Вы находитесь на странице: 1из 28

Improve Phase

Designing Experiments

M# 504
Designing Experiments

Welcome to Improve

Process Modeling: Regression

Advanced Process Modeling:


MLR Reasons for Experiments

Designing Experiments Graphical Analysis

Experimental Methods DOE Methodology

Full Factorial Experiments

Fractional Factorial Experiments

Wrap Up & Action Items

M# 505
OSSS LSS Black Belt v10.3 - Improve Phase 2 © Open Source Six Sigma, LLC
Project Status Review

• Understand our problem and it’s impact on the business. (Define)


• Established firm objectives/goals for improvement. (Define)
• Quantified our output characteristic. (Define)
• Validated the measurement system for our output characteristic.
(Measure)
• Identified the process input variables in our process. (Measure)
• Narrowed our input variables to the potential “X’s” through
Statistical Analysis. (Analyze)
• Selected the vital few X’s to optimize the output response(s).
(Improve)
• Quantified the relationship of the Y’s to the X’s with Y=f(x).
(Improve)

M# 505
OSSS LSS Black Belt v10.3 - Improve Phase 3 © Open Source Six Sigma, LLC
Six Sigma Strategy

SIPOC
VOC
Project Scope

P-Map, X-Y,
(X1) (X2) (X3) (X4) (X8) (X11) (X9) FMEA
(X6) (X7) (X5) Capability
(X10)
(X4) (X1) (X11) Box Plot, Scatter
(X3)
(X8) Plots,
(X2) (X5) Regression

Fractional
(X5) (X3) Factorial
Full Factorial
(X11) Center Points
(X4)

M# 506
OSSS LSS Black Belt v10.3 - Improve Phase 4 © Open Source Six Sigma, LLC
Reasons for Experiments

The Analyze Phase narrowed down the many inputs to a critical few,
now it is necessary to determine the proper settings for the vital few
inputs because:
– The vital few potentially have interactions.
– The vital few will have preferred ranges to achieve optimal results.
– Confirm cause and effect relationships among factors identified in
Analyze Phase (e.g. regression)
Understanding the reason for an experiment can help in selecting
the design and focusing the efforts of an experiment.
Reasons for experimenting are:
– Problem Solving (Improving a process response)
– Optimizing (Highest yield or lowest customer complaints)
– Robustness (Constant response time)
– Screening (Further screening of the critical few to the vital few X’s)

Design where you’re going - be sure you get there!


M# 506
OSSS LSS Black Belt v10.3 - Improve Phase 5 © Open Source Six Sigma, LLC
Desired Results of Experiments

Problem Solving
– Eliminate defective products or services.
– Reduce cycle time of handling transactional processes.
Optimizing
– Produce a mathematical model to move the process response.
– Opportunity to meet differing customer requirements (specifications or
VOC).
Robust Design
– Provide consistent process or product performance.
– Desensitize the output response(s) to input variable changes including
NOISE variables.
– Design processes knowing which input variables are difficult to
maintain.
Screening
– Past process data is limited or statistical conclusions
prevented good narrowing of critical factors in Analyze
Phase. When it rains it POURS!
M# 507
OSSS LSS Black Belt v10.3 - Improve Phase 6 © Open Source Six Sigma, LLC
DOE Models vs. Physical Models

What are the differences between DOE modeling and physical


models?
– A physical model is known by theory using concepts of physics,
chemistry, biology, etc...
– Physical models explain outside area of immediate project needs and
include more variables than typical DOE models.
– DOE describes only a small region of the experimental space.

The objective is to
minimize the response.
The physical model is
not important for our
business objective. The
DOE Model will focus in
the region of interest.

M# 507
OSSS LSS Black Belt v10.3 - Improve Phase 7 © Open Source Six Sigma, LLC
Definition for Design of Experiments

Design of Experiments (DOE) is a scientific method of planning


and conducting an experiment that will yield the true cause and
effect relationship between the critical X variables and the Y
variables of interest.

DOE allows the experimenter to study the effect of many input


variables that may influence the product or process
simultaneously, as well as possible interaction effects (for
example synergistic effects).

The end result of many experiments is to describe the results


as a mathematical function.
Y = f (x)
The goal of DOE is to find a design that will produce the
information required at a minimum cost.

Properly designed DOE’s are more efficient experiments


M# 508
OSSS LSS Black Belt v10.3 - Improve Phase 8 © Open Source Six Sigma, LLC
One Factor at a Time is NOT a DOE

One Factor at a Time (OFAT) is an experimental style but not a


planned experiment or DOE.
The graphic shows yield contours for a process that are unknown to
the experimenter.
Trial Pressure Temp Yield
Yield Contours Are 1 125 30 74
Unknown To Experimenter 75 2 125 31 80
3 125 32 85
4 125 33 92
80 5 125 34 86
6 130 33 85
Pressure (psi)

7 120 33 90
135 85
6
130 90
125
1 2 3 4 5 Optimum identified
95 with OFAT
120
7

True Optimum available


30 31 32 33 34 35 with DOE
Temperature (C)
M# 508
OSSS LSS Black Belt v10.3 - Improve Phase 9 © Open Source Six Sigma, LLC
Types of Experimental Designs

The most common types of DOE’s are:


– Fractional Factorials (aka Screening DOEs)
• 4-15 input variables
– Full Factorials
• General Full Factorials
• 2k Full Factorials
• 2-5 input variables
– Response Surface Methods (RSM)
• 2-3 input variables Response
Surface
Full Factorial

Fractional Factorials

M# 509
OSSS LSS Black Belt v10.3 - Improve Phase 10 © Open Source Six Sigma, LLC
Nomenclature for Factorial Experiments

The general notation used to designate a 2-level full factorial


design is given by:

– Where k is the number of input variables or factors.


– 2 is the number of “levels” that will be used for each factor.
• Quantitative or qualitative factors can be used.

This is a 24 example (4 factors at 2 levels)


Factors (Coding) Temp (oF) Pressure (psi) Solvent Supplier Office Location
Low Factor (-1) 300 500 Fisher London
High Factor (+1) 350 600 Baker New Delhi

M# 509
OSSS LSS Black Belt v10.3 - Improve Phase 11 © Open Source Six Sigma, LLC
Visualization of 2 Level Full Factorial

600 (-1,+1) (+1,+1)


300
Temp
350
22 Press
500
Press
600 500
Uncoded levels for factors (-1,-1) (+1,-1)

300F Temp 350F

T P T*P
-1 -1 +1 Four experimental runs:
+1 -1 -1 • Temp = 300, Press = 500
-1 +1 -1 • Temp = 350, Press = 500
+1 +1 +1 • Temp = 300, Press = 600
Coded levels for factors • Temp = 350, Press = 600
M# 510
OSSS LSS Black Belt v10.3 - Improve Phase 12 © Open Source Six Sigma, LLC
Graphical DOE Analysis - The Cube Plot

Stop Angle Start


Angle
Consider a 23 design on a catapult...

Fulcrum
8.2 4.55 A B C Response

Run Start Stop Meters


Number Angle Angle Fulcrum Traveled
3.35 1.5 1 -1 -1 -1 2.10
2 1 -1 -1 0.90
Stop Angle

3 -1 1 -1 3.35
5.15 2.4 4 1 1 -1 1.50
5 -1 -1 1 5.15
6 1 -1 1 2.40
Fulcrum
7 -1 1 1 8.20
8 1 1 1 4.55
2.1 Start Angle 0.9

What are the inputs being manipulated in this design?


How many runs are there in this experiment?
M# 510
OSSS LSS Black Belt v10.3 - Improve Phase 13 © Open Source Six Sigma, LLC
Graphical DOE Analysis - The Cube Plot

Open the MINITABTM worksheet “Catapult.mtw”


Stat > DOE > Factorial > Factorial Plots…
check Cube Plot box and click Setup…
select response and factors

Cube Plot (data means) for Distance

8.20 4.55

3.35 1.50
This graph is used by 1
the experimenter to
visualize how the What
response data is are Stop Angle 5.15 2.40
distributed across the
these? 1
experimental space.
Fulcrum
How do you read or 2.10 0.90
-1
interpret this plot? -1 1
-1

Start Angle
M# 511
OSSS LSS Black Belt v10.3 - Improve Phase 14 © Open Source Six Sigma, LLC
Graphical DOE Analysis - The Main Effects Plot

Stat > DOE > Factorial > Factorial Plots…


check “Main Effects Plot” box and click “Setup…”
to select response and factors

This graph is used to see the


relative effect of each factor
on the output response.

Which factor’s main


effect has the
largest impact on
the output?
Hint: Check the slope! M# 511

OSSS LSS Black Belt v10.3 - Improve Phase 15 © Open Source Six Sigma, LLC
Main Effects Plots’ Creation

Avg Distance at Low Setting of Start Angle: 2.10 + 3.35 + 5.15 + 8.20 = 18.8/4 = 4.70
Main Effects Plot (data means) for Distance
-1 1 -1 1 -1 1
5.2

4.4
Dist

3.6

2.8

2.0
Start Angle Stop Angle Fulcrum

Avg. distance at High Setting of Start Angle: 0.90 + 1.50 + 2.40 + 4.55 = 9.40/4 = 2.34
Run # Start Angle Stop Angle Fulcrum Distance
1 -1 -1 -1 2.10
2 1 -1 -1 0.90
3 -1 1 -1 3.35
4 1 1 -1 1.50
5 -1 -1 1 5.15
6 1 -1 1 2.40
7 -1 1 1 8.20
8 1 1 1 4.55
M# 512
OSSS LSS Black Belt v10.3 - Improve Phase 16 © Open Source Six Sigma, LLC
Interaction Definition

Interactions occur when variables act together to impact the output


of the process. Interaction plots are constructed by plotting both
variables together on the same graph. They take the form of the graph
below. Note that in this graph, the relationship between variable “A”
and output Y changes as the level of variable “B” changes. When “B”
is at its high (+) level, variable “A” has almost no effect on Y. When “B”
is at its low (-) level, A has a strong effect on Y. The feature of
interactions is non-parallelism between the two lines.

High
B-
When B changes
Y from low to high,
with A at high,
Output

When B changes the output


from low to high, changes
with A at low, the dramatically.
output changes
B+
Low
very little.
A- A+
M# 512
OSSS LSS Black Belt v10.3 - Improve Phase 17 © Open Source Six Sigma, LLC
Degrees of Interaction Effect

Some Interaction No Interaction Full Reversal


High High High
B- B-
B-

Y B+
B+ Y B+ Y

B+
Low Low Low
- A + - A + - A +
Strong Interaction Moderate Reversal
High High
B- B-

Y Y

B+
B+ B+
Low Low
- A + - A +
M# 513
OSSS LSS Black Belt v10.3 - Improve Phase 18 © Open Source Six Sigma, LLC
Interaction Plot Creation

Interaction Plot (data means) for Distance


Start Angle
6.5
-1
1
5.5

4.5

Mean
3.5

2.5
(4.55 + 2.40)/2 = 3.48
1.5

(0.90 + 1.50)/2 = 1.20 -1 1


Fulcrum

Run # Start Angle Stop Angle Fulcrum Distance


1 -1 -1 -1 2.10
2 1 -1 -1 0.90
3 -1 1 -1 3.35
4 1 1 -1 1.50
5 -1 -1 1 5.15
6 1 -1 1 2.40
7 -1 1 1 8.20
8 1 1 1 4.55

M# 513
OSSS LSS Black Belt v10.3 - Improve Phase 19 © Open Source Six Sigma, LLC
Graphical DOE Analysis - The Interaction Plots

Stat > DOE > Factorial > Factorial Plots…


Check “Interaction Plot” box and click “Setup…”
to input response and factors

Interaction Plot for Distance


When you select more than two Data Means
variables, MINITABTM generates an -1 1 -1 1

Start
Interaction Plot which allows you to 6
Angle
look at all factor interactions 4
-1
Star t A ngle 1
simultaneously. The plot at the
2
upper right shows the effects of Start
Angle on Y at the two different levels 6
Stop
Angle
of Fulcrum. The red dashed line 4
-1
Stop A ngle 1
shows the effects of Fulcrum on Y
2
when Start Angle is at its high level.
The black solid line represents the
effects of Fulcrum on Y when Start
Fulcr um
Angle is at its low level. M# 514
OSSS LSS Black Belt v10.3 - Improve Phase 20 © Open Source Six Sigma, LLC
Graphical DOE Analysis - The Interaction Plots

Stat > DOE > Factorial > Factorial Plots…


Check “Interaction Plot” box, click “Setup..” button, then click “Options...” button

Check this option for


these additional plots

Interaction Plot for Distance


The plots at the lower left in Data Means
-1 1
the graph below (outlined in 6
Start

blue) are the “mirror image” 4


Angle
-1
Star t A ngle 1
plots of those in the upper 2

right. It is often useful to 6


Stop
Angle
look at each interaction in Stop A ngle
4
-1
1
both representations. 2

6
Fulcrum
-1
1
4
Fulcr um

-1 1 -1 1 M# 514
OSSS LSS Black Belt v10.3 - Improve Phase 21 © Open Source Six Sigma, LLC
DOE Methodology

1. Define the Practical Problem


2. Establish the Experimental Objective
3. Select the Output (response) Variable
4. Select the Input (independent) Variables
5. Choose the Levels for the Input Variables
6. Select the Experimental Design
7. Execute the experiment and Collect Data
8. Analyze the data from the designed experiment and draw
Statistical Conclusions
9. Draw Practical Solutions
10. Validate the experimental results
11. Implement Solutions

M# 515
OSSS LSS Black Belt v10.3 - Improve Phase 22 © Open Source Six Sigma, LLC
Generate 2-level Full Factorial Designs in MINITABTM

Stat > DOE > Factorial > Create Factorial Design…


A 23 Full Factorial
design will result in 8
experimental runs.
Full Factorial
Experimental Designs
are always Green!

M# 515
OSSS LSS Black Belt v10.3 - Improve Phase 23 © Open Source Six Sigma, LLC
Create Three Factor 2-level Full Factorial Design

Stat > DOE > Factorial > Create Factorial Design…

M# 516
OSSS LSS Black Belt v10.3 - Improve Phase 24 © Open Source Six Sigma, LLC
Create Three Factor 2-level Full Factorial Design

M# 516
OSSS LSS Black Belt v10.3 - Improve Phase 25 © Open Source Six Sigma, LLC
Create Three Factor 2-level Full Factorial Design

M# 517
OSSS LSS Black Belt v10.3 - Improve Phase 26 © Open Source Six Sigma, LLC
Three Factor 2-level Full Factorial Design

Hold on! Here we go….


M# 517
OSSS LSS Black Belt v10.3 - Improve Phase 27 © Open Source Six Sigma, LLC
Summary

At this point, you should be able to:

• Determine the reason for experimenting

• Describe the difference between a physical model and a DOE model

• Explain an OFAT experiment and its primary weakness

• When shown a Main Effects Plot and Interaction Plot, determine


which effects and interactions may be significant.

• Create a Full Factorial Design

M# 518
OSSS LSS Black Belt v10.3 - Improve Phase 28 © Open Source Six Sigma, LLC

Вам также может понравиться