Outline OF EXPERIMENTS

0
0
1794 days ago, 614 views
PowerPoint PPT Presentation
THE ANALYST\'S PURPOSES. UNDERSTANDING (a charge of general circumstances and end results connections connected with a specific phenomenon)EXPLANATION (use of chose connections to a specific observation)PREDICTION (stretching out information of connections to a future event)CONTROL (mediating with the goal of either fortifying or keeping a specific occasion).

Presentation Transcript

Slide 1

Outline OF EXPERIMENTS M. Piczak November 2005

Slide 2

THE ANALYST'S PURPOSES UNDERSTANDING (an order of general circumstances and end results connections related with a specific wonder) EXPLANATION (utilization of chose connections to a specific perception) PREDICTION (stretching out information of connections to a future occasion) CONTROL (mediating with the plan of either fortifying or keeping a specific occasion)

Slide 3

PREFERRED PREDICTORS GYPSIES BOOKIES TAROT CARDS PALM READERS FARMER'S ALMANAC HOROSCOPE FORTUNE COOKIES EQUATIONS

Slide 4

POINT PREDICTION VERSUS GENERAL STATEMENTS

Slide 5

NARROWING THE KNOWLEDGE GAP

Slide 6

THE DOE OBJECTIVE: To set up a causal connection between chose autonomous factors X's and specific ward Y factors To confine or unravel the free impact being applied on a reaction variable

Slide 7

POINT OF DEPARTURE Ho: That there exists no relationship between the X and Y variable (expect that the estimation of the coefficient for Xi = 0) Ha: That there exists a relationship between the X and Y variable The test is to negate Ho and in this way, acknowledge Ha

Slide 8

OPTIONS PRESENT THEMSELVES REGRESSION METHODS TAGUCHI METHODS SHAININ METHODS 6 SIGMA METHODS CLASSIC DESIGN OF EXPERIMENTS

Slide 9

SOUND EXPERIMENTAL PROCEDURE Start with a consistency, normality or oddity deserving of examination (See " What is Worth Studying ') Establish the exploration questions Undertake a writing survey to recognize other key logical factors (See " Weird Predictors ') Explicate the significant hypothesis for every X variable as it identifies with Y, from the earlier Establish the measures for X and Y Choose a test configuration utilizing Minitab Execute the investigation Gather the information Analyze the information Draw conclusions Repeat as important to cumulate learning

Slide 10

LEARNING WITH AN EXAMPLE F = M * A Where: F = Force M = Mass A = Acceleration

Slide 11

SETTING UP THE DATA F = M * A

Slide 12

ADDING THE Y DATA F = M * A

Slide 13

CALCULATING COEFFICIENTS F = M * A

Slide 14

STAT, DOE, FACTORIAL, CREATE FACTORIAL DESIGN

Slide 15

WITH PERFECT RESULTS

Slide 16

STAT, DOE, FACTORIAL, ANALYZE FACTORIAL DESIGN

Slide 17

WILL REGRESSION GIVE THE SAME RESULTS? CODED DATA

Slide 18

RUNNING UNCODED ALL X's

Slide 19

RUNNING F = M * An ALONE

Slide 20

THE THEORY Thus, the real model:

Slide 21

TRANSFORMING THE CODED MODEL INTO ACTUAL Ma & Mc   or Ma = 7.5 + 2.5Mc Therefore

Slide 22

RELATIONSHIP BETWEEN CODED & UNCODED VALUES

Slide 23

SIMILARLY FOR ACCELERATION or Aa = 150 + 50 Ac Therefore

Slide 24

SUBSTITUTING TERMS F = M * A

Slide 25

"TA DA"… F = M * A

Slide 26

BUT THE ANSWER WAS THERE THE WHOLE TIME… WHERE IS IT?

Slide 27

OTHER RESOURCES http://www.airacad.com/PaperDOE.aspx http://www.statease.com/bars/popcorn.pdf http://thequalityportal.com/q_know02.htm http://www.hq.nasa.gov/office/hqlibrary/ppm/ppm35.htm http://www.isixsigma.com/tt/doe/

Slide 28

DESIGN OF EXPERIMENTS M. Piczak November 2005 THE END

SPONSORS