Use of data , display of data

& The Basic Seven (B7)

Tools of Quality Improvement

"As much as 95% of quality related problems in the factory can be solved
with seven fundamental quantitative tools." - Kaoru Ishikawa
Why Are We Doing This?
The Deming Chain

Improve Quality
Decrease Costs
Improve Productivity
Decrease Price
Increase Market
Stay in Business
Provide More Jobs
Return on Investment
 Six Problem Solving Steps
 Identify
 recognize the symptoms
 Define
 Agree on the problem and set boundaries
 Investigate
 Collect data
 Analyze
 Use quality tools to aid
 Solve
 Develop the solution and implement
 Confirm
 Follow up to ensure that the solution is effective
“We need research and development focused on
information integrity, data standardization, data capture,
and use.”

“In God we trust.

All others bring data.”

W. Edwards
Deming
Certified Health Data (Clinical
Data) Analyst

A New Role
 Health Data Analyst Duties &
Responsibilities
 Be responsible for the implementation, operation and
enhancement of clinical data validation systems and
processes
 Participate in ongoing clinical data analysis and
documentation
 Participate in the ongoing invention, testing and use of
clinical data quality systems
 Work closely with Service management in clinical data
areas
 Health Data Analyst Required
Experience and Skills
 Bachelors Degree in a patient care or health information management
discipline or work equivalent
 Experience working directly with physicians, nurses, and other healthcare
personnel
 Minimum of 2 years of clinical data management experience
 Working knowledge of health information systems
 Working knowledge of standard clinical terminologies and coding systems
 Ability to analyze user requests, define requirements, develop project plans
and report conclusions
 Ability to work creatively and flexibly, both independently and as part of a
team
 Attention to fine details and work processes
 Desire and ability to learn new skills, systems and processes
 Good organizational, and written and oral communications skills
Data Capture & Sharing
The Basic Seven (B7)
Tools of Quality Improvement
Where did the Basic Seven
come from?

Kaoru Ishikawa

 Known for “Democratizing Statistics”

 The Basic Seven Tools made statistical analysis less

complicated for the average person
What are the Basic Seven Tools
of Quality Improvement?
 Fishbone Diagrams
 Histograms
 Pareto Analysis
 Flowcharts
 Scatter Plots(diagram)
 Run Charts
 Control Charts
 Fishbone Diagram

 No statistics involved

 Maps out a process/problem

 Makes improvement easier

 Looks like a “Fish Skeleton”

 Cause & Effect Diagrams
Sample
Manpower Materials

Typos
Source info incorrect
Wrong source info
Didn’t follow proc.

Po
Dyslexic
Wrong purchase order
or
Transposition
t ra
in i
ng Incorrect
shipping
Glare on documents
Temp. display Corrupt
data
No training
Environment
No procedure
Keyboard sticks
No communications

Software problem

Methods Machine
Fishbone Diagram Overview
 Definition
 Uses
 Ishikawa

 Use within organizations

 benefits

 Creation of the Diagram

 Steps 1-9
 Example
 Service example
 Exercise
 Ham Industries
Fishbone (Cause and Effect or
Ishikawa) Diagrams (1 of 4)

 Named after Kaoru Ishikawa

 Japanese Quality pioneer
 Resembles skeleton of a fish
 Focus on causes rather than symptoms of a
problem
 Emphasizes group communication and
brainstorming
 Stimulates discussion
Fishbone (Cause and Effect or
Ishikawa) Diagrams (2 of 4)

 One of Seven basic tools of Japanese

Quality
 Leads to increased understanding of
complex problems
 Visual and presentational tool
Fishbone (Cause and Effect or Ishikawa)
Diagrams (3 of 4)

 Typically done on paper or chalkboard

 Recently some computer programs have
been created to make Fishbone Diagrams
Use in Organizations
 Can be used to improve any product,
process, or service
 Any area of the company that is experiencing
a problem
 Isolates all relevant causes

 Helps bring a problem into light

 Group discussion and brainstorming
 Finds reasons for quality variations, and the
relationships between them
Creating Fishbone Diagrams
(1 of 4)

• As a group:
1. Establish problem (effect)
-state in clear terms
-agreed upon by entire group

2. Problem becomes the

“head” of the fish
-draw line to head (“backbone”)
Creating a Fishbone Diagram
(2 of 4)

3. Decide major causes of the

problem
- by brainstorming
- if the effect or problem is part of a process the
major steps in the process can be used

4. Connect major causes to backbone of

the fish with slanting arrows
Creating a Fishbone Diagram
(3 of 4)

5. Brainstorm secondary causes for

each of the major causes
6. Connect these secondary causes to
their respective major causes
7. Repeat steps 5 & 6 for sub-causes
dividing with increased specificity
- usually four or five levels
Creating a Fishbone Diagram
(4 of 4)

8. Analyze and evaluate causes and sub-

causes
-may require the use of statistical, analytical, and
graphical tools
9. Decide and take action
Example (1 of 4)
 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

accuracy Poor Service

One on one
courtesy service
dependability

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
 Histograms

 Bar chart

 Used to graphically represent groups of data

Overview

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?
 A Histogram is a variation of a bar chart in
which data values are grouped together
and put into different classes.
 This grouping allows you see how
frequently data in each class occur in the
data set.
 Higher bars represent more data values
in a class.
 Lower bars represent fewer data values in
a class.
Example of a Histogram
 Uses for a Histogram
A Histogram can be used:
 to display large amounts of data values in a
relatively simple chart form.
 to tell relative frequency of occurrence.
 to easily see the distribution of the data.
 to see if there is variation in the data.
 to make future predictions based on the data.
How do Histograms Work?
 First, you need to pick a process to analyze.
 Next, you need a large amount of data, at least
100 data values so that patterns can become
visible.
 Then, you need to assemble a table of the data
values that you collected with regards to
frequency of data values.
 Next, you need to calculate some statistics for
the Histogram, including: mean, minimum,
maximum, standard deviation, class width,
number of classes.
 How do Histograms Work?
(cont)
 After you have created a Histogram, it
will take one of five shapes:

 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.

 This analysis will help you to make better

decisions toward quality improvements.
SEE YOU NEXT LECTURE

Thank You
 Pareto Analysis

 Very similar to Histograms

 Use of the 80/20 rule

 Use of percentages to show importance

Pareto Principle
 Vilfredo Pareto (1848-1923) Italian
economist
 20% of the population has 80% of the wealth
 Juran used the term “vital few, trivial
many”. He noted that 20% of the quality
problems caused 80% of the dollar loss.
Histogram
25

20
Frequency

15

10

Category
 Chart
Pareto

Percent from each cause

10
20
30
40
50
60
70

0
Po
or
W De
ro s
n g ig
di n
(64)

m
De en
f ec
s io
ns
tiv
(13)

M e
ac pa
hi rts
ne
(10)

ca
O lib
pe ra
ra t io
to
(6)

re ns
De rro
f ec rs
tiv
(3)

e
m
Causes of poor quality

Su at
rfa er
ce ia
ls
(2)

ab
ra
s io
n s
(2)
Pareto Diagram
Method Usage Result

Pareto Diagram Used to identify and Allows clarification of

(No. of Occurrences)
prioritize a problem. important tasks.
Allows identification of
Used to review the effects
a starting point (which
of an action taken. task to start with).
Allows projection of
Used to prioritize the effects of a
actions.(Used during measure to be taken
phases to monitor the
situation, analyze
causes, and review
effectiveness of an
action.)
Pareto Analysis, how to use it
 1. Gather facts about the problem, using Check Sheets or
Brainstorming, depending on the availability of information.

 2. Rank the contributions to the problem in order of

frequency.

 3. Draw the value (errors, facts, etc) as a bar chart.

 4. It can also be helpful to add a line showing the

cumulative percentage of errors as each category is
added. This helps to identify the categories contributing to
80% of the problem.

 5. Review the chart – if an 80/20 combination is not

obvious, you may need to redefine your classifications and
go back to Stage 1 or 2.
 Pareto Charts
Purpose:
Prioritize problems.

How is it done?
 Create a preliminary list of
problem classifications.
 Tally the occurrences in each
problem classification.
 Arrange each classification
in order from highest to
lowest
 Construct the bar chart
 Pareto Charts
Benefits: 120

 Pareto analysis 100

helps graphically
display results so 80

the significant few

Quantity
problems emerge 60

from the general

40
background
 It tells you what to 20

work on first
0
Dent Scratch Hole Others Crack Stain Gap
Defects 104 42 20 14 10 6 4
 Pareto Charts

Pareto Charts Weighted Pareto

 Weighted Pareto charts
use the quantity of
defects multiplied by 900

their cost to determine 800

the order. 700

Weighted 600
Defect Total Cost cost
Gap 4 200 800 Weighted Cost 500

Dent 104 2 208

400
Hole 20 5 100
Crack 10 8 80 300
Scratch 42 1 42
Others 14 1 14 200

Stain 6 1 6
100

0
Gap Dent Hole Crack Scratch Others Stain
Weighted cost 800 208 100 80 42 14 6
Acme Pizza (Example 1)
Slices Frequency %
0 1 .3
1 33 13.09
2 65 25.79
3 8 3.17
4 12 4.76
5 0 0
6 0 0
7 1 .3
Acme Pizza (Example 1)
 The completed Pareto Analysis results in the following graph:

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?

 How can this be a useful tool to the Acme

boss?
A series of Pareto charts drill down
to more detail (Example 2) :
Fault by Main Cause
1st level
70
100 Analysis 2nd level
60 80 gives “Design” Analysis gives
50 as main cause breakdown of

Percent
60
Count

40 of failure “Design”
30 40

20
20
10

0 0
Design Faults
t
n en er
Defect s ig on ild
De mp Bu Oth
Co
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
le n
ule rs rt du atio
od oto Sta Mo libr n
Defect nn
ec
tM
rqu
eM
Co
ld du
ce
r
IC
Ca IOP Imo
Co To ns AS
Tra
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
 Flowcharts
 A graphical picture of a PROCESS

Process Decision The process flow

 Flow Charts
Purpose:
Visual illustration of the sequence of operations required to
complete a task
 Schematic drawing of the process to measure or improve.
 Starting point for process improvement
 Potential weakness in the process are made visual.
 Picture of process as it should be.
Benefits:
 Identify process improvements
 Understand the process
 Shows duplicated effort and other non-value-added steps
 Clarify working relationships between people and organizations
 Target specific steps in the process for improvement.
 Flow Charts
Top Down
Benefits
• Simplest of all flowcharts
• Used for planning new
processes or examining Measure Analyze Improve Control
existing one
Hardware Fleet leader
• Keep people focused on Problem report Customer input procurement reports
the whole process Hardware return Stress analysis
Customer
Service reports
coordination
How is it done? Failure analysis
Heat transfer Compliance Operational
analysis verification statistics
• List major steps
• Write them across top of Life analysis Documentation

the chart Substantiation FAA approval

• List sub-steps under each
in order they occur
Benefits
Flow charts
Linear
 Show what actually happens
at each step in the process
 Show what happens when
non-standard events occur Toolbox
 Graphically display
processes to identify
redundancies and other
wasted effort
How is it done?
 Write the process step inside
each symbol
 Connect the Symbols with
arrows showing the direction
of flow
 Flow chart
Sample Linear Flow

END
5 - Action Assignee
performs detail
analysis of failure. No
Start Requests failure
analysis as needed. 11 - Fleet Analysis
Still monitors failure to
failing? ensure corrective
action is effective.
1- Fleet Analysis
Yes
utilizes data
Yes
warehouse reports to 6 - Action Assignee
create and distribute documents
a selection matrix. 10 - FRB determines
investigation
required corrective
findings.
action - i.e. QAM or
supplier corrective
2 - Other Groups action.
compile data as
determined by FRB.
7 - Action Assignee 9 - FRB Categorize
reports investigation Failure: Workmanship,
3 - FRB meets to results to FRB. Still component, material,
No
analyze data. failing? maintenance, or
design. Also fleet
wide or RSU.

4 - FRB selects 8 - Fleet Analysis

candidate problems monitors failed item
for additional to ensure failure has
investigation. been corrected.
Flowcharts
Don’t Forget to:

 Define symbols before beginning

 Stay consistent

 Check that process is accurate

 Scatter Plots(Diagram)

2 Dimensional X/Y plots

 Used to show relationship between
independent(x) and dependent(y)
variables
Scatter Diagram
1. What is Scatter Diagram ?

A scatter diagram is used to “examine the relationship between

the two, paired, interrelated data types, ” such as “height and
weight of a person.” A scatter diagram provides a means to
find whether or not these two data types are interrelated. It is
also utilized to determine how closely they are related to
identify a problem point that should be controlled or improved.
 Scatter Diagram

2. When is it used and what results will be obtained?

The table on next slide shows some examples of scatter diagram’s
usage. If, for example, there is a relationship where “an increase in the
number of rotations (x) causes an increase in abrasion (y),” there exists
“positive correlation.” If, on the other hand, the existence of a
relationship where “an increase in the number of rotations (x) causes a
decline in abrasion (y)” indicates that there is “negative correlation.”

59
Scatter Diagram
Method Usage Result
•Used to identify a relationship
Scatter Diagram
between two matters. •Can identify
•Used to identify a relationship
ｙ Axis
between two matters and
cause and effect
establish countermeasures relation.
Abrasion

based on their cause and effect •Can understand

relation.
the relationship
Example Usage
1.Relationship between thermal between two
x Axis
Number of treatment temperature of a steel results.
Rotations material and its tensile
strengths
2.Relationship between visit
made by a salesman and
volume of sales 3.Relationship
between the number of persons
visiting a department store and
volume of sales
4 Others
[Used during phases to monitor the
situation, analyze causes, and
review effectiveness of an action.]
Scatter Diagram

ｙ Axis
Abrasion (micron)

x Axis
Number of
Rotations
Acme Pizza
(Scatter Diagram) Example 1
Minutes Cooking Defective Pies
10 1
45 8
30 5
75 20
60 14
20 4
25 6

In this simple example, you can find the existing relationship without
much difficulty but…
Scatter Diagrams
25
•Easier to see direct
relationship 20

Defective Pizzas
15

10

0
0 20 40 60 80

Time Cooking (minutes)

Scatter Diagrams
As a quality tool
 What does this tell Acme management
about their processes?
 Improvements?
25

20

Defective Pizzas 15

10

0
0 20 40 60 80

Time Cooking (minutes)

 Run charts

Measurement
What is it?
 Run Charts are representing change
Time
 in measurement over a sequence or time
How does it Work?
 Gather Data
 Organize Data
 Measurements (y) must be confronted with time or
sequence of the events.
 Chart Data
 Interpreting Data
What is its use?
 Determining Cyclic Events and there average character
Run Charts
Example
 Oil consumption of a specific machine over a period of time.
Measurement

Time
 Control Charts

 Deviation from Mean

 Upper and Lower Spec’s

 Range
Control Charts
Y

What is it? Upper limit

Average/Spec
 Statistical tool, showing whether
Lower limit
a process is in control or not

How does it Work? X

 Define Upper limit, lower limit and medium value

 Draw Chart.
 Gather values and draw them into chart

What is its use?

 Taking samples of a process and detect possibility of process being out of
control
Control Charts

Y
Upper limit

Average/Spec

Lower limit

X

NB:
NB:
Check
Check Sheet
Sheet
Shifts

   

 
Defect Type

 

 

NB:
NB:
Check
Check Sheet
Sheet
COMPONENTS REPLACED BY LAB
TIME PERIOD: 22 Feb to 27 Feb 1998
REPAIR TECHNICIAN: Bob

TV SET MODEL 1013

Integrated Circuits ||||
Capacitors |||| |||| |||| |||| |||| ||
Resistors ||
Transformers ||||
Commands
CRT |

NB:
NB:
Check
Check Sheet
Sheet
Billing Errors Monday
Wrong Account

Wrong Amount

A/R Errors

Wrong Account

Wrong Amount
Logical Order for B7 Tools
Big Data Data Problem
Picture Collection Analysis Identification Prioritization

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”
Thank You