Statistical Process Control

Mohammed Alamgir
Department of Marketing , University of Chittagong

06/02/2025
 Statistical
Quality
Control
(SQC)

Statistical Process
Descriptive Acceptance
Control
Statistics Sampling
(SPC
)
 Descriptive Statistics
Statistics used to describe quality
characteristics and relationships

Acceptance Sampling
The process of randomly inspecting a
Statistical Process
sample of goods and deciding whether
Control (SPC)
to accept the entire lot based on the
A statistical tool that
involves inspecting a results
random sample of the
output from a process and
deciding whether the
process is producing Process Capability
products with
characteristics that fall The ability of a production
within a predetermined process to meet or exceed
range preset specifications

All three of these statistical quality control categories are helpful in measuring and
evaluating

the quality of products or services. However, statistical process control (SPC)

tools are used most frequently because they identify quality problems during
Descriptive Statistics
Procedures used to summarize, organize, and make sense of a set of scores
or observations
Typically presented graphically, in tabular form (in tables), or as summary
statistics (single values)
The of Dat
Presentation a
Tables & Graphs
Table Graph
s s
Absolute Frequenc
Frequency y
Distributio Histogra
n ms
Relative Bar & Column
Frequency Charts
Distributio
n
Cumulative Line
Frequency Graphs
Distributio
n
Pie
Charts

Stem-&-
Leaf
Diagram
s
Box-&-
Whisker
Diagram
s
Inferential Statistics
Procedures used that allow researchers to infer or generalize observations made
with samples to the larger population from which they were selected
 Inferentia Analysi
l s

Hypothes Relationship
is among
Testin Variables
g
Why SPC is the Most
Important Tool of the
SQC?
 Measure the value of a quality
characteristic
 Help to identify a change or
variation in
some quality characteristic of the
product
or
process
Some Information about
SPC
 SPC can be applied to any
process.
 There is inherent variation in any process which
can be
measured and “controlled”.
 SPC doesn’t eliminate variation, but it does allow the
user to
track special cause variation.
 “SPC is a statistical method of separating variation
resulting
from special causes from natural variation and to
establish and
maintain consistency in the process, enabling process
improvement.” (Goetsch & Davis, 2003. p. 631)
 Source of Variatio
s n

Common Causes Assignable Causes

Based on random of Variatio of can be
causes n Variatio identified
precisely
&
that cannot be n eliminated
identified,
unavoidable
& due to
slight
differences in
processing
(SPC)
• A methodology for monitoring a process to identify special causes of
variation and signal the need to take corrective action when appropriate
 Variability is inherent in every process
 Natural or common causes
 Special or assignable causes
 Provides a statistical signal when assignable causes are present
 Detect and eliminate assignable causes of variation

• SPC relies on control charts

11
06/02/2025
Common
Causes

Special Causes

12
06/02/2025
Histograms do not
take into account
changes over time.

Control charts
can tell us when
a process
changes

13
06/02/2025
Natural Variations
 Also called common causes
 Affect virtually all production processes
 Expected amount of variation
 Output measures follow a probability distribution
 For any distribution there is a measure of central tendency and
dispersion
 If the distribution of outputs falls within acceptable limits, the
process is said to be “in control”

06/02/2025
Assignable Variations
 Also called special causes of variation
 Generally this is some change in the process
 Variations that can be traced to a specific reason
 The objective is to discover when assignable causes are present
 Eliminate the bad causes
 Incorporate the good causes
Descriptive
Statistics
 Describing certain
characteristics of a
product &
a process
 Measures of Central
Tendency
(mean)
 Measures of Variability
(standard deviation &
range)
 Measures of the
Distribution
of Data
Statistic Process
al
Method – 7Control
Basic Tool
s Quality s

Control Check Pareto

Chart Sheet Chart

Cause-&-
Flow Histogra
Effect
Chart m
Diagra
m

Scatter
Diagram
1. Control Chart
 A graph that shows whether a
sample
of data falls within the common or
normal range of variation
 A control chart has upper and
lower
control limits that separate
common from assignable causes
 A process is out of control when a
plotof variation.
of data reveals that one or more
samples fall outside the control
limits.
Types of Control
Chart
Characteristics measured
by
Control
Chart

Variable Attribute
s s

A product characteristic that can be A product characteristic

measured and has a continuum of that
values has a discrete value and
(e.g.,height, weight, or volume). can be
counted

P&C
Charts
The Four Process States of Control Chart

• Processes fall into one of four states: 1) the ideal,

2) the threshold, 3) the brink of chaos and 4) the
state of chaos (Figure 1).3

• When a process operates in the ideal state, that process is in statistical control
and produces 100 percent conformance. This process has proven stability and
target performance over time. This process is predictable and its output meets
customer expectations.

06/02/2025
The Four Process States of Control Chart

• A process that is in the threshold state is

characterized by being in statistical control but
still producing the occasional nonconformance.
This type of process will produce a constant level
of nonconformances and exhibits low capability.
Although predictable, this process does not
consistently meet customer needs.

• The brink of chaos state reflects a process that is not in statistical control, but
also is not producing defects. In other words, the process is unpredictable, but
the outputs of the process still meet customer requirements. The lack of
defects leads to a false sense of security, however, as such a process can
produce nonconformances at any moment. It is only a matter of time.

06/02/2025
 The Four Process States of Control Chart
• The fourth process state is the state of chaos. Here, the process is not in
statistical control and produces unpredictable levels of nonconformance.

• Every process falls into one of these states at

any given time, but will not remain in that state.
• All processes will migrate toward the state of
chaos.
• Companies typically begin some type of
improvement effort when a process reaches the
state of chaos (although arguably they would be
better served to initiate improvement plans at
the brink of chaos or threshold state).
• Control charts are robust and effective tools to
use as part of the strategy used to detect this
natural process degradation (Figure 2).
06/02/2025
Elements of a Control Chart

There are three main elements of a control chart as shown in Figure 3.

• A control chart begins with a time series graph.
• A central line (X) is added as a visual reference for detecting shifts or trends –
this is also referred to as the process location.
• Upper and lower control limits (UCL and LCL) are computed from available data
and placed equidistant from the central line. This is also referred to as process
dispersion.

06/02/2025
Elements of a Control Chart

• Control limits (CLs) ensure time is not wasted looking for unnecessary trouble –
the goal of any process improvement practitioner should be to only take action
when warranted. Control limits are calculated by:
• Estimating the standard deviation, σ, of the sample data
• Multiplying that number by three
• Adding (3 x σ to the average) for the UCL and subtracting (3 x σ from the average) for the
LCL
• Mathematically, the calculation of control limits looks like:

06/02/2025
Control Chart limits

• Because control limits are calculated from process data, they are independent
of customer expectations or specification limits.

• Control rules take advantage of the normal curve in which 68.26 percent of all
data is within plus or minus one standard deviation from the average, 95.44
percent of all data is within plus or minus two standard deviations from the
average, and 99.73 percent of data will be within plus or minus three standard
deviations from the average.

06/02/2025
Controlled Variation

• Controlled variation is characterized by a stable and consistent pattern of

variation over time, and is associated with common causes. A process
operating with controlled variation has an outcome that is predictable within
the bounds of the control limits.

06/02/2025
Uncontrolled Variation

• Uncontrolled variation is characterized by variation that changes over time and

is associated with special causes. The outcomes of this process are
unpredictable; a customer may be satisfied or unsatisfied given this
unpredictability.

06/02/2025
Control Charts for
Variables
Range (R)

Charts

Contro Chart for
l s Attributes
P-Charts C-Charts
 
 Types of Sampling Plans
Single-Sampling Plan Sequential-Sampling
A decision to accept or Plan
reject a lot based on the A plan in which the consumer
results of one random randomly selects items from the
sample from the lot. lot and inspects them one by
one.

Double-Sampling Plan
A plan in which management
specifies two sample sizes and
two acceptance numbers; if
the quality of the lot is very
good or very bad, the
consumer can make a decision
to accept or reject the lot on
the basis of the first sample,
which is smaller than in the
single-sampling plan.

Sampling by
Attribute Samplin by
g Variable
8-3 Introduction to Control Charts

8-3.1 Basic Principles

• A process that is operating with only chance causes of

variation present is said to be in statistical control.
• A process that is operating in the presence of assignable
causes is said to be out of control.
• The eventual goal of SPC is the elimination of variability in
the process.
8-3 Introduction to Control Charts
8-3.1 Basic Principles
A typical control chart has control limits set at values such that if the
process is in control, nearly all points will lie within the upper control
limit (UCL) and the lower control limit (LCL).
8-3 Introduction to Control Charts
8-3.1 Basic Principles
8-3 Introduction to Control Charts
8-3.1 Basic Principles
8-3 Introduction to Control Charts
8-3.1 Basic Principles
Important uses of the control chart
1. Most processes do not operate in a state of statistical control
2. Consequently, the routine and attentive use of control charts will identify
assignable causes. If these causes can be eliminated from the process,
variability will be reduced and the process will be improved
3. The control chart only detects assignable causes. Management, operator,
and engineering action will be necessary to eliminate the assignable causes.
8-3 Introduction to Control Charts
8-3.1 Basic Principles
Types the control chart
• Variables Control Charts
– These charts are applied to data that follow a continuous
distribution.
• Attributes Control Charts
– These charts are applied to data that follow a discrete
distribution.
8-3 Introduction to Control Charts
8-3.1 Basic Principles
Popularity of control charts

1) Control charts are a proven technique for improving productivity.

2) Control charts are effective in defect prevention.
3) Control charts prevent unnecessary process adjustment.
4) Control charts provide diagnostic information.
5) Control charts provide information about process capability.
8-3 Introduction to Control Charts
8-3.2 Design of a Control Chart

Suppose we have a process that we assume the true process mean

is  = 74 and the process standard deviation is  = 0.01. Samples of
size 5 are taken giving a standard deviation of the sample average,
is

 0.01
x   0.0045
n 5
8-3 Introduction to Control Charts
8-3.2 Design of a Control Chart
• Control limits can be set at 3 standard deviations from the
mean in both directions.
• “3-Sigma Control Limits”
UCL = 74 + 3(0.0045) = 74.0135
CL= 74
LCL = 74 - 3(0.0045) = 73.9865
8-3 Introduction to Control Charts
8-3.2 Design of a Control Chart
 Check sheets
• Also known as Data Collection sheets and Tally charts

• Check sheets are non-statistical and relatively simple.

• They are used to capture data in a manual, reliable, formalized way so
that decisions can be made based on facts.
• As the data is collected, it becomes a graphical representation of itself. Areas for
improvement can then be identified, either directly from the check sheet, or by
feeding the data into one of the other seven basic tools.
 Check sheets
• Many organizations:
They are :”DATA RICH, INFORMATION POOR”
• Check sheet can be a valuable tool in wide applications.
• Purpose: To make it easy to collect data for specific purposes or to
convert into valuable information.
 Function of Check Sheets
According to Ishikawa 1982, check sheets have the following functions:
1. Production Process distribution checks
2. Defective item checks
3. Defective location checks
4. Defective cause checks
5. Check-up confirmation checks
6. Others
 Example of a simple check sheet.
(for car valet operation)
Car type
Car registration
Ford Focus
W357 PHR
Interior vacuumed √

Upholstery cleaned √

Dash board cleaned √

De odorised √

Body washed √

Washed waxed & Polished √

Under car washed √

Wheels washed √

Tyres blacked √

Comments: Front bumper badly scratched on delivery, this can not be covered

Performed / Checked by J Bloggs Date 2 May 2008

E.g; Weekly Summary of Shaft
Dimensional Tolerance Results
 Check sheet
• Figure 15-11 reports how the works being produced relates to the shaft
length specifications.
• Machine setup limits 1.120 – 1.130 inches. Outside range  waste!
• So Figure 15-12 is the check sheet set up to display useful information.
• It produces histogram.
Pareto analysis
• Pareto analysis: Analyses of the frequency of events described on a
Pareto chart that contribute to an outcome. In the quality profession,
outcomes could be rejects, scrap, and other contributors to cost of
quality such as incorrect invoices, purchase orders, missing information,
etc.
 1. Pareto Chart
 Named after Italian economist and sociologist Vilfredo Pareto.
 A simple statistical tool that ranks contributing factors to an outcome
according to either cost or frequency of occurrence.
 This allows for easy prioritization of contributing factors for analysis,
thereby keeping cost of analysis low by focusing on the vital few and
temporarily not analyzing the trivial many.
 Pareto charts are important because they can help an organization decide
where to focus limited resources.
 On a Pareto chart, data are arrayed along an X-axis and a Y-axis.
 1. Pareto Chart
• Pareto charts are important because they can help an organization decide
where to focus limited resources.
• On a Pareto chart, data are arrayed along an X-axis and a Y-axis.
 Example
• In a factory, 20% of problems will produce 80% of defects --> 80% of
defect’s cost will be assigned to only 20% of the total number of defect
types occurring.
• So, 80% of defect costs will spring 20% of total cost element.
 Purpose of Pareto
• Pareto can show you where to apply your resources by “revealing few
from the trivial many”..  (Highlight few most important issues out of
many)
Pareto Chart

Figure -1
 Pareto Chart
• Figure -1 represents customers A,B, C, D, E and All others.
• 75% sales are from 2 customers; A,B
• All others include many more customers but brings insignificant sales
(>5%)
• Which customers should be kept happy?
Pareto Chart
Pareto Chart
• Figure -2 shows sales of particular model of automobile by age group of the
buyers.
• The manufacturer has limited budget in advertising.
• The chart reveals the most logical choice to target to advertise.
• Concentrating on advertising on 26-45 age will result in the best return of
investment. (75%)
• The significant few  26-45 age
• The insignificant many are those under 26 & above 45
 Pareto Chart
• Figure -3
 Pareto Chart
• Figure -3 shows 80% of the cost was related to 5 defect causes.
• All the other (about 30 more) were insignificant.
• The longest bar ($70k) accounted for 40%, if solved, immediate reduction
in rework cost will happen.
• After eliminate the longest bar, the team sorted data again to develop
level 2 Pareto Chart
 Pareto Chart
• Figure 15-4
 Steps in Constructing Pareto Chart
1. Select the subject of the chart
2. Determine what data to be gathered
3. Gather the data related to the quality problem
4. Make a check sheet of the gathered data, record the total numbers in each category.
5. Determine total numbers of nonconformities, calculate percentage each.
6. Select scales of the chart
7. Draw PARETO Chart from largest category to smallest.
8. Analyze the chart
Exercise
 Histograms
• Used to chart frequency of occurrence. (How often does something
happen?)
• Commonly associated with processes: attributes and variables

DATA TYPES EXAMPLES

Attributes Has / has not

Good / bad
Pass / fail
Accept / Reject
Conform / non-conform

Variables Measured values (Dimension, weight, voltage, surface,

etc.)
 Histograms
• Figure 15-14
 Histograms
• Attribute data: Go/no go information.
• Variable data: measurement information.
• Looking at Figure 15-14, we are using attributes data; either they passed or
they failed the screening.
• But, it does not reveal about the process contributing to the adjustment.
• Also, does not tell the robust process. This is why variables data is needed.
Histograms and Statistics
• Example: textbook – page 500
BEAD EXPERIMENT

There are 900 white beads, 100 red beads.

1. The beads mixed thoroughly.

2. 50 beads are drawn at random. – Count how many red beads. – Check mark is entered in
histogram.
3. All the beads are put back into container and mixed again.
4. Repeat Step 1  Step 3

The process does not change, but the output changed!

If these steps are taken over and over, Histogram as in Figure 15-15 will occur
Histograms and Statistics
• Figure 15-15 and Figure 15-16
 Histograms and Statistics
• The flatter and wider the frequency distribution curve, the greater the
process variability.
• Taller and narrower the curve, the less process variability.
• 2 things in process variability;
• Standard deviation, σ
• Mean, μ
 Histograms and Statistics
• Mean is the sum of the observations divided by the number of observations
• Also describes the central location of the data in the chart,

• Standard Deviation describes the spread.

Calculating the Mean, μ

μ=ΣX÷n
X=product of the number of beads in a sample times the number of samples containing the
number of beads.

*See Figure 15-17b, page 503 for further understanding.

Histograms and Statistics

μ =ΣX÷n
=510 ÷ 100
=5.1
Histograms and Statistics
 Histograms and Statistics
• Calculating Standard Deviation, σ

d = The deviation of any unit from the mean

n = the number of units sampled.

 From Figure 15-17c,

 n(100)

σ = 1.49, 2 σ = 2.99, 3 σ = 4.47

• Figure 15-18
Shapes of Histograms