Vous êtes sur la page 1sur 29

Chapter 4

PF/CE/CNX/SOP

1
Stemplots

Used to analyze data graphically to


check the overall distribution
Used hand-in-hand with the
histogram and the control chart
An efficient data organizer that has
the first measurement digits in
the first column and the
remaining digits in the second
column
The goal is to get a picture of the
distribution, so a split stem may
be necessary

2.2 7
2.3 6 9 1
2.4 3 5 4 3
2.5 1 6 7 8
2.6 3 1 2 7
2.7 2 5
2.8 3Stem 8 1 21
Stemplot exercise

Create a stemplot for the following


set of numbers

4.90 4.64 4.51 4.37 4.


4.36 4.65 4.35 4.54 4.
4.47 4.80 4.53 4.27 4.
4.04 4.43 4.26 4.18 4.
 Worksheet

 Add specification limits and ask yourself,


“What is the location, shape and
spread?”

3
 What about a split stem?
Tally Chart

A graphical table that shows


all the different values from a
set of data, the pattern of the
data and number of times
they occur in the data set.
Objective
Organize the data
Condense the data information
if the information is to vast
Obtain a distribution pattern
(*use the G chart – see
Appendix B.2)
2.2 X
 2.3 X X X
2.4 X X X X X
2.5 X X X X
2.6 X X X X
2.7 X X
2.8 X X X X X
2.9 X X
4
Tally Chart Exercise

Create a tally chart for the


following set of numbers

4.90 4.64 4.51 4.37 4.


4.36 4.65 4.35 4.54 4.
4.47 4.80 4.53 4.27 4.
4.04 4.43 4.26 4.18 4.
 Worksheet

 Add specification limits and ask yourself,


“What is the location, shape and
spread?”
5
Frequency Distribution

A table that shows all the


different values from a set of
data and the number of
times they occur in the data
set.
Objective
Organize the data
Condense the data information
if the information is too vast
Obtain a distribution pattern
(*use the G chart pg.579 or
Measurement Frequency
 ¯n) 2.2 1
2.3 3
2.4 5
2.5 4
2.6 4
2.7 2
2.8 4
2.9 2
total n= 25 6
Frequency Distribution
Exercise

Create a frequency distribution for


the following set of numbers

4.90 4.64 4.51 4.37 4.


4.36 4.65 4.35 4.54 4.
 4.47
Worksheet 4.80 4.53 * 4.27 4.
 4.04 4.43 4.26 4.18 4.

Measurement Frequency


total n=
 * Start with a tally chart 7
Pareto Charts
Reasons For Late Flights at "Wingin' it" Airlines

80

70

60
# of Late Flights

50

40

30

20

10

0
Airplane Mechanical Moldy Food Late Inclimate Air Traffic Fueling
Arrived Late Difficulty Luggage Weather Control Problem
Problem
Reasons

Focus on the “Critical Few” vs. the “Trivial Many”

Where would you focus your


attention?

8
Pareto Diagram (cont.)

Pareto Diagram of IC Board Defects Using M onetary


Loss

120%

1200
100%
1000
Monetary Loss ($)

80%
800

600
60% Etching #1
400
40% Problem
200 20% $$ wise
0 0%

Etching Soldering M olding Cracking Other


Problem

Pareto Diagram of IC Board Using Frequency of


Defects
120%

140
Frequency of Defects

100%

Soldering #1 120

80%

Problem
100

80

Frequency wise
60%

60
40%

40

20%
20

0 0%

Soldering Etching Molding Cracking Other


Problem

9
Defect Check Sheet

Departm ent Beginning Date Ending Date


Assy & Test Jan 1 Jan 31
Description of Problem
Late shipments to customer
Reasons
No parts xxxxx
Late parts xxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Nonconforming parts xxxxxxxxxxxxxxxxxxxx
No paperwork xxxxxxxxxx
Incorrect paperwork xxxxxxxxxxxxxxx

Based on the information above, and using


the next page, make a Pareto chart.

10
Pareto Chart exercise

30 100%

28
90%
26

24 80%

22
70%
20

18 60%

16
50%
14

12 40%

10
30%
8

6 20%

4
10%
2

11
Form Team

 Teams should be comprised of key individuals


directly involved with the process being addressed.
Other functional representatives can be brought in
as needed. Each individual brings unique
expertise and perspective to the team and melds
that with the talents of the other team members.

 The team process leads to decisions of higher


quality than those arrived at by individuals.
Moreover, members of a team are much more
willing to take ownership in the resultant decisions
and to actively pursue their successful
implementation.

Q u a lity
T o o lin g E n g i n e e rin g S u p p lie r P la n n in g P u rc h a s i n g F in a n c e A s s u ra n c e Shop

12
Brainstorming

on
onideas
ideas
nes aren'tapplicable
es aren't applicable
n those actionswill
those actions willbe
bedone
doneby.
by.
sfixed.
fixed.
nfixed,
fixed,and
andwhat
whatproblems
problemsstill
stillneed
needto
tobe
beworked
workedon
on
make
make sure that everyone understands who is Responsible,who
sure that everyone understands who is Responsible, whoisisAccountable,
Accountable,wh
w

13
Process Flow Chart

A diagramming tool that


is used to trace the
process from start to
finish
Symbols represent
actions
Used to identify waste
and make the process
clear to the team

14
Why create a process flow
Any Process Has At Least Three Versions

What You THINK It Is

What It ACTUALLY Is

What it SHOULD Be

15
Process mapping symbols

Activity
BLUE (Operation)

Transportation
GREEN (Product Movement)

YELLOW Inspection
I (Decision)

Storage
TAN
S (File)

PINK Delay

To next
BLACK product step

16
Process mapping

17
Process Mapping “How to”

1.Determine the boundaries of


the process
2.Determine the major steps in
the process
3.Sequence the major steps
4.Fill in the minor steps
5.Draw the flowchart with the
appropriate symbols
6.Test the chart for
completeness
7.Finalize the chart
8.

18
Class Exercise

Create a Process Flow diagram for


completing a homework assignment

19
Cause & Effect Diagram
The 6M’s

Measurement Method Machine

(X) Placement of
tube in machine
(X) Tools used for sizing (Tony / (X) Collet only has
to measure Haus) three sides

(X) Different
methods for
(C) Location of measurment (Jim (C) Plug gages
where
measurement is
(X) # of points
taken (Jim A) (X) Cutoff
to measure

Measure
.433-.435

(C) Raw tube is


oversize

(N) Different
certs/size

(N) Material
(X) Training hardness
measuring
tools (N) Wall (N) Fluid temp
(Brent) thickness

Manpower Materials Mothernature

Also known as a Fishbone or Ishikawa diagram

20
Cause & Effect Diagram

Causes Effect
( Sources of Variation or Variables ) ( Response or Output Variable )

M ETH O D M O THER N ATUR E M A T E R IA L

R ESPO NSE

Partition the Variables:


C=
N=
M EA SU R EM EN T M A C H IN E M A N PO W ER
X=

21
Cause & Effect C/N/X’s

C = those variables which must be held


constant and require standard operating
procedures to insure consistency.
Consider the following examples: the
method used to enter information on a
billing form, the method used to load
material in a milling or drilling process,
the autoclave temperature setting.

N = those variables which are noise or uncontrolled


variables and cannot be cheaply or easily held
constant. Examples are room temperature or
humidity.

X = those variables considered to be key process


(or experimental) variables to be tested in order
to determine what effect each has on the outputs
and what their optimal settings should be to
achieve customer-desired performance.

22
Class Exercise
Create a Cause and Effect diagram for not
receiving an “A” on an exam

Did not receive


an “A” on
exam

23
Scatter plots

A graph of measurement pairs that


shows whether there is correlation
between the measurements
When correlation exist, changes in
one measurement shows an effect
in another
If you draw a best fit line and
figure out an equation for that
line, you would have a ‘model’
that represents the data.

Scatter Chart (Weight vs mpg)

35 y = -0.0152x + 63.507
2
30 R = 0.9191
25
20
mpg

15
10
5
0
1900 2400 2900 3400 3900
Weight
24
Looking at correlation from
a scatter plot:

‘Correlation’ is a fancy word for how well


the model predicts the response from
the factors.

25
Is there really an effect?

26
Scatter Diagram example:
Shoe Height
Size
Height

Shoe Size 27
Least Squares Method

yˆ = mx + B
S xy
m=
S x2
B = y − mx
( ∑ x )( ∑ y )
S xy = ∑ xy −
n
(∑ x) 2

S x2 = ∑ x −
2

n
28
2
R

r= r 2

 ∑ ( x − x )( y − y ) 
 
 n − 1 
r=  
( s x )( s y )

sx =
∑(x − x) 2

n −1

sy =
∑ ( y − y) 2

n −1
29

Vous aimerez peut-être aussi