Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Tools of Quality
"As much as 95% of quality related problems in the factory can be solved with
seven fundamental quantitative tools." - Kaoru Ishikawa
By
Zaipul Anwar
Business & Advanced Technology Centre,
Universiti Teknologi Malaysia
What are the Basic Seven
Tools of Quality?
Fishbone Diagrams
Histograms
Pareto Analysis
Flowcharts
Scatter Plots
Run Charts
Control Charts
Where did the Basic
Seven come from?
Kaoru Ishikawa
Known for Democratizing Statistics
Fishbone Diagrams
No statistics involved
Step 1 & 2:
Poor Service
(backbone)
(head)
Example (2 of 4)
Step 3 & 4:
Responsiveness
Appearance
Poor Service
Attention Reliability
Example (3 of 4)
Step 5, 6, & 7:
Appearance Responsiveness
equipment time
personnel
facility
Attention Reliability
Example (4 of 4)
Step 8 & 9:
Use tools to analyze and evaluate causes
Pareto diagrams, charts, and graphs
Statistical analysis for causes in processes
Decide and take action
Use fishbone diagram, analysis and evaluations to
find causes that can be fixed
Take action to eliminate and fix problem causes
Summary (1 of 3)
Fishbone Diagrams
- visual diagram
- resembles fish skeleton
- identifies the causes of a problem (effect),
and their relationships
- created by Kaoru Ishikawa for Quality
Management
Summary (2 of 3)
Organizational Uses
Increases communication about
problems
Used to improve any product, process,
or service
Important part of quality management
Summary (3 of 3)
Bar chart
1) What is a Histogram?
2) What are some possible uses for a
Histogram?
3) Where did the Histogram come
from?
4) How do Histograms work?
5) A real world example.
6) An exercise.
What is a Histogram?
Normal Distribution:
How do Histograms
Work? (cont)
Positively Skewed:
Negatively Skewed:
How do Histograms
Work? (cont)
Bi-Modal Distribution:
Multi-Modal Distribution:
How do Histograms
Work? (cont)
Once your Histogram is complete,
you can analyze its shape, as well as
the statistics that you came up with.
mean (x)
Max
Min
Range (max-min)
Constructing a Histogram
Use range to estimate beginning and
end
Calculate the width of each column
by dividing the range by the number
of columns
Range = Width
# of Columns
Acme Pizza Example
Lets say the owner wants a
distribution of Acmes Thursday
Night Sales
Data Set from last Thursday(slices)
02122413121224341432232122122142212
12212121212121222121211222314223222
12322422441222322122421242172122312
1121222122121222424
Acme Pizza Example
Mean = 2.032258
Max = 7
Min = 0
Range = 7
Question
For 7 columns what would the width be?
Range/Columns=7/7=1 slice
Acme Pizza Example
Histogram
70 65
60
# times ordered
50
40 33
30
20
12
8
10
0 0 1
0
1 2 3 4 5 6 7
Slices of Pizza
Constructing a Histogram
How is this helpful to Acme?
2 slices of pizza most common order
placed
Distribution of sales useful for forecasting
next Thursdays late night demand
70
# times ordered
60
50
40
30
20
10
0
21 1
2 43 34 75 56 67
Slices of Pizza
Acme Pizza (part 2)
Critical Thinking
How does the Pareto Analysis differ
from the Histogram?
Percent
60
Count
40 of failure Design
30 40
20
20
10
0 0
Design Faults
Defect
100
Count 57 13 4 2
50
Percent 75.0 17.1 5.3 2.6
80
Cum % 75.0 92.1 97.4 100.0
40
Percent
60
Count
30
40
20
10 20
0 0
Defect
Count 21 10 8 8 5 3 2
Percent 36.8 17.5 14.0 14.0 8.8 5.3 3.5
Cum % 36.8 54.4 68.4 82.5 91.2 96.5 100.0
The Basic Seven (B7)
Tools of Quality
Flowcharts
Process Decision
Stay consistent
yes
Get Pizza
no
Lockup
Put More in
Oven 2 Pies
Available?
no
yes
Time
to close? no
Take to Customer yes
How can we use the flowchart to
analyze improvement ideas from
the Histogram?
Window Take Customer Money?
(start) Order
yes
Get Pizza
no
Lockup
Put More in
Oven 2 Pies
Available?
no
yes
Time
to close? no
Take to Customer yes
Want some practice?
Make a flowchart for:
Taking a shower
Cooking dinner
Driving a car
Having a party
Creating a Flowchart
Defective Pizzas
15
10
0
0 20 40 60 80
20
Defective Pizzas 15
10
0
0 20 40 60 80
Cyclical
8 9 10 11 12 1 2 3 4 8 9 10 11 12 1 2 3 4 8 9 10 11 12 1 2 3 4 Time
PM- AM PM- AM PM- AM
Thursday Thursday Thursday
The Basic Seven (B7)
Tools of Quality
Control Charts
Deviation from Mean
Range
Control Charts
Upper Limit
Lower Limit
Unacceptable
deviation
Control Charts
Acme Pizza Management wants to get
in on the control chart action
Average Diameter = 16 inches
Upper Limit = 17 inches
Lower Limit = 15 inches
Acme example
Control Charts
Upper Limit
17 inches
16 inches=X
Lower Limit
15 Inches
Small Pie
Acme example #50
Control Charts
Cause
Cause
Flow
Flow Check
Check Pareto
Pareto
Histograms
Histograms &&
Chart
Chart Sheet
Sheet Analysis
Analysis
Effect
Effect
Scatter
Scatter
Diagrams
Diagrams
Control
Control
Charts
Charts
Summary
Basic Seven Tools of Quality
Measuring data
Quality Analysis
Democratized statistics
Bibliography
Foster, Thomas. Managing Quality. An IntegrativeApproach.
Upper Saddle River : Prentice Hall, 2001.