 |
Syllabus
Required textbook:
| |
Applied Statistics and Probability for Engineers, 3rd
edition, 2003,by D. C. Montgomery and G. C. Runger, John Wiley and Sons, with
eText CD |
Other (not required) reading material:
| |
Statistical Methods for Industrial Process Control
Handbook of Experimental Methods for Process Improvement both by David
Drain, Chapman and Hall, 1997 |
IEE 598 DOE/SPC for Semiconductor Processing
Week
|
Lecture
|
Lecture Topics
|
1
|
Chapter 1
Chapter 2
|
-
Course overview/review of syllabus/student information, role of statistics in
engineering (1-1 through 1-4)
-
Introduction to probability and random variables (rv’s) (2-1 through 2-8)
|
2
|
Chapter
3
Chapter 4
|
-
Probability, probability density functions of discrete rv’s, and the
binomial distribution (3-1, 3-2, 3-4, 3-6, 3-9)
-
Probability plots, probability mass functions of continuous rv’s, and the
normal distribution (4-1, 4-2, 4-4, 4-6, 4-7)
|
3
|
Chapter
6
|
-
Correlation and independence, random sampling, and the central limit theorem
(CLT) (6-1 to 6-7)
|
4
|
Chapter
8 Chapter 9
|
-
Statistical intervals for a single sample, point estimation, confidence
intervals for means and variances (8-1 to 8-4), semiconductor examples
-
Tests of hypotheses for a single sample, t tests for means, and tests for
variance (9-1 to 9-4)
|
5
|
Chapter 10
Midterm
|
-
Inference for a difference in two means with known and unknown variances,
variances of two normal populations, and review summary table for two-sample
inference procedures, paired t-test (10-1 through 10-5)
-
Midterm Exam
|
6
|
Chapter
13
|
-
Introduction to designed experiments and analysis of variance (ANOVA) for a
single factor (13-1, 13-2)
|
7
|
Chapter
13
|
-
Randomized block designs (13-4), role in semiconductor processing
|
8
|
Chapter
14
|
-
Multiple-factor designed experiments (DOE) (14-1 through 14-5)
-
2k designs for
factors (14-7) and examples for semiconductor processing
|
9
|
Chapter
14
|
-
Single replicate of a 2k design (14-7)
-
Blocking of a 2 k design (14-8)
|
10
|
Chapter
14
|
-
Fractional replication of a 2k design (14-9)
- Addition of center points to a 2k design (14-7 supplemental)
|
11
|
|
-
Examples for semiconductor processing
-
Midterm Exam
|
12
|
Chapter
16
|
-
Introduction to statistical process control (SPC) (16-1 to 16-4)
-
Control charting, and the
 control
chart and special concerns for semiconductor manufacturing (16-5)
|
13
|
Chapter
16
|
-
Rational subgroups, control charts for individuals (16-6)
-
Process capability (16-7)
|
14
|
Chapter
16
|
-
Introduction to attribute data
-
Attribute control charts (16-8)
|
15
|
Chapter
16
|
-
Control chart performance and special topics (16-9)
|
|
|
|
Course
Syllabus: IEE 598 Design of Experiments/ Statistical Process Control for
Semiconductor Processing
Instructor: George C. Runger,
Required textbook:
| |
Applied Statistics and Probability for Engineers, 3
rd edition, 2003,by D. C. Montgomery and G. C. Runger, John Wiley and Sons,
with eText CD |
Other (not required) reading material:
| |
Statistical Methods for Industrial Process Control
Handbook of Experimental Methods for Process Improvement both by David
Drain, Chapman and Hall, 1997
|
About the course:
-
A course in statistical process control and improvements through designed
experiments that focuses on semiconductor processing
-
Intended for engineers, and physical/chemical scientists, and deals with the
types of control charts and experiments that are frequently run in industrial
settings
-
A basic working knowledge of introductory statistical methods would be useful
background, but introductory material will be covered at the start of the
course
-
The introductory material that will be covered at the start of the course
includes the following:
-
Compute and interpret the sample mean and standard deviation
-
Use the normal distribution
-
Test a hypothesis (the t-test, for example)
-
Construct and interpret a confidence interval
Course objective:
-
Interpret data summaries
-
Compute basic probabilities for risk assessment
-
Reason statistically from a sample to a process
-
Plan, design, conduct, and analyze experiments efficiently and effectively
-
Assess process control and capability and to develop and use basic control
charts
-
Adjust analyses for important characteristics of semiconductor data
Opportunities to use the principles taught in the course arise in all phases of
engineering work, including new product design and development, process
development, and manufacturing process improvement. Methods will be customized
to semiconductor manufacturing and examples will be drawn from this field. Some
important modifications to standard methods are needed for semiconductor
processes.
All experiments conducted by engineers and scientists are designed experiments;
some of them are poorly designed, and others are well designed. The
well-designed ones allow you to obtain the desired results faster, easier, and
with fewer resources. That’s what you will learn how to do in this
course. A well-designed experiment can lead to reduced development lead-time
for new processes and products, improved manufacturing process performance, and
products that have superior function and reliability.
Computer software: Computer software to implement the methods
presented will be illustrated, and you will have opportunities to use it for
homework assignments and project. Campus labs provide Minitab, but any one of
several commercial packages can be used and the relationships between these and
Minitab should be easy to follow.
Grading: Your grade in the course will be determined by two
mid-term exams (50%), a final exam (25%) and homework projects (25%).
Additional homework exercises will be assigned, with solutions, but these will
not be scored.
Homework and Projects: Important!! You should work as many
exercises from the book as you feel are necessary to become familiar with the
material. These will not be turned in, but selected solutions will be provided.
Small projects will be assigned approximately every two weeks that will be more
comprehensive applications of the material. These will be turned in and they
will sometimes require you to collect and analyze data present to the class. No
proprietary information should be used. Additional details will be provided.
|
