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).

Outline OF EXPERIMENTS M. Piczak November 2005

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)

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

POINT PREDICTION VERSUS GENERAL STATEMENTS

NARROWING THE KNOWLEDGE GAP

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

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

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

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

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

SETTING UP THE DATA F = M * A

ADDING THE Y DATA F = M * A

CALCULATING COEFFICIENTS F = M * A

STAT, DOE, FACTORIAL, CREATE FACTORIAL DESIGN

WITH PERFECT RESULTS

STAT, DOE, FACTORIAL, ANALYZE FACTORIAL DESIGN

WILL REGRESSION GIVE THE SAME RESULTS? CODED DATA

RUNNING UNCODED ALL X's

RUNNING F = M * An ALONE

THE THEORY Thus, the real model:

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

RELATIONSHIP BETWEEN CODED & UNCODED VALUES

SIMILARLY FOR ACCELERATION or Aa = 150 + 50 Ac Therefore

SUBSTITUTING TERMS F = M * A

"TA DA"… F = M * A

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

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/

DESIGN OF EXPERIMENTS M. Piczak November 2005 THE END

SPONSORS

No comments found.

SPONSORS

SPONSORS