a) understand the concept of design of experiments and confounding of factors b) set up full and fractional factorial experiments c) set up multi‐level experiments and response surface methods • To provide an introduction to multivariate statistical methods for analyzing and using large databases commonly arising in science, engineering and business. Outline: 1. Visualizing data: creating high‐density, efficient graphics that highlight the data. 2. Univariate data analysis: Probability distributions, confidence intervals, and hypothesis tests. Process monitoring, statistical process control (SPC), for tracking process behaviour. 3. Least squares regression modelling: correlation, covariance, ordinary and multiple least squares models. 4. Design and analysis of experimental data and response surface methods for continual process improvement and optimization. 5. Introduction to latent variable modelling: a general overview of latent variable models and their use in chemical engineering processes.
If time permits, we will also cover topics from non‐linear regression and principal components analysis. Evaluation Assignments 20% (estimate 5 assignments) Midterm Test 30% (February 10, 2015, 2 hr exam, standard calculators only) Final Exam 50% The Final Exam will be 3 hours with an additional take‐home component. For the supervised exam component, only standard calculators may be used and the Course Notes (paper copy). Grade adjustment techniques may be used. Note 1) Students must receive a passing grade on the final examination to pass this course. 2) Assignments are to be handed in during class on the specified due date provided by the Instructor. Extensions only for medical (with doctor’s note), compassionate or emergency reasons. Course Schedule (changes will be communicated as soon as known) Week of January 5, 2015 Introduction, Data Visualization ‐ No Tutorial this week
Week of January 12, 2015 Univariate Data Analysis Assignment 1 given out Week of January 19, 2015 Process Monitoring Assignment 1 due Week of January 26, 2015 Least Squares Modelling Assignment 2 given out Week of February 2, 2015 Least Squares Modelling Assignment 2 due
Week of February 9, 2015 Mid Term Exam
Week of February 16, 2015 BREAK WEEK BREAK WEEK
Week of February 23, 2015 Design and analysis of Experiments
Assignment 3 given out Week of March 2, 2015 Design and analysis of
Experiments
Assignment 3 due Week of March 9, 2015 Design and analysis of
Experiments
Assignment 4 given out Week of March 16, 2015 Design and analysis of
Experiments
Assignment 4 due Week of March 23, 2015 Latent Variable Methods Assignment 5 given out Week of March 30, 2015 Latent Variable Methods Assignment 5 due Week of April 6, 2015 Latent Variable Methods, Review Repeating All Components of the Course In accordance with Senate and Faculty Policy, students who have failed an Engineering course (i.e. <50%) must repeat all components of the course. No special permissions will be granted enabling a student to retain laboratory, assignment or test marks from previous years. Previously completed assignments and laboratories cannot be resubmitted for grading by the student in subsequent years. Use of English In accordance with Senate and Faculty Policy, students may be penalized up to 10% of the marks on all assignments, tests, and examinations for the improper use of English. Additionally, poorly written work with the exception of the final examination may be returned without grading. If resubmission of the work is permitted, it may be graded with marks deducted for poor English and/or late submission. Attendance Any student who, in the opinion of the instructor, is absent too frequently from class or laboratory periods in any course, will be reported to the Associate Chair (Graduate) (after due warning has been given). On the recommendation of the Department concerned, and with the permission of the Associate Chair (Graduate), the student will be debarred from taking the regular examination in the course. Cheating University policy states that cheating is a scholastic offence. The commission of a scholastic offence is attended by academic penalties, which might include expulsion from the program. If
you are caught cheating, there will be no second warning (see Scholastic Offence Policy in the Western Academic Calendar). Plagiarism Students must write their essays and assignments in their own words. Whenever students take an idea or a passage from another author, they must acknowledge their debt both by using quotation marks where appropriate and by proper referencing such as footnotes or citations. Plagiarism is a major academic offence (see Scholastic Offence Policy in the Western Academic Calendar).
The University of Western Ontario has software for plagiarism checking. Students may be required to submit their work in electronic form for plagiarism checking. Sickness and Other Problems Students should immediately consult with the instructor or Associate Chair (Graduate) if they have problems that could affect their performance in the course. The student should seek advice from the Instructor or Associate Chair (Graduate) regarding how best to deal with the problem. Failure to notify the Instructor or the Associate Chair (Graduate) immediately (or as soon as possible thereafter) will have a negative effect on any appeal. Please contact the course instructor if you require material in an alternate format or if any other arrangements can make this course more accessible to you. You may also wish to contact Services for Students with Disabilities (SSD) at 661‐2111 x 82147 for any specific question regarding an accommodation. Notice Students are responsible for regularly checking their Western email and notices posted on their Instructor’s doors. Course cancellations will also be sent out via Twitter. Consultation Students are encouraged to discuss problems with their teaching assistant and/or instructors in tutorial sessions. Office hours will be arranged for the students to see the instructor and teaching assistants. Other individual consultations can be arranged by appointment with the appropriate instructor. Accreditation (AU) Breakdown Engineering Science = 50% Engineering Design = 50% January 5, 2015/ah
