Problem Solving Techniques

QC tools Seven Wonders

V.Thiagarajan, General Manager (TQC & HR), WABCO INDIA Limited

1 of 54

Significance of 7 QC tools
Unearthing problems in the workplace, analysing them, taking
countermeasures and establishing control requires some

special tools such as 7 QC tools

The 7 QC tools used in solving problems range from simple methods such as drawing graphs to sophisticated techniques requiring computers for their application

2 of 54

Significance of 7 QC tools - Continued

1. 7 QC tools are easy to use. They do not make people to do difficult calculations or complicated drawings. They can

be mastered in few hours of study

2. 7 QC tools are easy to understand and are immediately comprehensible visually. 3. 7 QC tools are capable of being used by everyone together.

Dr.Kaoru Ishikawa, the leading figure in development of Quality Control in Japan, used to say that in his experience everybody could solve 95% of the problems around them by using 7 QC tools skillfully
3 of 54

QC tools

4 of 54

Seven QC Tools
Scatter Diagram

Stratification
Histogram

Graph & Control Chart

Cause & Effect Diagram Pareto Diagram

Check Sheet

5 of 54

Application of 7 QC tools
Sl.No
1 2 3

Tools
Check Sheet / Data Sheet Pareto Daigram Cause & Effect Diagram Graph & Control Chart Histogram Stratification Scatter Diagram

Function
To Collect the data in a simple manner and to prevent omitting the checks To pick up the important few problems from the trivial many. To reorganise the factors (causes) which are influencing the problem Presentation of data in a pictorial form for better understanding and see if the process is under stable conditions. To see the distribution pattern compared against the standard values. To segregate data according to contributing sources.(Suppliers,machines,operators etc) To see relation between two sets of data
6 of 54

5 6 7

Check Sheet
Solder Bath Temperature
Checked by : Raised by : Take reading of temperature nearest to degree. Time Temp (0C) Time Temp (0C) 0800 60 1300 61 0900 62 1400 58 1000 59 1500 63 1100 58 1600 63 1200 59 1700 59 Note : Power failure from 12.00 to 12.15 Hours

7 of 54

What is Check Sheet ?

A Check sheet is a paper form on which item to be

checked have been printed already so that data can be

collected easily and concisely. Its main purposes are two-fold : 1) To make data-gathering easy ;

2) To arrange data automatically so that they can be used

easily later on

Two Types of Data :

1. Measurement data or Variable data. (Examples: Sales,

PBT%, Pressure, Temperature, Torque, Head count)

2. Attribute data. (Yes or No type Go / No go, IT return filed / Not filed)
8 of 54

Check Sheet example

Check sheet to collect data on various reasons for misplaced letters in a post office Data collected by: XYZ Period of data collection: November 2007
Defect Wrong mail box Wrong city Wrong Pin Code Nov 10th Nov 11th Nov 12th Nov 13th Total defects 16 18 25

Old office symbol

Total defects 17 17 17 16

8
67

9 of 54

Check Sheet example

Defect location check sheet examples

Painting defects happening in a car factory Data collected for defect reduction in door panel painting process

Defects happening in a ready made shirt Data collected for defect reduction

10 of 54

Benefits of Check Sheets

Data Collection have the following benefits : Right decision can be made.

Errors due to subjective feeling or personal bias are avoided.

Measurement understandable to all.

Assessments of magnitude of improvements can be decided.

Discovery of causes affecting quality and productivity.

11 of 54

Pareto Diagram
No of Components
300 250 200
156 85.28 91.53 96.75 100

100

60
105

150
50.98

100 50 0 Cold shut oil shot Damage Chip off Bad app
20 16

40 20
9

Types of Defects

% Cumulative

80

12 of 54

What is Pareto Diagram ?

Pareto diagrams are presentation technique used to

show facts and they help us to separate the vital few.

Pareto diagram was first thought out by an Italian economist,

Pareto (1848 - 1923), when he used it as a method for national

income analysis to show that a large proportion of the wealth was centered around a small minority of people. After that, an American Quality Control Authority,Mr.Juran used it in the field of Quality Control.

13 of 54

Pareto diagram example

Annual department budget requirements projected by various functional departments have been given to Finance department.
Finance department wishes to study the requirements and wishes to prioritise the allocation based on demands.

Department Marketing Manufacturing Purchase Ouality Assurance Prdocution Engineering Research and Development Human Resources Information Systems Others

Budget projection (Rs. In Crores) 6.25 15.75 33.55 3.45 65.15 21.50 5.65 4.15 5.00
14 of 54

Pareto diagram example

Requirements of three departments, Production engineering, Purchase and R&D account for 75% of the total demand.
Pareto of projected budget requirements of various departments
Budget requirement (Rs. in Cr.)
160 84.7 74.9 120 61.5 65.15 88.6 92.1 94.7 96.9 100.0 100

80

50 40.6 33.55 21.5 15.75 6.25

et in g

40

25 5.65
es

4.15
ys te m s

3.45
th er s

5 0

0
En gi ne er in g ev el op m en t M an uf ac tu ri n g ha se M ar k Pu rc ss ur an ce es ou rc O

In fo rm at io n

um an

Pr do cu tio n

an d

es ea rc h

Departments
15 of 54

ua li t y

Cumulative %

75

Pareto diagram example

Pareto diagram for various delays 50% of delays due to hydrotest

Before improvement

After improvement

Pareto diagram before and after the improvement Defects reduced. Defect A shifted to 3rd level after improvement

16 of 54

Cause & Effect Diagram

Man Machine

Characteristics

Material

Method

17 of 54

What is Cause & Effect Diagram ?

Any defect in a component,a product or service could be due to one or more causes.To find out the relationship between

the causes and effect, a diagram is drawn systematically by

mapping out all the probable causes influencing the effect. This is called a Cause & Effect Diagram Cause & effect diagram was introduced by Dr.K.Ishikawa. He used it in Kawasaki Iron Works in 1943. Since final diagram looks like a fish bone it is also called as Fish Bone Diagram. Also it is known as Fish Bone Diagram
18 of 54

Format of a Cause and Effect Diagram

Main Cause Main Cause A Level 3 Cause Level 2 Cause
Countermeasure will emerge by asking How? How?

Level 1 Cause

Characteristics

Main Cause

Main Cause

To arrive at root causes keep on asking Why? Why?

Above diagram depicts the basic format of Cause and Effect diagram. There is a hierarchical relationship of the effect to the main causes and their subsequent relationship to the subcauses.
19 of 54

Example of Cause-and-Effect Diagram:

Health
Relaxation Rest Amusement Time Sleep Quantity Nutrition Calories Meal Patience Carefulness Concentration Confidence Power Motion Speed Cooperation Schedule Exercise Quality Quantity Advice Form Repetition
20 of 54

Spirit

Encouragement
Pride Fighting spirit

Devotion Calmness
Composure

Depth
Information Theory Planning Rules Common sense Study of opponent Analysis

Defeat in a sports match

Observation

Teamwork
Function

Judgement of situation Experience of matches

Model

Strategy

Technique

Example of Cause-and-Effect Diagram:

21 of 54

Graph & Control Chart

UCL C/L LCL

22 of 54

What is Graph :
Graph is a pictorial representation of data which is easily understandable at a glance.It is a visual representation of data made up of points,lines,letters,words,numbers,shades etc.

23 of 54

Types of Graphs :
Line graph Bar chart / graph Pie chart or Circle graph Radar chart Compound graphs - bar and line Gantt chart

24 of 54

Line Graph:

Production details

25 of 54

Bar Chart / Graph :

10 9 8 7 6

6.3 4

Hrs

5 4 3 2 1 0 Jan'07 Feb'07

3.4

4 3

Mar'07

Apr'07

May'07

Jun'07

HMC Machine Breakdown

26 of 54

Pie chart or Circle graph:

Product sales value Consumer expenditure

27 of 54

Radar chart

Plant performance : 2007-2008

unit1
100% 100%

50%

unit5
85% 0%

unit2
80%

unit4
60%

unit3
90%

28 of 54

Compound graph - Bar and Line

Traffic census data

29 of 54

Gantt chart

Project plan

30 of 54

Control chart
It is a tool used to determine whether a manufacturing or business process is in a state of statistical control or not.

31 of 54

Histogram
40 34 38 35 30

25

20

15 9 9.9

10 4 0 0 4

4.5 2.5 2.5

3.3 3.35 3.4 3.45 3.5 3.55 3.6 3.65 3.7

32 of 54

What is Histogram:
Histogram is a graphical representation of a frequency distribution which is a summary of variation in a product or process.

33 of 54

Histogram example
Annual gross salary paid to all the employees are given below.
The company wishes to determine the frequency distribution of gross salary under different ranges.
Employee 1 Employee 2 Employee 3 Employee 4 Employee 5 Employee 6 Employee 7 Employee 8 Employee 9 Employee 10 Employee 11 Employee 12 Employee 13 Employee 14 Employee 15 107.45 339.45 925.10 109.56 421.50 476.15 481.25 265.75 241.95 817.17 831.45 635.35 675.45 620.25 525.25 Employee 16 Employee 17 Employee 18 Employee 19 Employee 20 Employee 21 Employee 22 Employee 23 Employee 24 Employee 25 Employee 26 Employee 27 Employee 28 Employee 29 Employee 30 568.25 571.45 346.85 325.15 278.50 212.87 335.65 375.41 442.78 459.10 511.40 521.26 820.15 715.30 948.75 Employee 31 Employee 32 Employee 33 Employee 34 Employee 35 Employee 36 Employee 37 Employee 38 Employee 39 Employee 40 Employee 41 Employee 42 Employee 43 Employee 44 Employee 45 588.28 765.45 650.51 777.87 917.45 641.56 586.15 619.81 735.75 576.45 625.16 630.46 611.45 561.29 106.85
34 of 54

Gross salary Rs. x1000

Histogram example
Annual gross salary paid to all the employees are given below.
The company wishes to determine the frequency distribution of gross salary under different ranges.
Histogram of salary of employees
10 9 8 7

Frequency

6 5 4 3 2 1 0 Less than 100 100 to 200 200 to 300

5 4 3

5 4 3 3

0
300 400 400 to 500 500 to 600 600 to 700 700 to 800 800 to 900 More than 900

Gross salary range ( x 1000 Rs.)

35 of 54

Interpretation of Histogram
Minimum Limit Average or Mean
1

Maximum Limit

Minimum Limit Mean shifted and high variation process

Average or Mean
4

Maximum Limit

Controlled process
2

Mean shifted process

Very high variation process

Process with mix up of two populations

Process on the verge of going out of control

36 of 54

Uses of Histogram :

To know the pattern of variation.

To assess states of control To assess conformance to specifications To assess spread or variation with reference to specification. To assess process capability.

To get clues for bringing process under control - whether to

shift mean or to reduce variation or both. To get clues for possible assignable causes for observed variation - mixture of lots,suppliers,instruments / measurements errors,bias of written results etc.
37 of 54

Stratification
Quality Defective (Nos)
50 40 30 20 10 0
A B C D E

48%

16%

18% 12% 6%

Types of Defect

38 of 54

What is Stratification:
Stratification is the process of separation of data into

categories.It is normally done for identifying the categories

contributing to the problem tackled.

39 of 54

Types of Stratification:
Material Based:
48%
50 40

Defective %

30 20 10 0
A B C

16 %

15%

Supplier

Defective Quality - Supplier Wise.

40 of 54

Types of Stratification:
Quality Based:
Quality Defective(Nos)
50 40 30 20 10 0
A B C D C

48%

16%

18% 12% 6%

Type of Defect

Defects Data For Each Hour in a Shift

41 of 54

Types of Stratification:
Operator Based:
30

No of Defective

25 20 15 10 5 0
A

22%

5%

3%
B C

Operator

Defective Assemblies
42 of 54

Types of Stratification:
Machine Based:
100 90 80 70 60 50 40 30 20 10 0

Production %

57% 43%

Machine

Production Machine Wise

43 of 54

Types of Stratification:
Processing Based:
Power Consumption %
50 40

32%
30 20

25% 12%

26%

10 0
A B C

5%
D E

Process Area

Power Consumption Process Area Wise

44 of 54

Scatter Diagram
4.5 4.4

Overall rating of the Session

4.3 4.2 4.1 4 3.9 3.8 3.7 3.6 3.5 3.4 100 150 200 250 300 350 400 450 500 550 600 650

4.0 y axis 430 x axis

Average Experience of Training Team (in days)

45 of 54

What is Scatter Diagram:

Scatter diagram is a graphical representation of relationship between two variables. It can be between a Cause & Effect and between causes.
Variable 2 Variable 2 Variable 1

Variable 1

In 1837 J.F.W.Herswchal,an Englishman,used scatter diagram. In 1950s Dr.K.Ishikawa popularized the use of scatter diagram.
46 of 54

Type of relationships
a. Positive relationship (Positive correlation)

b. No correlation.

c. Negative relationship (Negative correlation)

d. Positive and Negative correlation

e. Negative and Positive correlation

47 of 54

Scatter diagram examples

n = 30
y 6 4 2 0 0 2 4 6 x

r = 0.9
y
6 4

n = 30

r = 0.6
y 6 4 2 0

n = 30

r = 0.0

2
0 0 2 4 6 x

Positive Correlation

Positive Correlation May Be Present

No Correlation r is correlation coefficient

n = 30
y 6 4 2 0 0 2 4

r = - 0.9
y 6 4 2 0 6 x 0

n = 30

r = - 0.6

r is positive for positive correlation r is negative for negative correlation r can vary from 1 to +1

Negative Correlation

Negative Correlation May Be Present

If r is more than 0.7, or less than 0.7, the correlation is very strong
48 of 54

Scatter diagram example

A company has collected data on R&D expenditure and profit attained for several years.
The company wishes to determine, whether R&D expenditures affect profit. Also the company wants to estimate the profit, if the R&D expenditure is raised to 12 Million rupees.

R&D expenditure 5 3 7

Profit attained 30 40 60

6
10 4

60
80 40
49 of 54

Scatter diagram example

Scatter diagram showing relationship between R&D expenditure and profit

80

Profit - Rs. in Million

60 40 20 y = 6.5405x + 13.514 0 0 3 6 9 R&D expenditure - Rs. in Million 12

R&D expenditure has positive correlation with Profit Correlation coefficient is 0.885 If R&D expenditure is increased to 12 Million Rupees, the expected profit can be extrapolated as, Profit = 6.5405 x 12 + 13.514 = 92 Million Rupees
50 of 54

Scatter diagram example

A super market that has a chain of 15 retail outlets in a city, wants to predict the monthly sales behaviour based on the number of customer arriving at the outlet. So data was collected from each of the 15 outlets for a month. The data is presented here
Retail Outlet No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Number of customers Sales in visited Rs. Lakhs 1814 22.4 1852 22.1 1012 13.68 1482 18.42 1578 18.84 1778 20.16 1748 18.9 1020 13.46 1058 14.48 840 12.24 1358 15.26 1744 18.86 1848 18.92 1214 15.28 904 13.84

Can we help the supermarket by analyzing the data to find out whether there is any correlation between number of customer visiting and sales value. Also to estimate how many customers have to visit the market, if sales value has to be Rs. 24 Lakhs.
51 of 54

Scatter diagram example

Sales value - Rs. In Lakhs
Scatter diagram showing relationship between No. of customers visiting supermarket and sales value

24 21 18 15 y = 0.0082x + 5.4891 12 800 1100 1400 1700 No. of customers visiting super market 2000

Number of customers visiting the super market has a strong positive correlation with sales, with a correlation coefficient of 0.95 If the super market wants to achieve a sales of Rs. 24 Lakhs, the number of customers, who will visit the market has to be = (24 5.4891) / 0.0082 = 2257 persons
52 of 54

Summary
Essential tools in the continuous Improvement Process. Provides objective analysis of problems based on facts and data Encourages and enhances teamwork as problems are addressed through groups

Able to anticipate potential problems and improve quality

Greater customer satisfaction through superior quality

53 of 54

Thank You

54 of 54