7QC tools

OBJECTIVE

Objective of this training:

 Understand the basic concept of 7QC tools
 Able to identify which tools suitable for application
 Develop working plan using appropriate tools
INTERMISSION
7QC tools
Seven QC tools are fundamental instruments to improve the
quality of the product. They are used to analyze the
production process, identify the major problems, control
fluctuations of product quality, and provide solutions to
avoid future defects. Statistical literacy is necessary to
effectively use the seven QC tools. These tools use
statistical techniques and knowledge to accumulate data
and analyze them
Quality Control
 7 QC TOOLS

1. Stratification
2. Check sheet
3. Pareto diagram
4. Histogram
5. Fishbone diagram
6. Scatter diagram
7. Control Chart
1. STRATIFICATION
 STRATIFICATION
Definisi:
 Stratification is a technique used in combination with other data
analysis tools. When data from a variety of sources or categories
have been lumped together, the meaning of the data can be
impossible to see. This technique separates the data so that
patterns can be seen.

E.g: General stratification categorize data according to

1. Who: Department, individual, type of customer
2. What: Type of complaint, category of defect
3. When: Monthly, weekly, daily
4. Where: part, specific location
 STRATIFICATION (DATA ANALYSIS)
Example
WITHOUT STRATIFICATION WITH STRATIFICATION

REJECTED

REJECTED

REJECTED

Conclusion:
Line C equipment has the most rejected items (75 unit)
Stratification video
 EXAMPLE : STRATIFICATION
WITHOUT STRATIFICATION WITH STRATIFICATION

Reviewing the two figures above, what is your Conclusion?

 Activity
TV Manufacturing Company
Sub
Sub Woofer
Woofer Sub
Sub Woofer
Woofer

TV-A TV-B TV-C

A TV manufacturing company owns 4 different of products. Every product has 4

production lines with a capacity of 1000 units per day, running non stop for 4 TV products
with different sizes from small to big with 3000 workers. A few suppliers are involved in the
supply chain.

Then at one time, in a month a total of 2400 customers complaint about spotty pictures.
The company had to recall all TV in all sizes from the market. How would you try to sort
the defects? What data that could be potentially stratified?.
2. CHECK SHEET
 CHECK SHEET
Definition:
Planned action to collect data

Benefit:
1. Assist and simplify data collection
2. Standardize data collection
3. To track event
4. To identify problems
CHECK SHEET
HOW TO CREATE:
Use the basic questions
(5W+1H):
1. What?
2. Where?
3. Who?
4. When?
5. Why?
6. How?
CHECK SHEET
Example : Telephone interruptions for front desk
 PLANNING FOR DATA COLLECTION
(ROLE OF CHECK SHEET IN DATA COLLECTION)
1. Select What to Measure
2. Develop Operational Definitions
3. Identify Data Sources
4. Prepare a data Collection and Sampling
Plan
5. Implement and Refine the data collection
process
PLANNING FOR DATA COLLECTION
1. Select what to measure

Criteria for selecting measures/ parameter data:

• Value / usefulness
 Link to high priority customer requirements
 Accuracy of the data
 Areas of concern or potential opportunity
 Can be benchmarked to other organizations
 Can be helpful ongoing measure
• Feasibility
 Availability of data
 Lead time required
 Cost of getting the data
 Complexity
 Likely resistance to “fear factor” associated with a particular type of
measure
PLANNING FOR DATA COLLECTION
1. Select what to measure
TOOL : MEASUREMENT ASSESSMENT TREE

STEPS

1. Identify a customer related defect in a key output, and write it above

the designated line on the chart
2. Brainstorm a list of questions that relate to that defect, and write
them on the left side of the tree
1. What patterns do you suspect you might find?
2. What factors do you think might influence the type or amount of that
defect?
3. Identify stratification factors that will help you answer the questions
about the output. Write this on the branches to the right of output
4. Identify specific types of data you could collect that would answer
the question of how the stratification factor did or did not affect the
output
 PLANNING FOR DATA COLLECTION
1. Select what to measure
TOOL : MEASUREMENT ASSESSMENT TREE

STEPS

5. When the diagram is complete, review each of the metrics and rate
them as follows:

1. Y : metric potentially help to predict the output

2. Y : data exist to support this metric

6. Use this analysis to decide which of the metrics will be most useful
for your project
PLANNING FOR DATA COLLECTION
1. Select what to measure
TOOL : MEASUREMENT ASSESSMENT TREE

What are selected measures ?

 PLANNING FOR DATA COLLECTION
2. Develop Operational Definition

Operational definitions : a clear, understandable description of what’s to

be observed and measured, such that different people taking or
interpreting the data will do so consistently

Elements of an operational DEFINITION

1. What you are trying to measure
• Number of surface defects on the rear panel
2. What the measure isn’t
• Does “surface defects” include smears or only scratches and dents?
3. Basic Definition of the measure
• Surface defect = any dent or scratch visible from a distance of 3 feet under
normal light
4. How to take the measurement (procedures)
• …………………..
 PLANNING FOR DATA COLLECTION

3. Identify Data Sources

• Historical data
• New data

4. Prepare a data Collection and sampling Plan

• Identify or confirm the stratification factors
• Develop a sampling scheme
 Random sampling
 Stratified sampling : ex. dividing 1000 customer into four groups: large, medium
small, and infrequent buyers
 Systematic sampling ; ex. every half hour or every 20 th item)
• Create data collection form
 Keep it simple
 Label it well
 Include space for date , time, and collector’s name
 Organize the data collection form and compiling sheet
 Include key factors to stratify the data
 PLANNING FOR DATA COLLECTION

5. Implement and Refine the data Collection Process

• Review and finalize your data collection plans
• Prepare the workplace
• Test your data collection procedures
• Collect the data
• Monitor accuracy and refine procedures as appropriate
example
• Data Set from last Thursday(slices)
021224131212243414322321221221422121221212121
21212221212112223142232221232242244122232212
24212421721223121121222122121222424
• These numbers represent the customers order at the order window at the
pizza store.
• For example, the first customer didn’t order any pizza, the second ordered
2 slices, the third ordered 1 and so on.
• It should be noticed that the highest order was 7 slices and the lowest was
0. This is used to find the range which is used to find the column width
for the histogram.
• HOW TO IMPROVE THE DATA COLLECTION METHODS
ABOVE?
3. PARETO CHART
 PARETO CHART

DEFINITION:
A Pareto chart is a bar graph.
The lengths of the bars represent frequency or cost (time or
money), and are arranged with longest bars on the left and the
shortest to the right. In this way the chart visually depicts
which situations are more significant.
 PARETO CHART

BENEFIT:
 Improvement or solution could be based on
principle of “Vital Few” from “Trivial Many”.
 Focusing resource on cause or problems that
could yield highest return
 Comparing frequency from different root cause
 PARETO CHART

Conclusion: Biggest defect is Chipping (40 pcs = 37.38%)

 COMPARING PARETO
Data before
Data after implementation
implementation
20 Cum 20
Week 5 – Week 6
Week 1 – Week 2
100%
15 15

10 Cumulative
10 10
100%
5 5
5 5 3
3
2 2
1
0 0
Daun Tnm Tnm Ber Tnm Daun Tnm Tnm Ber Tnm
me- Kerdil - Tumbang me- Kerdil - Tumbang
ngunin Penyakit ngunin Penyakit
g g

Conclusion

How many percent of improvement?

An example case study

Team should focus on 3 biggest

problem that contributes to 77.1%
of the total problems
Synergy effect of Pareto
1 Improvement effects all

On the first project, team concentrates on 1 object. It also effects other objects
4. Histogram
 HISTOGRAM
DEFINISI:
A frequency distribution showing how often each different value
in a set of data occurs. A histogram is the most commonly
used graph to show frequency distributions.
BENEFIT:
Useful to test form and distribution of data:
 To see range and distribution of data (example: weight of sent
items, purchasing in PO)
 To see variation and level of completion in
specification/customer satisfaction (size, cycle time,
temperature, dsb).
EXAMPLE : HISTOGRAM
DATA FOR HISTOGRAM
(EMPLOYEE HEIGHT - WEIGHT (KG)) CHECK SHEET FOR HISTOGRAM

Number of data values = n = 100  k = sqrt (100) = 10

Range = max – min = 124 – 77 = 47
H = Range / k = 47 / 10 = 4,7
HISTOGRAM
CONTOH
 HISTOGRAM
STEPS
 Determine variable from a measurement. Example : Time, Height
measurement, width.
 Collect data 50 - 250 points
 Complete table of data frequency
 Calculate total data = n
 Determine range data, max – min = r
 Determine interval/total bars from origin n = k
 Determine distance every interval , r/k = h
 Form table of data based on values a – d
 Input data into histogram
 HISTOGRAM
STEPS
 Give label on X axis with value on every bar (from first bar until
bar k)
• First bar : min + h
• Second bar : min + 2h
• …..
• kth Bar : min + (kxh)
 Calculate total data from evey bar
 Vertical axis (Y) shows total data from evey bar
 Horizontal axis (X) shows value from evey bar
 Analise histogram and study the pattern distribution
 HISTOGRAM PATTERN
S

er
35
S
30
er
25
S
20 er

15 S

er
10

Normal

Normal Distribution (Symmetrical)

You could give customer specification on the histogram visually and you could know how far the process
satisfy customer requirements.

You could also show average values and standard deviation as numbers that represent process.
HISTOGRAM PATTERN
25

20

15

10

Skewed Distribution

Data approaches one end of the axis

Example: Process time, cycle time

Analyse condition on what differentiate one end and another. If this

condition is not desirable, improvement need to implemented. Eliminate
cause but if this is the intended condition maintain result.
30

25

20

15

10

Bimodal distribution

Data has two peaks

Pattern appears when assumption of 1 process is actually 2 process

Stratification could help identify source of each data

30

25

20

15

10

Evenly distributed data values

Rarely happens

Pattern appears when a gauge or measuring tool is no longer

sensitive in detection of difference between units
 EXAMPLE : HISTOGRAM
51 48 51 60 51
CASE:
You work in a soy sauce 49 52 57 63 51
manufacturing company. A total of 50 45 57 54 49 50
samples of soy sauce were 44 51 54 47 58
measured for its salinity. Make a
40 60 44 45 55
histogram from the information given
41 50 47 48 43

50 42 52 38 52

44 43 61 51 52

40 50 53 48 53

51 52 49 50 43
 Contoh Histogram perbandingan

Conclusion?
5. FISHBONE DIAGRAM
 CAUSE AND EFFECT DIAGRAM

DEFINISI:
The fishbone diagram identifies many possible causes for an
effect or problem. It can be used to structure a brainstorming
session. It immediately sorts ideas into useful categories.
 Could also be used to brainstorm methods to achieve a
purpose
 Developed in 1943 by Prof Kaoru Ishikawa. Often known as
Ishikawa diagram of fishbone diagram because it is similar to
a fishbone
 CAUSE AND EFFECT DIAGRAM

BENEFIT:
1. Identifying major causes to a problem
2. Identifying root cause to a problem
3. Providing alternatives to problem solving
method
CAUSE AND EFFECT DIAGRAM
 STEPS IN USING CAUSE AND
EFFECT DIAGRAM
1. Agree on a problem statement (effect). Write it at the
center right of the flipchart or whiteboard. Draw a box
around it and draw a horizontal arrow running to it.
2. Brainstorm the major categories of causes of the
problem. If this is difficult use generic headings:
Method
Machines (equipments)
People (manpower)
Materials
Environment
Metrology
3. Write the categories of causes as branches from the
main arrow.
 STEPS IN USING CAUSE AND
EFFECT DIAGRAM

4. Brainstorm all the possible causes of the problem. Ask:

“Why does this happen?” As each idea is given, the
facilitator writes it as a branch from the appropriate
category. Causes can be written in several places if
they relate to several categories.
5. Again ask “why does this happen?” about each cause.
Write sub-causes branching off the causes. Continue
to ask “Why?” and generate deeper levels of causes.
Layers of branches indicate causal relationships.
6. When the group runs out of ideas, focus attention to
places on the chart where ideas are few.
Contoh 1
 Contoh 2

Selected Root cause is:

Employee medications requests ( the number of employee prescription requests by telephone to the pharmacy was causing a
delay in medication delivery
6. SCATTER DIAGRAM
 SCATTER DIAGRAM

DEFINITION:
The scatter diagram graphs pairs of numerical data,
with one variable on each axis, to look for a
relationship between them. If the variables are
correlated, the points will fall along a line or curve.
The better the correlation, the tighter the points will
hug the line.
 SCATTER DIAGRAM

BENEFIT:
 Present data to confirm a hypothesis on a correlation between
2 variables
 To understand how tight the correlation between 2 variables
 As a tool to verify root cause from using cause and effect
diagram
 SCATTER DIAGRAM
EXAMPLE CORRELATION:
 Between customer visit to sales result
 Between customer complaint and revenue
 Between hours of work and work result
 Between number of salesman to number of sales
 Between age of machine and total breakdown time
 Between number of samples inspected with total defects
 Between frequency of repairs and number of rejects
 Between inventory levels and wastes
 Between training hours and work related accidents
TYPES OF CORRELATION BETWEEN TWO
TYPES OF
VARIABLES
SCATTER DIAGRAM

Positive Possible positive No correlation

correlation correlation

Negative correlation Possible negative Dual curve correlation

correlation
Total
Sales

Number of visits

Conclusion: because “r” = 0.735 close to 1, it could be said that

variable x and y are strongly correlated
7. CONTROL CHART
 CONTROL CHART

Definition
 Tools to determine whether a process is in a state
of statistical control
 CONTROL CHART

Benefit
 Helps us to see visually whether we have a
capable process
 Whether our process is deviating from
specification
CONTROL CHART