Statistical Process Control

An awareness session
What we will cover

• What is Statistics ?
• What is SPC ?
• Importance of SPC
• Benefits of SPC ?
• SPC tools.
• How to apply SPC in our organizational context.
Discussion

• What do you measure in your process?

• Why do those measures matter?
• Are those measures consistently the same?
• Why not?
Have you ever…
• Shot a rifle?
• Played darts?
• Shot a round of golf?
• Played basketball?

What is the point of these sports? What makes them hard?

Who is better ?

• See the target hits of two players

Emmet & Jake.

Emmett

Jake

Who is the better shot?

 Concepts to understand before SPC session
• Statistics -a branch of mathematics dealing with the collection, analysis,
interpretation, and presentation of masses of numerical data”

• Population - the entire set of raw data that you

may have available for a test or experiment

• Sample – Statistics allows to take some values

that represent a population , to analyze them as to
make meaningful conclusions and decisions.
This is called sampling and the number of samples
taken is represented by sample size.
 Descriptive statistics
• Mean is the central value normally called average.

Mean corresponds to a single value which most appropriately describes a group

of data
If x1, x2, x3, … xn is the group of data and mean is denoted by x̄ , n is count of data.
x̄ = (x1+ x2+ …+ xn)/n or
Descriptive statistics
• Median – Middle most value when arranged in ascending or descending order.
• Mode – Most repeated value

Example : Consider the data (4, 2, 4, 3, 2, 2)

Mean = (4+2+4+3+2+2) / 6 = 2.83

Median = 2, 2, 2, 3, 4, 4 , median = (2+3)/2 = 2.5

Mode = 2 (appears 3 times)

Distribution
Example : for 10 employees , 9 of them in the salary bracket 10K, 1 in the salary bracket 110k, the mean will be
20K. But their actual salary is only 10K so there is no true representation of the population with the mean.

A distribution chart (normally a histogram) is used to display frequency by which each value appear in the set to
get a better representation of data.

Frequency - Number of times a value

occurs.

Example : in the example, there are

about 25 persons in age range 35-40.
So the frequency of the age range 35-
40 is 25. similarly, for 55-60y is 140..
Etc.
Normal distribution
Referred to as bell curve due to its shape (Symmetrical)
Most values will be clustered around the central peak and further away values distributed equally on either sides.
Most natural process will form a normal distribution.

Note : Normal processes will

resemble the bell curve , and may not
exactly look like a bell curve.
Concepts of variation & Range
Variation is a measure of how spread out the data is around the center of the data.
• Common measure of variation is by Range & Standard deviation.
If x1, x2, x3, … xn is the group of data, Range R = X max – X min

In the example, youngest Nobel prize winner is 17 years

and oldest is 97 years. So the range is 97-17 = 80 years.

Concept of Standard deviation.
Standard deviation (σ) measures how far a 'typical' observation is from the average
of the data (μ). SD is most used measure of variation and important in statistics.

It is a common measure of variation for data that has a normal distribution.

A low standard deviation indicates that the values tend to lie quite close to the
mean, while a high standard deviation indicates that the values are more spread
out.
Step 1: Find the mean.
Step 2: For each data point, find the square of its distance
to the mean.
Step 3: Sum the values from Step 2.
Step 4: Divide by the number of data points.
Step 5: Take the square root
Variability
8
7
10
8
9
• Deviation = distance between observations
and the mean (or average) Emmett

Observations Deviations
10 10 - 8.4 = 1.6
9 9 – 8.4 = 0.6
8 8 – 8.4 = -0.4 Jake
8 8 – 8.4 = -0.4
7 7 – 8.4 = -1.4
averages 8.4 0.0
Variability

• Deviation = distance between observations

and the mean (or average) Emmett

Observations Deviations 7
7 7 – 6.6 = 0.4 6
7
7 7 – 6.6 = 0.4 7
7 7 – 6.6 = 0.4 6
Jake
6 6 – 6.6 = -0.6
6 6 – 6.6 = -0.6
averages 6.6 0.0
Variability
8
7
10
8
9
• Variance = average distance between
observations and the mean squared Emmett

Observations Deviations Squared Deviations

10 10 - 8.4 = 1.6 2.56
9 9 – 8.4 = 0.6 0.36
8 8 – 8.4 = -0.4 0.16 Jake
8 8 – 8.4 = -0.4 0.16
7 7 – 8.4 = -1.4 1.96
averages 8.4 0.0 1.0 Variance
Variability

• Variance = average distance between

observations and the mean squared Emmett

Observations Deviations Squared Deviations 7

7 6
7
7 7
7 6
Jake
6
6
averages
Variability

• Variance = average distance between

observations and the mean squared Emmett

Observations Deviations Squared Deviations 7

7 7 - 6.6 = 0.4 0.16 6
7
7 7 - 6.6 = 0.4 0.16 7
7 7 - 6.6 = 0.4 0.16 6
Jake
6 6 – 6.6 = -0.6 0.36
6 6 – 6.6 = -0.6 0.36
averages 6.6 0.0 0.24 Variance
Variability

• Standard deviation = square root of

variance Emmett

Variance Standard
Deviation
Emmett 1.0 1.0
Jake 0.24 0.4898979 Jake

But what good is a standard deviation

Variability

The world tends to

be bell-shaped

Even very rare Fewer Most outcomes Fewer Even very rare
outcomes are in the occur in the in the outcomes are
possible “tails” middle “tails” possible
(probability > 0) (lower) (upper) (probability > 0)
 Variability

Here is why:
Even outcomes that are equally likely (like dice), when you add them up,
become bell shaped

Add up the dots on the dice

0.2

0.15
Probability

1 die
0.1 2 dice

0.05 3 dice

0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sum of dots
“Normal” bell shaped curve

Add up about 30 of most things

and you start to be “normal”

Normal distributions are divide up

into 3 standard deviations on
each side of the mean

Once your that, you

know a lot about
what is going on

And that is what a standard deviation

is good for
Usual or unusual?
1. One observation falls outside 3
standard deviations?
2. One observation falls in zone A?
3. 2 out of 3 observations fall in one
zone A?
4. 2 out of 3 observations fall in one
zone B or beyond?
5. 4 out of 5 observations fall in one
zone B or beyond?
6. 8 consecutive points above the
mean, rising, or falling?

X XX
XX1X XX
2X 3 4 5 6 7 8
Causes of Variability

• Common Causes:
– Random variation (usual)
– No pattern
– Inherent in process
– adjusting the process increases its variation
• Special Causes
– Non-random variation (unusual)
– May exhibit a pattern
– Assignable, explainable, controllable
– adjusting the process decreases its variation

SPC uses samples to identify that special causes have occurred

Limits

• Process and Control limits:

– Statistical
– Process limits are used for individual items
– Control limits are used with averages
– Limits = μ ± 3σ
– Define usual (common causes) & unusual (special causes)
• Specification limits:
– Engineered
– Limits = target ± tolerance
– Define acceptable & unacceptable
Process vs. control limits

Distribution of averages

Control limits
Specification limits

Variance of averages < variance of individual items

Distribution of individuals

Process limits
Usual v. Unusual,
Acceptable v. Defective

A B C D E

μ Target
More about limits

Good quality:
defects are
rare (Cpk>1)
μ
target

Poor quality:
defects are
common (Cpk<1)
μ
target

Cpk measures “Process Capability”

If process limits and control limits are at the same location, C pk = 1. Cpk ≥ 2 is exceptional.
Process Capability Indices

Process capability indices are the measures to check

the capability of the process to produce within the
specification.
What is a specification??
e.g.- Length = 10 ± 2 mm

SPEC has TWO factors

10 is the Nominal value
4 is the Tolerance limit

Nominal value is the most desired central value

Tolerance limit is maximum scatter that is allowed
Tolerance v/s Standard deviation

Cp= Tolerance
6 x Standard deviation

This checks whether the variability of the process is within the allowed
tolerance
Nominal v/s Mean

Cpk = minimum value of

USL-Mean , Mean- LSL

3 x St dev 3 x St. dev

This compares the nominal with mean and simultaneously scatter with
Tolerance.
Process capability

Good quality: defects are rare (Cpk>1)

Poor quality: defects are common (Cpk<1)
=
USL – x 24 – 20
3σ = = .667
3(2)
Cpk = min
=
x - LSL 20 – 15
= = .833
3σ 3(2)

14 20 26
15 24
= =
3σ = (UPL – x, or x – LPL)
Going out of control

• When an observation is unusual, what can we conclude?

The mean
has changed

μ1 μ2
Going out of control

• When an observation is unusual, what can we conclude?

The standard deviation

σ1
has changed
σ2

X
Setting up control charts:
Calculating the limits

1. Sample n items (often 4 or 5)

2. Find the mean of the sample (x-bar)x
3. Find the range of the sample R
4. Plot on the chart
5.
x x
Plot the R on an R chart
6. Repeat steps 1-5 thirty times
7. Average the ’s to create (x-bar-bar)
8. x create (R-bar)x
Average the R’s to
R
Setting up control charts:
Calculating the limits

9. Find A2 on table (A2 times R estimates 3σ)

10. Use formula to find limits for x-bar chart:

X A2 R
11. Use formulas to find limits for R chart:

LCL D3 R UCL D4 R

Let’s try a small problem

smpl 1 smpl 2 smpl 3 smpl 4 smpl 5 smpl 6

observation 1 7 11 6 7 10 10
observation 2 7 8 10 8 5 5
observation 3 8 10 12 7 6 8
x-bar
R

X-bar chart R chart

UCL

Centerline

LCL
Let’s try a small problem

smpl 1 smpl 2 smpl 3 smpl 4 smpl 5 smpl 6 Avg.

observation 1 7 11 6 7 10 10
observation 2 7 8 10 8 5 5
observation 3 8 10 12 7 6 8
X-bar 7.3333 9.6667 9.3333 7.3333 7 7.6667 8.0556
R 1 3 6 1 5 5 3.5

X-bar chart R chart

UCL 11.6361 9.0125
Centerline 8.0556 3.5
LCL 4.4751 0
X-bar chart

14.0000
12.0000 11.6361

10.0000
8.0000 8.0556

6.0000
4.4751
4.0000
2.0000
0.0000
1 2 3 4 5 6
R chart

10
9.0125
8
6

4 3.5

0 0

1 2 3 4 5 6
Interpreting charts

• Observations outside control limits indicate the process is probably “out-

of-control”
• Significant patterns in the observations indicate the process is probably
“out-of-control”
• Random causes will on rare occasions indicate the process is probably
“out-of-control” when it actually is not
Interpreting charts

• In the excel spreadsheet, look for these shifts:

A B

C D

Show real time examples of charts here

Lots of other charts exist

P chart C charts U charts Cusum & EWMA

For yes-no For counting Average count Advanced charts
questions like number defects per unit (similar
“is it defective?” where most items to C chart)
(binomial data) have ≥1 defects
(eg. custom built
houses)

“V” shaped or
p (1  p ) u
p 3 c 3 c u 3 Curved control
limits (calculate
n n
them by hiring a
statistician)
Selecting rational samples

• Chosen so that variation within the sample is considered to be from

common causes
• Special causes should only occur between samples
• Special causes to avoid in sampling
– passage of time
– workers
– shifts
– machines
– Locations
Chart advice

• Larger samples are more accurate

• Sample costs money, but so does being out-of-control
• Don’t convert measurement data to “yes/no” binomial data (X’s to P’s)
• Not all out-of control points are bad
• Don’t combine data (or mix product)
• Have out-of-control procedures (what do I do now?)
• Actual production volume matters (Average Run Length)
Introduction

• SPC is a quality control technique that uses statistical methods to monitor

and control a process. It provides a way to detect and correct any variation
or defects in a process to ensure that the product or service meets the
desired quality standards.

• The purpose of SPC is to maintain a process in control, which means the

process is stable and predictable, and produces consistent output that
meets customer requirements.
Benefits of SPC

• Improve quality: By detecting and correcting process variations, SPC can

improve the quality of the product or service.

• Reduce waste: SPC helps to identify and eliminate waste, such as defects,
rework, and scrap, which can save time and money.

• Increase efficiency: SPC can increase efficiency by reducing variability in the

process and improving productivity.

• Customer satisfaction: SPC ensures that the product or service meets

customer requirements, which can lead to increased customer satisfaction.
Key Concepts of SPC
• Control charts: SPC uses control charts to monitor and control a process.
Control charts plot data over time and indicate whether the process is in
control or out of control.

• Variation: SPC focuses on reducing variation in the process, which can lead
to improved quality and productivity.

• Statistical analysis: SPC uses statistical analysis to determine whether the

process is in control or out of control, and to identify the causes of variation.

• Process capability: SPC measures the process capability, which is the ability
of the process to produce output that meets customer requirements.
Steps in Implementing SPC
• Define the process to be controlled.
• Determine the critical characteristics and customer requirements.
• Collect and analyze data to establish a baseline.
• Determine the appropriate control chart and set up the chart.
• Monitor the process using the control chart.
• Take action when the process goes out of control.
• Continuously improve the process.
Control Charts
• Control charts are the primary tool used in SPC.

• They plot data over time, with the mean and control limits calculated based
on the data.

• The control limits are calculated based on the variation in the data, and they
indicate the range of variation that is expected in a stable process.

• Control charts can be used to detect and correct any variation or defects in a
process.
Types of Control Charts
• There are several types of control charts, including:X-bar and R chart: used
to monitor the mean and range of the process.

• Individuals and moving range chart: used when only one observation is
taken at a time.

• p-chart and c-chart: used to monitor the proportion of defective items or

defects per unit.
7. Control Charts

Purpose
• A control chart is nothing but a line chart.
• The primary purpose of a control chart is to predict expected product outcome.
• It can be generated when we have upper and lower control limits present for the data
• we can check whether the control points are lying between the actual upper and lower limits or going out of those.
• If the control points are lying well within limits, then the process is said to be in control

51
Method
Step 1. Find Mean or Average. This is the central control line in our chart.
Mean = (Sum of all values / Number of values) A standard deviation (or σ) is a measure of how
dispersed the data is in relation to the mean. Low
standard deviation means data are clustered around the
mean, and high standard deviation indicates data are
more spread out

Step 2. Find Standard deviation

52
Why we find standard deviation
As per SPC , processes are going normal (process is in control) if 99.73% observations are scattered around the Control Line
within 3 standard deviations above and below the same

Step 4. Find Upper control limit (UCL)

= Mean + 3 * SD

Step 5 . Find Lower control Limit (LCL)

= Mean – 3 * SD

Step 6 . List the data points in time order to Plot the chart

Work file

53
 PROCESS VARIATION..

Can be of TWO types

1) Common cause variation/ Chance cause Variation.

2) Assignable cause variation/ Special cause variation

Common Cause Variation

e.g..
General shift starts at 8:30 AM.
If you plot your time of arrival for various days, it could be 8:20 ,
8:21 ,8:34, 8:40, 8:15, 8:25 etc….
This variation happens by chance is known as common Cause variation /
Chance variation.
Special cause Variation
One day there was a roadblock at Thripunithura and you arrived here at
10:30 AM.

This is a Special cause/ Assignable cause/ Sporadic variation…..

SPC charts help to differentiate Assignable causes from common causes

• Common Cause variation (Approx. 80%)
-- Common cause variation is inherent in all process and cant be avoided.
A process with only common cause variation is highly predictable.

-- A process that has a significant inherent common cause variation will not be capable of producing products that meet predetermined specifications. Significant
changes in process & design is required to reduce the common cause variations.

Eg : lack of clear instructions, measurement errors, normal wear and tear.

• Assignable cause variation (Approx. 20%)

An assignable cause is a source of variation that is intermittent, not predictable.

• Control charts help to detect and thus eliminate assignable-cause variation:

– Get timely data so that you see the effect of an assignable cause soon after it occurs.

– As soon as you see something that indicates that an assignable cause of variation has happened, search for the cause.

Eg: Machine malfunctions, Power cut,

 Understanding process variation

11
Time of Arrival 10
Series1
9
8 1 2 3 4 5 6 7

Series1 8.2 8.15 8.34 10.3 8.4 8.15 8.25

Date of the month

Assignable Cause
TYPES OF CONTROL CHARTS

• X-bar R-bar Charts.

-Control chart for variables
• P-Charts
-Control chart for proportions or %defectives
• nP Charts
-Control chart for number of defectives
• U- charts
- Control chart for no of defects/unit
• C-Chart
-Control Charts for No of defects in a lot
Determining statistical stability of
the Process
• In the absence of any assignable cause , process is considered to be
statistically stable.

• Remember- Cp, Cpk is valid only when the process is statistically stable .
Basic Control Charts interpretation rules:
Following rules can be used to properly interpret the control harts.

Rule 1 – One point beyond the 3 σ control limit

One point beyond the 3 σ control limits, seeks to identify points that are random or
outliers
following are potential special causes to consider:

Man : E.g. New operator, inadequate training , distracted , overcompensating

adjustments.
Machines and Tooling : E.g. Improper start / set up, Tool breakage , machine failure
Method - Process : E.g. Equipment not stabilized , process steps missed or altered.
Method - Inspection : E.g. improper measuring or not calibrated equipment.
Materials : E.g. Change in materials , Components , handling damage or expired
materials
Environment : E.g. Temperature / Humidity
Basic Control Charts interpretation rules:
Rule 2 – Eight or more points on one side of the centerline without crossing

Eight or more points on one side of the centerline without crossing, is a

prominent shift
following are potential special causes to consider:

Man : E.g. New operator, inadequate training , distracted , shift change

Machines and Tooling : E.g. Improper start / set up, Tool breakage , tool wear.
Method - Process : New or incorrect process parameters, improved or
degraded process.
Method - Inspection : E.g. damaged measuring equipment or not properly
calibrated.
Materials : E.g. Change in materials , Components , handling damage or
expired materials
Environment : E.g. Temperature / Humidity
Basic Control Charts interpretation rules:
Rule 3 – Four out of five points in zone B or beyond

Four out of five points in zone B or beyond, is considered a small shift

following are potential special causes to consider:

Man : E.g. New operator, inadequate training , distracted , shift change

Machines and Tooling : E.g. Improper set up or machine, intermitted support
system failure
Method - Process : New or incorrect process parameters, improved or degraded
process.
Method - Inspection : E.g. damaged measuring equipment or not properly
calibrated.
Materials : E.g. mixed materials or components.
Environment : E.g. Temperature / Humidity suddenly too high or low
Basic Control Charts interpretation rules:
Rule 4 – Six points or more in a row steadily increasing or decreasing

Six points or more in a row steadily increasing or decreasing, is considered a

trend.
following are potential special causes to consider:

Man : E.g. distracted operator.

Machines and Tooling : E.g. gradual support system failure (compressed air,
vacuum) , tool wear.
Method - Process : slowly degrading process.
Method - Inspection : E.g. measuring equipment not adequate for intended
use or damaged.
Materials : E.g. variation in materials or components.
Environment : E.g. Temperature / Humidity gradually drifting too high or low
Basic Control Charts interpretation rules:
Rule 5– Two out of three points in zone A

Two out of three points in zone A, is considered a large shift. .

following are potential special causes to consider:

Man : E.g. New operator, inadequate training , distracted , shift change

Machines and Tooling : E.g. Improper equipment or its set up, support system
failure
Method - Process : New or incorrect process parameters, degraded process.
Method - Inspection : E.g. damaged , not adequate or not calibrated measuring
equipment
Materials : E.g. mixed materials or components.
Environment : E.g. Temperature / Humidity suddenly too high or low
Basic Control Charts interpretation rules:
Rule 6 – 14 points in a row alternating up and down

14 points in a row alternating up and down, is generally considered to be over

control.
following are potential special causes to consider:

Man : E.g. New operator, inadequate training , overcompensating adjustments,

not waiting for process to stabilize before making adjustments. (but may be
other assignable causes as well)
Machines and Tooling : E.g. Improper equipment or its set up, support system
failure
Method - Process : Equipment not stabilized, new / missed / incorrect process
parameters
Method - Inspection : E.g. damaged , not adequate or not calibrated measuring
equipment
Materials : E.g. mixed materials or components.
Environment : E.g. Temperature / Humidity intermittently too high or low
 Basic Control Charts interpretation rules:
Rule 7 – Any noticeable/predictable pattern, cycle, or trend

following are potential special causes to consider:

Man : E.g. New operator, inadequate training , interrupted or distracted,

multiple shifts.
Machines and Tooling : E.g. Improper equipment or its set up, intermittent or
gradual support system failure , Tool Wear.
Method - Process : Equipment not stabilized, new / missed / incorrect process
parameters. Two or more process
Method - Inspection : E.g. damaged , not adequate or not calibrated measuring
equipment
Materials : E.g. mixed materials or components. Variation in raw materials.
Environment : E.g. Temperature / Humidity gradually or intermittently drifting
too high or low

The points circled in red are considered out of control.

Stratification
Very stable process within 1sigma frequently.

When stratification is identified, it is generally due to one of two issues.

1. The operators are purposefully truncating the measurements, or
2. The process has improved significantly, which will require the
recalculation of the statistical control limits.
Conclusion
• SPC is a powerful quality control tool that can help organizations improve
the quality of their products or services.

• By monitoring and controlling a process, SPC can reduce waste, increase

efficiency, and improve customer satisfaction.

• By implementing SPC, organizations can achieve better quality and increased

productivity.
70