lect18

lect18 - IE 410: Design of Experiments Lecture 18 Factor...

Info iconThis preview shows pages 1–12. Sign up to view the full content.

View Full Document Right Arrow Icon
IE 410: Design of Experiments Lecture 18 Factor Screening Designs: INTRODUCTION Chapters 6, 7, and 8 are essentially contiguous.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
The topics are 2 k designs (Chapter 6), Confounded 2 k designs, or 2 k in blocks (Chapter 7), and Fractional Factorial Designs, (Chapter 8). The main results of interest are the Fractional Factorial designs of chapter 8. 6 and 7 are in many ways leading to chapter 8.
Background image of page 2
A common and important situation arising in the design of experiments is the need to investigate many factors. Typically one might construct a list of factors that MAY affect the response. The experiment is designed to find which factors do indeed have a significant response.
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Thus we try to "screen" out the important from the unimportant. This accounts for the name "Factor Screening Designs" often simplified to "Screening Designs"
Background image of page 4
We have seen that factorial designs have many desirable properties, and are in general, preferred. However, as the number of factors grows, the number of required observations grows very quickly (exponentially).
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Our goal is to examine via experiment a large number of factors in a small number of runs. To accomplish this goal we shall do the following:
Background image of page 6
1. Reduce the number of levels of each factor to 2 If we have k factors to investigate and each is at 2 levels, the result is a design with 2 k cells. 2. We shall reduce the number of obs. per cell to 1.
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
3. We shall show how to block without additional runs. 4. We shall show how to reduce the design to even fewer runs than 2 k
Background image of page 8
Finally, we shall show a process of experimentation by which we can narrow in on the important factors. This procedure is sequential in that it augments preliminary data with additional data over time.
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
It is this last process of experimentation that is most important, and which the 3 chapters lead up to. Chapter 6 is devoted primarily to points 1 and 2 above.
Background image of page 10
CHAPTER 6: 2 k Designs Each factor is at 2 levels to reduce the number of runs.
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 12
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 58

lect18 - IE 410: Design of Experiments Lecture 18 Factor...

This preview shows document pages 1 - 12. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online