Corporate

Quality

Statistical Process Control

Rules for the program

Corporate
Quality

Put the mobiles switched off / Silent mode.

Interaction is much more important
Maintain the break timings

Program Agenda

Corporate
Quality

TS, Medical & Aerospace requirements on SPC?

What is defect, defective?
What is variation?
What is the common cause and special Cause?
What is meant by Population and Sample?
What are the measures of central tendency & Dispersion?
What is standard deviation?
How Sigma differs from mean, median and mode?
What is variable and attribute?
What is control limit and specification limit?
What is meant by stability and capability?
What is CP, CpK, Range, etc?

SPC & TS 16949

Corporate
Quality

TS 16949:2009 and SPC requirements

as per TS

Corporate
Quality

Process as per TS 16949

With Who?
(Competent
personnel)

With What?
(Materials,
Infrastructure,
Work environment)

Customer who
has a need

How?
(Procedures,
Methods,
Instructions)

Process

Customer who
has a need met

With Which?
(Key process and
Product measures)

Goal of TS 16949

Corporate
Quality

The goal of this Technical Specification is the development of

fundamental quality management system that provides for
continual improvement, emphasizing defect prevention
and the reduction of variation and waste in the supply chain.

Defect Vs Defective

Corporate
Quality

A defect is any non-conformance of the unit of product with the

specified requirements

A defective is a unit of product which contains one or more defects.

Detection Vs Prevention

Corporate
Quality

Detection Approach of Quality :

Production

Output

Usable
output

Quality
Control

Customers

Not usable
output

Scrap

Rework

Quality Control

Detection
Tolerates
waste
Scrap + rework = Waste of money, time, materials and efforts (Loss to company)

Corporate
Quality

Detection Vs Prevention
Prevention Approach of Quality :
Prevention
Avoids
waste

Statistical
process Control
Voice of
Process
Input

Production

Output

Customers

Voice of Customer

Do things right the First time & Every time Zero defect concept

7 + 1 Wastes

Corporate
Quality

Can be remembered as T I M W O O D S
10

8 QMS Principles

Corporate
Quality

11

TS 16949 clauses referring to SPC

Corporate
Quality

7.4.3.1 Incoming product conformity to requirements

The organization shall have a process to assure the quality of purchased
product (see 7.4.3) utilizing one or more of the following methods:..
receipt of, and evaluation of, statistical data

7.5.1.1 Control plan

The control plan shall. initiate the specified reaction plan (see 8.2.3.1)
when the process becomes unstable or not statistically capable
7.5.1.3 Verification of job set-ups
Job set-ups shall be verified whenever performed, such as an initial run of a job,
material changeover or job change. . The organization shall use statistical
methods of verification, where applicable

12

TS 16949 clauses referring to SPC

Corporate
Quality

7.6.1 Measurement system analysis

Statistical studies shall be conducted to analyze the variation present in the
results of each type of measuring and test equipment system. This requirement
shall apply to measurement systems referenced in the CP.

8.1.1 Identification of statistical tools

Appropriate statistical tools for each process shall be determined during advance
quality planning and included in the control plan.
8.1.2 Knowledge of basic statistical concepts
Basic statistical concepts, such as variation, control (stability), process capability
and over-adjustment shall be understood and utilized throughout the organization

13

TS 16949 clauses referring to SPC

Corporate
Quality

Clause No: 8.2.3.1

The organization shall initiate a reaction plan from the control plan for
characteristics that are either not statistically capable or are unstable.
These reaction plans shall include containment of product and 100 % inspection,
as appropriate. A corrective action plan shall then be completed by the organization,
indicating specific timing and assigned responsibilities to assure that the
process becomes stable and capable. The plans shall be reviewed with
and approved by the customer when so required.

14

ISO 13485 clause referring to SPC

Corporate
Quality

Clause No: 8.1

The organization shall plan and implement the monitoring, measurement,
analysis and improvement processes needed
a) to demonstrate conformity of the product,
b) to ensure conformity of the quality management system, and
c) to maintain the effectiveness of the quality management system.
This shall include determination of applicable methods, including statistical techniques,
and the extent of their use.
NOTE National or regional regulations might require documented procedures for
implementation and control of the application of statistical techniques.

15

AS 9100 clause referring to SPC

Corporate
Quality

Clause No: 7.4.3

Verification of Purchased Product: The organization shall establish and implement the
inspection or other activities necessary for ensuring that purchased product meets
specified purchase requirements.
Note 1 Customer verification activities performed at any level of the supply chain
Should not be used by the organization or the supplier as evidence of effective control
of Quality and does not absolve the organisation of its responsibility to provide
acceptable product and comply with all requirements.
Note 2 Verification activities may include
- obtaining objective evidence of the quality of the product from suppliers
(e.g., accompanying documentation, certificate of conformity, test records,
statistical records, process control records),

16

AS 9100 clause referring to SPC

Corporate
Quality

Clause No: 8.1

The organization shall plan and implement the monitoring, measurement, analysis and
improvement processes needed
a) to demonstrate conformity to product requirements,
b) to ensure conformity of the quality management system, and
c) to continually improve the effectiveness of the quality management system.
This shall include determination of applicable methods, including statistical techniques,
and the extent of their use.
NOTE: According to the nature of the product and depending on the specified
requirements, statistical techniques can be used to support:
- design verification (e.g., reliability, maintainability, safety);
- process control:
- selection and inspection of key characteristics;
- process capability measurements;
- statistical process control;
- design of experiment;
- Inspection and
- failure mode, effect and criticality analysis.

17

QMS and Importance of Data

Corporate
Quality

Management system (QMS) provides a framework for recording data and a

set of tools may be applied to interpret fully and derive maximum use of the data.

In statistical process control numbers and information will form the basis for
decisions and actions, and a thorough data recording system is essential.

Simple methods will offer any organization a means of collecting, presenting

and analyzing most of its data

18

Corporate
Quality

7 QC Tools overview

19

7 QC Tools overview

Corporate
Quality

Check sheets / tally charts how often is it done?

Histograms what does the variation look like?
Pareto analysis which are the big problems?
Cause and effect analysis - causes the problems?
Stratification material, operator, machine time/season, supplier,
customer, department wise, etc (Defect Concentration Diagram )
Scatter diagrams what are the relationships between factors?
(Generally input and output variables (X and Y)
Control charts which variations to control and how?

20

7 QC Tools application

Problem Identification

Check Sheet

Corporate
Quality

Problem Analysis

Pareto diagram

Histogram

Stratification

Scatter Diagram

Cause & Effect

diagram

Control Chart

21

Contents of this Program

Corporate
Quality

a) Introduction and Basics of SPC

b) Process Capability and Stability
c) Control Charts for Variables
d) X bar R Chart in detail
e) Control Chart Patterns and Analysis

22

SPC

Corporate
Quality

Statistical Process Control :

Statistical process control (SPC) is a process of monitoring and measuring
Variability in the performance of a process or a machine through the
Interpretation of statistical techniques and it is employed to manage
in-process quality
Statistical process control (SPC) is an element of Quality assurance in production
And a tool for continuous Improvement

23

SPC

Corporate
Quality

Activity for variability

24

Corporate
Quality

Variability
Random

Non-Random

1. Common causes

1. Special causes

2. Inherent in a process

2. Due to identifiable factors

3. Natural cause

3. Can be modified through operator

4. Can be eliminated only through

or Management action

Improvements in the system

25

Natural Variations

Corporate
Quality

These are to be expected

Output measures follow a probability distribution
For any distribution there is a measure of central tendency and dispersion
Variations that can be traced to a specific reason (machine wear, misadjusted
equipment, fatigued or untrained workers)

26

Variability

Corporate
Quality

Variability is inherent in the process due to

Chance Causes
Assignable causes
Provides a statistical signal when assignable causes are present

Objective is
To detect and eliminate assignable causes of variation

27

Measurement of Variation

Corporate
Quality

How to measure the variation :

By Measuring the Population (or)
By Measuring the Samples

Population is all the data that exist for a process

Sample is a small group of data randomly taken
from a population that is representative of the
whole population

28

Samples

Corporate
Quality

After adequate samples are taken from a stable process,

they form a pattern called a distribution

Frequency

Solid line represents

Distribution

Weight
To measure the process, take samples and analyse the sample statistics using appropriate chart

29

Type of distributions

Corporate
Quality

There are three common patterns of variation

Normal A bell shaped curve

Bi-Model has two or more peaks

Skewed Looks like a normal

curve pushed over to one side

30

Measures of variation

Corporate
Quality

Central tendency / Accuracy

Mean :
The arithmetic mean is the average of all the values of the variate in the sample
Median :
When all the observations (n) are arranged in ascending or descending order,
then the median is the magnitude of the middle case, it has half the observation
Above it and half below it.
Median = (n+1/2) if n is odd, ((n/2)+1) if n is even.
Mode :
Mode is the value that occurs most frequently, in a frequency histogram
or frequency polygon, it is the observed value corresponding to the high point
of the graph.
E.g : 2,3,2,4,5,2,7,6,2,4,2 Since 2 occurs more frequently mode is 2.

31

Measures of variation

Corporate
Quality

Spread / Dispersion :
Range :
Range is the simplest measure of dispersion in a sample. It is used particularly
In the control chart. It is the difference between the largest observed value and
Smallest observed value.
Standard Deviation (
) :
Standard deviation is defined as the root mean square of the differences between
the observations and the mean.
X = ((X1+X2+X3++Xn)/n), = (X
i X)2/n
Variance :
Sum of the squares of the deviations from the arithmetic mean divided by the
Number of observations n. In other words, the variance is the square of the standard
Deviation (variance = 2 )

32

Measures of variation

Corporate
Quality

Central tendency / Accuracy

Spread / Precision

33

Corporate
Quality

If you say there is good process control,

what is important then?

34

Stability & Predictability

Corporate
Quality

If only Natural causes of variation are present, the output of a process forms a
distribution that is stable over time and is predictable.

Process capability is determined by variations that come from common causes.

This represents the best process.
35

Stability & Predictability

Corporate
Quality

If assignable causes are present, the process output is not stable over time
and is not predictable.

36

Prediction

Corporate
Quality

Who will be the GOOD student?

Why?
37

Corporate
Quality

Types of Data
Variable

Attribute

1. Characteristics that can take any real value

1. Defect related characteristics

2. May be in whole or in fractional numbers

2. Classify products as either good or bad

3. Continuous random variables

(or) count of defects

3. Categorical or discrete random variables

Note :
Variable data could be measured on a continuous scale while attribute data would be
a number
Variable data : Length of the box, Weight of the material, etc..
Attribute data : Population, number of defectives, defects, no. of customer complaints,
Number of items passed and failed.

38

Need of a Control chart

Corporate
Quality

Purpose
Distinguish between natural variations and variations due to assignable causes
through the distribution curve
To monitor, control, and improve process performance over time by studying variation
and its source.

39

What Control charts do?

Corporate
Quality

Focuses attention on detecting and monitoring process variation over time

Distinguishes special from common causes of variation, as a guide to local or
management action
Serves as a tool for ongoing control of a process
Helps to improve a process to perform consistently and predictably for higher quality,
lower cost, and higher effective capacity
Provides a common language for discussing process performance

40

What Control charts do?

Corporate
Quality

Give a statistical signal that a process has gone out of control

so the operator can adjust to bring into control.
The chart itself cannot control the process. It provides the signal
so the operator can take action in the process.
A process that is stable is in control. If the process changes, we get
a signal from the control chart that it has gone out of control.

Note :
Control charts can be used to measure product characteristics or in-process parameters.
Measuring an in-process parameter is the best way to monitor a process because it may
provide a signal that the process is going out of control before the bad product is made.

41

Control limit & Specification limit

Corporate
Quality

Control Limits :
For plotting the control charts, generally 3 sigma limits are selected
and they are termed as Control limits.
Specification Limits :
Specification data are provided in the blue print by customer.
For e.g., Shaft diameter, 35.50 +/- 0.25 mm.
Upper Specification limit (USL) 35.75 mm
Lower Specification limit (LSL) 35.25 mm

42

Corporate
Quality

Types of Control charts

Data

Attribute Data
(Number or Classification)

Variable Data
(Measurement)

Number

Classification

Defects per unit

Defective parts

Const.
sample
size

Variable
sample
size

Const.
sample
size

Variable
sample
size

Sample
size = 1

Const.
sample
size

Variable
sample
size

C chart

U chart

NP
chart

P chart

Xi chart
(Indi chart)

X,R
chart

X,S
chart

Poisson distribution

Binomial distribution

Normal distribution
43

Control charts for variables

Corporate
Quality

For variables that have continuous dimensions,

Weight, Speed, Length, Strength, etc..
X charts are to control the central tendency
R charts are to control the dispersion

Why range chart required ?

44

Mean (X bar) and Range (R) charts

These sampling
Distributions
Results in the
Charts below

Corporate
Quality

Sampling mean
Is shifting upward
But range is
consistent

X chart

(X bar chart
Detects shift
In central
Tendency)

R Chart

(R chart does not

Detect change in
mean)
45

Mean (X bar) and Range (R) charts

These sampling
distributions
results in the
charts below

Corporate
Quality

Sampling mean
is constant but
dispersion is
increasing

X chart

(X bar chart
does not detect
the increase in
dispersion)

R Chart

(R chart detects
increase in
dispersion)
46

Sampling for Variable and Attribute data

Corporate
Quality

Attribute charts requires larger sample sizes

50 100 parts in a sample
Variable charts require smaller samples
2 to 10 parts in a sample
Note :
For process control most widely used charts are X bar and R charts
For inspection and testing applications, Attribute charts are used

47

Steps to construct a Control chart

Corporate
Quality

1. Determine the data to be collected

2. Collect and enter the data by subgroup
3. Calculate and enter the average for each subgroup.
4. Calculate and enter the range for each subgroup
5. Calculate the grand mean of the subgroups average
6. Calculate the average of the subgroup ranges
7. Calculate the upper control limit (UCL) and lower control limit (LCL)
for the averages of the subgroups
8. Calculate the upper control limit for the ranges. When the subgroup
or sample size (n) is less than 7, there is no lower control limit
9. Select the scales and plot the control limits, centreline, and data
points, in each plotting area.
10. Provide the appropriate documentation
48

Setting control limits for X bar chart

Corporate
Quality

For control charts when we dont know :

Upper Control limit (UCL) = X + A2R
Lower control Limit (LCL) = X - A2R

Where,
R = Average range of the samples
A2 = Control factor found in the table
X = Mean of the Sample means

49

Control chart factors

Corporate
Quality

50

R Chart

Corporate
Quality

Type of variable control chart

Shows sample ranges over time
Difference between smallest and largest values in sample
Monitors process variability
Independent from process mean

51

Setting control limits for R chart

Corporate
Quality

For control charts when we dont know :

Upper Control limit (UCLR) = D4R
Lower control Limit (LCLR) = D3R

Where,
R = Average range of the samples
D3 and D4 = Control factor found in the table

52

Setting control limits for R chart

Corporate
Quality

For control charts when we know :

Upper Control limit (UCL) = X + Zx
Lower control Limit (LCL) = X Zx
Where,
x = mean of the sample means or a target value set for the process
z = number of normal standard deviations, generally it is 3.
x = standard deviation of the sample means
= / n
= population standard deviation
n = sample size
53

Setting control limits thro

Corporate
Quality

For X bar chart :

Upper Control limit (UCL) = X + A1
Lower control Limit (LCL) = X A1
= /,
N = Number of sub groups, using the C2 factor from Table B to estimate
= /C2
And 3x = 3/ n or
3x = 3/ C2 n

A1 = 3/ C2 n

For R chart :
Upper Control limit (UCLR) = D2
Lower control Limit (LCLR) = D1
54

Setting control limits thro

Corporate
Quality

For R chart :
Upper Control limit (UCL) = X + A1
Lower control Limit (LCL) = X A1
' = /,
N = Number of sub groups, using the C2 factor from Table B to estimate
= /C2
And 3x = 3/ n or
3x = 3/ C2 n

A1 = 3/ C2 n

55

Plotting X bar and R chart

Corporate
Quality

X bar chart :
Central line X

Solid Horizontal line

Upper & Lower control limits Dotted horizontal line

R chart :
Central line X

Solid Horizontal line

Upper & Lower control limits Dotted horizontal line

Note :
If the sub group size is six or less, the lower control limit for R is Zero.
If the sub group size is seven or more, the lower control limit for R should be drawn
in Dotted horizontal line

56

Interpretation of Control charts

Corporate
Quality

In a control chart, the 3 standard deviations are identified by zones.

Each zones dividing line is exactly one-third the distance from the
centreline to either the UCL or LCL.
Zone A is defined as the area between 2 and 3 standard deviations
from the centreline on both the plus and minus sides of centreline.
Zone B is defined as the area between 1 and 2 standard deviations
from the centreline on both sides of centreline.
Zone C is defined as the area between centre line and 1 standard deviation
from the centreline on both sides of centreline.

57

Control charts patterns

Corporate
Quality

Control chart
patterns

Chance (Natural) pattern

of variation

Assignable cause (unnatural)

pattern of variation

58

Chance pattern of variation

Corporate
Quality

A control chart having a chance pattern of variation will have the

following three characteristics :
Most of the points will lie near the central line
Very few points will be near the control limits
None of the points (except 3 in 1000) fall outside the control limits.

59

Assignable cause pattern of variation

Corporate
Quality

Rule 1 :
Whenever a single point falls outside the 3 sigma control limits, a lack of
control is indicated. Since the probability of this happening is rather small,
it is very likely not due to chance.

60

Assignable cause pattern of variation

Corporate
Quality

Rule 2 :
Whenever at least 2 out of 3 successive values fall on the same side of the centreline
and more than 2 sigma units away from the centreline (in Zone A or beyond), a lack of
control is indicated. Note that the third point can be on either side of the centreline.

61

Assignable cause pattern of variation

Corporate
Quality

Rule 3 :
Whenever at least 4 out of 5 successive values fall on the same side of the centreline
and more than one sigma unit away from the centreline (in Zones A or B or beyond),
a lack of control is indicated. Note that the fifth point can be on either side of
the centreline

62

Assignable cause pattern of variation

Corporate
Quality

Rule 4 :
Whenever at least 8 successive values fall on the same side of the centerline,
a lack of control is indicated

63

Assignable cause pattern of variation

Corporate
Quality

Other Rules :
Evaluate the graph to see if the process is out-of-control.
The graph is out-of-control if any of the following are true:
Any point falls beyond the red zone (above or below the 3-sigma line).
8 consecutive points fall on one side of the centerline.
2 of 3 consecutive points fall within zone A.
4 of 5 consecutive points fall within zone A and/or zone B.
15 consecutive points are within Zone C.
8 consecutive points not in zone C.

64

Summary of control chart for X and R chart

Corporate
Quality

1. Collect and Calculate Subgroup Data

Collect (at least) 20 subgroups of data from the process
2. Calculate the Centrelines and Control Limits
Apply the formula for Control Chart (control chart factors in table).
Set the scales for the charts
Add the centrelines and control limits
3. Plot the Data
Plot on both the X-bar and the R Charts
4. Interpret the Control Chart As per Rules
5. Take ActionUse the calculated limits if the process is stable.
Improve the process if it is not stable
You cannot use a control chart on an unstable process
65

Revising control limits for X and R chart

Corporate
Quality

1. Initial limits are called Trial control limits

2. Take periodic samples and determine the process is in control
3. Trail control limits can be used to control the process for sizable volume
4. After gathering sufficient data, control limits may be revised time to time.
a) Once in a week
b) Once in every month
c) Once in every 25, 50 or 100 subgroups

66

Run chart

Corporate
Quality

Run chart / Trend Chart :

Recording and depiction of data in the sequence in which it occurs.
It can be used for analysis of variance.

Procedure :
Determine the amount of data to be obtained.
Estimate the range of values for scaling purposes
Plot values in the sequence in which they occur.
View and interpret
If necessary conduct data evaluation

67

Corporate
Quality

Run chart

mm
9.7
9.65

USL

9.6
9.55
9.5
9.45

LSL

9.4
9.35
9.3

Job No.

68

Corporate
Quality

Process Capability study

69

Process capability

Corporate
Quality

Process capability is the natural variation in a process that

results from common causes.
Cp = (USL LSL) / 6
Process capability compares the process output with the
customers specification
Process capability determines statistically if the process can
meet the customers specification

70

Process capability

Corporate
Quality

R
-----d2

71

Process capability

Corporate
Quality

When Cp = 1, the natural variation is the same as the design

specification width (b).
When Cp < 1, a significant percentage of output will not conform
to the specifications (a)
Cp > 1, indicates good capability (c); in fact, many firms require
Cp values of 1.66 or greater from their suppliers
The value of Cp does not depend on the mean of the process;
thus, a process may be off-center, (d), and still show an acceptable value of Cp
A capable process is Stable and not changing
A capable process can fit within the customers specification with a
little extra room (usually 25%) to spare.

72

Process capability

Corporate
Quality

73

Process capability index

Corporate
Quality

Process capability Index is the indication of shift of process from the mean.

Cpk = Minimum of [(USL-X)/3sigma, (X-LSL)/3sigma]

Computing Cpk :
Net weight Specification 9.0g +/- 0.5g
Process Mean 8.80 g
Process standard deviation 0.12
Cpk = Minimum of [(USL-X)/3sigma, (X-LSL)/3sigma]
= [(8.80 8.50/3(0.12)), (9.50 8.80/3(0.12))]
74

Process capability index

Corporate
Quality

Interpretation of Cp / Cpk :

Acceptable values for process capability indices

75

Corporate
Quality

Flow chart - Process capability

Verify measurement
system

Implement control
chart

Is it under
control

Establish goals for

capability

No

Implement corrective action

Is process
average =
target

No

Can process
average be moved
to target
Yes

Yes
Monitor controls
and improve

Yes

No

Yes
Is capability
meeting the
goal

Select key
characteristics

No

Implement
corrective action

Adjust

A
76

Uses of Process capability

Corporate
Quality

Provide information to facilitate design of product characteristics / tolerances

Assist process planners to select or modify a process
Assist in establishing the interval between sampling for process control
Specify the performance requirement of a new process
Compare the capability of various processes
Selection of operators
Assigning work to machines
Selecting between competing vendors
Determining the economic nominal for an operation
Reducing the variability in a process

77

What supervisor to do?

Corporate
Quality

Prepare SPC instruction chart

To prepare blank chart
Fix the scales in the graph
Calculate UCL, LCL, Limit marks (UM,LM) for both X bar and R chart
Calculate Cp & Cpk (Process capability Index)
Support operator in interpretation of charts and remedial actions
Analysis of Cp & Cpk for different machines and fix actions for improvement
of processes
Maintain records and charts of SPC
Train operators on the job
Audit and plot graph in different colour with date & signature
78

What operator to do?

Corporate
Quality

Follow the instruction chart

Do a setup approval of the machine
Measure the dimension for the characteristic and enter the measured values
Enter the X bar and Range values
Plot the graph for both X bar and Range chart
Write the date, time and signature
Interpret the charts and act accordingly for actions on process. Take
support from foreman.

79

Benefits of SPC

Corporate
Quality

Economical Quality Assurance

Reduced scrap and repair
High product quality
Keeping up the schedule of production
Improves productivity
Build better experience with process
Provides a live record for feature investigation
Defects are avoided online at source
Prediction of process control
Satisfies the requirements of standard
Satisfied customer

80

Corporate
Quality

only
can

Make it Happen, Let us...

81

Corporate
Quality

82

Corporate
Quality

83