methods for quality improvement - anthony bonifontemethods for quality improvement isye 3039 ic 113...
TRANSCRIPT
Course Goals At the end of the course,
students should be able
to:
Perform data analysis and
suggest improvement plans
Design experiments to
efficiently optimize settings
Analyze factorial experiments
and draw statistically sound
conclusions
Develop control charts for
monitoring continuous and
discrete quality characteristics
Model time series data and
monitor for changes
Assess statistical process
capability
Apply methods of quality
control to manufacturing and
engineering settings
Course Description
Course Description
Instructor
Methods for Quality Improvement ISyE 3039 IC 113
“Statistics is the science of learning from experience.”
-Bradley Efron, 2006.
Topics include quality system requirements, designed
experiments, process capability analysis, measurement
capability, statistical process control, and acceptance
sampling plans.
In this course we will learn various statistical techniques for
improving the quality of engineering systems. Broadly, the steps in
this are summarized by the acronym DMAIC- Define, Measure,
Analyze, Improve, and Control. First we will study design of
experiments, which will enable us to plan how we will gather data.
We will practice the measure step by reviewing basic statistical
techniques. Analyze, improve, and control will be addressed by
studying statistical process control, which monitors when a process
has left its typical range of operation and needs repairing. The
content from this course is used in industrial systems,
manufacturing, health care, and any engineering environment
which involves the design and collection of data.
Anthony Bonifonte
Office: Main Building 312
Office Hours: Tuesday 11:00am - noon, Thursday 1:30pm – 2:30pm
If you are unable to attend these hours, you are encouraged to
email questions or schedule an appointment.
Email: [email protected]
Catalog Description
Fall 2016
Tu - Th 12:05pm – 1:25pm
Course Logistics Prerequisites: ISyE 2028
Textbook: Introduction to Statistical Quality
Control, 7th edition, by Montgomery
http://www.amazon.com/Statistical-
Quality-Control-Douglas-
Montgomery/dp/1118146816
It is important to have the current edition of
the book, as some homework problems may
be selected out of it.
TA: Junying (Jasper) He, [email protected]
Office: Main 450
Office Hours: Wednesday 11:00am – 1:00pm
Expectations Honor Code: I trust you to adhere to the Georgia
Tech Honor Code on all assignments. On
exams you will be required to write and
sign the Honor Pledge: ‚I have adhered to the
Honor Code on this assignment‛. How the
Honor Code applies on each assignment is
specified below, and any questions should be
directed to me. Cheating is wholly
unacceptable and will be dealt with harshly.
Class communication: All electronic communication
will be through announcements using T-
square and delivered to your Georgia Tech
email. You are responsible for checking these
messages periodically to stay informed of
important dates and potential changes to the
syllabus.
I am pleased to reply to questions via email at
[email protected]. Please include ‘ISyE
3039’ in the subject line. I check my email
frequently, but I reserve the right to a 48-hour
response period. This means questions
immediately before an exam or assignment
due date may not receive a timely response.
Please send all emails through your Georgia
Tech account so it does not get blocked by
spam filters.
Technology Policy: Please be respectful with your
use of laptops and other technology in class.
I request you only use them for class related
purposes, as I and others may find them
distracting. Cell phones should be kept silent
and away, and you can expect the same from
me.
Software: We will use computer software for some
of the course material. R is a powerful free
statistics program. Many of the homework
problems may be solved in Microsoft Excel,
but a small time investment to learn R will
greatly benefit you in the future.
Class forums: We have a forum on T-square. If
you have conceptual questions that may
benefit others, please post them there.
Class Project 15% Students will work in groups of 2-4 to identify a quality-
related problem, collect and analyze the necessary data,
draw conclusions, and present the solutions. A project
report (~7 pages) will be due Friday, 12/02/16. A 12-15
minute presentation will be made to the class during the
week of 11/29/16 – 12/01/16.
Of the project grade, 40% will derive from the written
report, 40% from the presentation, and 20% from feedback
given to fellow student’s presentations on standard
feedback forms.
Assignments and Grading
Homework 20% Homework problem sets will be assigned on a weekly basis. You are allowed to work in groups, however,
every student needs to submit individually and write up the solution in their own words. Problem sets will
be important to learn the skills you need to master course objectives. Some problems may be evaluated on
honest effort and others on accurate solutions. Solutions will be posted online in a timely manner.
Homework will be due at the start of class on the posted date. Please submit a legible hard copy. No late
homework will be accepted without an institute approved absence.
Re-grades: In the interest of fairness, re-grades for partial credit on homework will not be accepted. If you
believe a mistake on the grading was made, you have one week to submit a written explanation stapled to
the original assignment.
“Quality is never an accident; it is always the result of high intention, sincere effort,
intelligent direction and skillful execution; it represents the wise choice of many
alternatives.”
-NY Times, 1939
Final Course Grade: 90-100 = A; 80-89 = B;
70-79 = C; 60-69 = D; <60 = F
Exams (2 in class 20% each), Final Exam 25%
Exams: Two exams will take place during class on 9/20/16 and 10/18/16. To succeed on these exams, you will
need to demonstrate competence in the skills required for the homework. Exams will contain quantitative
questions on topic content and applications, and open response questions.
Final exam: the final exam will take place 12/15/16 from 11:30am to 2:20pm. The final exam will be
cumulative.
You are allowed to bring one single-sided sheet of notes to each midterm exam, and three sides of notes to
the final exam.
Additionally, you will be provided necessary statistical tables.
Calculators are permitted, though they may not have communication capability (no cell phone or laptop
calculators).
No resources of any sort may be shared during exams.
If you have accommodations through ADAPTS, be sure to inform me in advance and provide the
necessary documentation from http://adapts.gatech.edu/
Grade disputes must be submitted in written format stapled to the original exam within one week of them
being returned.
Individual make up exams: Individual make-up exams will only be offered if a student has valid reasons,
such as documented illness, severe illness or death in the family, accidents, or court appearances. ‚I didn’t feel
well‛ is insufficient. You must provide documentation prior to the make up.
Course Schedule
Suppose you have machines that produce a variable number of Gizmos© every day. Each course
topic will prepare you to answer the quoted question.
Dates Topics Textbook Sections
Exam Homework Due
8/23, 8/25 Introduction, statistics review (summary statistics, hypothesis testing) “How can I quantify and describe the output of a machine?”
3.1 - 4.4
8/30, 9/01 Stats Review continued 9/01
9/06, 9/08 Statistics review (regression, ANOVA) “How do I model and analyze the number of Gizmos produced as a function of various settings?”
4.5 – 4.6 9/08
9/13, 9/15 Design of experiments – factorial experiments “How can I set up an experiment to test the impact of changing binary settings on Gizmo production?”
13.1 – 13.5 9/15
9/20, 9/22 Design of experiments – factorial experiments continued 9/20
9/27, 9/29 Design of experiments – fractional factorial experiments “If I have limited resources and can only set up a few experiments, which settings should I change?”
13.6 9/29
10/04, 10/06 Design of experiments – response surface methodology “How can I interpolate quantitative settings and find an optimal setting for maximum production?”
14.1 – 14.3 10/06
10/13 10/11: Fall Break Design of experiments – in class project
10/18, 10/20 Statistical process control – methods and philosophies “How should I test and manage the quality of Gizmo production?
5.1 – 5.7 10/18
10/25, 10/27 Statistical process control – control charts for variables “If I observe the number of Gizmos produced every day, how do I test if the manufacturing process is in control?”
6.1 – 6.6 10/27
11/01, 11/03 Statistical process control – control charts for attributes “If I observe the fraction of defective Gizmos produced every day, how do I test if the manufacturing process is in control?”
7.1 – 7.5 11/03
11/08, 11/10 Statistical process control – multivariate control charts “How can I track several quality characteristics of Gizmos at once?”
11.1-11.3 11/10
11/17 Other quality methods – tolerance design “How do I how do I ensure Gizmos can fit in their packaging, if both are variable?”
8.8, 9.1 – 9.2
11/22 Tolerance design continued 11/24: Thanksgiving Break
11/29, 12/01 Student Projects
12/06 Review n/a 12/06
“Quality is doing the right thing when no one is looking.”
-Henry Ford