ADC604/AGC604
OPERATIONS &
SERVICE
MANAGEMENT
LESSON 12:
STATISTICAL
PROCESS
CONTROL
Illustrate the elements of good control systems, variation in
processes, the difference between common and special causes of
variation, quality control metrics, and the design of quality control
systems.
Compare variation in manufacturing and service processes, metrics
for quantifying variation, and the role of control charts and statistical
process control methods in helping managers control variation.
LEARNING
Justify and interpret simple control charts for both continuous and OUTCOMES
discrete data, to understand how to select the proper chart, and to
understand the role of SPC in processes approaching Six Sigma
capability.
Explain the concept of process capability and be able to analyze
process capability data, compute process capability indexes, and
interpret the results.
CHAPTER 17: QUALITY
CONTROL AND SPC
COLLIER, D.A. & EVANS J.R. (2023).
OPERATIONS AND SUPPLY CHAIN
MANAGEMENT (3RD ED.). CENGAGE.
Statistical Process
Control (SPC)
CONTENTS Process Capability
Six Sigma Level of
Process
STATISTICAL The objective of a
PROCESS process control
CONTROL system is to provide
a statistical signal
when assignable
causes of variation
are present
STATISTICAL PROCESS CONTROL (SPC)
1. VARIABILITY IS INHERENT IN EVERY PROCESS
- NATURAL OR COMMON CAUSES
- SPECIAL OR ASSIGNABLE CAUSES
2. PROVIDES A STATISTICAL SIGNAL WHEN ASSIGNABLE
CAUSES ARE PRESENT
3. DETECT AND ELIMINATE ASSIGNABLE CAUSES OF
VARIATION
FOUNDATIONS OF STATISTICAL
PROCESS CONTROL
Statistical process control (SPC): a methodology for
monitoring quality of manufacturing and service delivery
processes to help identify and eliminate unwanted causes of
variation.
Common cause variation: result of complex interactions of
variations in materials, tools, machines, information, workers,
and the environment.
Common cause variation accounts for 80 to 95 percent of
observed variation.
Only management has the power to change systems and
infrastructure that cause common cause variation.
Natural cause variation also called common cause
variation
Affect virtually all production processes
Expected amount of variation
UNDERSTANDING
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”
UNDERSTANDING VARIATION
Special cause variation also called assignable cause
variation.
Arises from external sources that are not inherent in the
process, appear sporadically, and disrupt the random pattern of
common causes.
Special cause variation accounts for 15 to 20 percent of
observed variation.
Front line employees and supervisors have the power to
identify and solve specific causes of variation.
UNDERSTANDING VARIATION
STABLE SYSTEM: A SYSTEM IN CONTROL: IF NO SPECIAL OUT OF CONTROL: WHEN
GOVERNED ONLY BY COMMON CAUSES AFFECT THE OUTPUT OF SPECIAL CAUSES ARE PRESENT IN
CAUSES. THE PROCESS. THE PROCESS.
SAMPLES (a) Samples of the product,
say five boxes of cereal
taken off the filling
machine line, vary from
To measure the each other in weight
process, we take
samples and
analyze the
sample statistics
following these
steps
(b) After enough samples
SAMPLES are taken from a
stable process, they
form a pattern called
a distribution
To measure the
process, we take
samples and
analyze the
sample statistics
following these
steps
SAMPLES
To measure the process, we take samples
and analyze the sample statistics following
these steps
(c) There are many types of distributions, including the normal
(bell-shaped) distribution, but distributions do differ in terms of
central tendency (mean), standard deviation or variance, and
shape
SAMPLES
To measure the process, we take samples
and analyze the sample statistics following
these steps
(d) If only natural causes
of variation are
present, the output of
a process forms a
distribution that is Prediction
Frequency
stable over time and is
predictable
Weight
SAMPLES
To measure the process, we take samples
and analyze the sample statistics following
these steps
?
?? ??
(e) If assignable causes are ? ?
present, the process ?
? ?
?
? ?
output is not stable over ??
? ??
?
time and is not
predicable Prediction
Frequency
Weight
CONTROL CHARTS
SPC uses control charts: a run chart to which two horizontal lines, called control limits, are
added; the upper control limit (UCL) and lower control limit (LCL).
Constructed from historical data, the purpose of control charts is to help distinguish
between natural variations and variations due to assignable causes
Control limits are chosen statistically to provide a high probability (generally greater
that 0.99) that points will fall between these limits if the process is in control.
As a problem-solving tool, control charts allow employees to identify quality problems as
they occur. Of course, control charts alone cannot determine the source of the problem.
CONTROL
CHARTS
The general structure of a
control chart is shown as
below:
PROCESS
CONTROL
Variables Attributes
Characteristics Defect-related
that can take characteristics
any real value Classify products
May be in as either good or Types of
whole or in bad or count
fractional defects Data
numbers Categorical or
Continuous discrete random
random variables
variables
Choosing the Right Control Chart
STEPS IN
CREATING
CONTROL
CHARTS
CONTROL
CHARTS FOR
VARIABLES
For variables that have
continuous dimensions: Weight,
speed, length, strength, etc.
x-charts are to control the
central tendency of the process
R-charts are to control the
dispersion of the process
These two charts must be used
together
SETTING CHART LIMITS
SETTING CONTROL LIMITS
SETTING CONTROL LIMITS
SETTING CHART LIMITS
CONTROL
CHART
FACTORS
R – CHART
SETTING
CHART LIMITS
FOR SALMON FILETS AT DARDEN RESTAURANTS
RESTAURANT CONTROL LIMITS
EXERCISE 12.1
The management of West Allis Industries is concerned about the production
of a special metal screw used by several of the company’s largest customers.
The diameter of the screw is critical to the customers. Data from five
samples appear in the accompanying table. The sample size is 4. Is the
process in statistical control?
OBSERVATIONS
Sample 1 2 3 4 R
Numbe
r
1 .5014 .5022 .5009 .5027 .0018 .5018
2 .5021 .5041 .5024 .5020 .0021 .5027
3 .5018 .5026 .5035 .5023 .0017 .5026
4 .5008 .5034 .5024 .5015 .0026 .5020
5 .5041 .5056 .5034 .5047 .0022 .5045
Average .0021 .5027
SOLUTION 12.1
Compute the range for each sample
and the control limits
UCLR = D4 R = 2.282(0.0021) = 0.00479 in.
LCLR = D3 R = 0(0.0021) = 0 in.
SOLUTION 12.1
Process variability is in statistical
Figure 3.9
control.
SOLUTION 12.1
Compute the mean for each sample and
the control limits.
UCLx = X + A2 R = 0.5027 + 0.729(0.0021) = 0.5042 in.
LCLx = X – A2 R = 0.5027 – 0.729(0.0021) = 0.5012 in.
SOLUTION 12.1
Process average is NOT in statistical
control.
CONTROL CHARTS FOR
ATTRIBUTES
For variables that are categorical
Good/bad, yes/no, acceptable/unacceptable
Measurement is typically counting defectives
Charts may measure
Percent defective (p-chart)
Number of defects in a unit (c-chart)
CONTROL LIMITS FOR P-CHARTS
Total defectives
p=
Total number of observations
P-CHART FOR DATA ENTRY
Sample Number Fraction Sample Number Fraction
Number of Errors Defective Number of Errors Defective
1 6 .06 11 6 .06
2 5 .05 12 1 .01
3 0 .00 13 8 .08
4 1 .01 14 7 .07
5 4 .04 15 5 .05
6 2 .02 16 4 .04
7 5 .05 17 11 .11
8 3 .03 18 3 .03
9 3 .03 19 0 .00
10 2 .02 20 4 .04
Total = 80
80 (.04)(1 - .04)
Sample size, n = 100 p= (100)(20)
= .04 σp^ = 100
= .02
P-CHART FOR DATA ENTRY
UCLp = p + zσp^ = .04 + 3(.02) = .10
Possible
LCLp = p - zσp^ = .04 - 3(.02) = 0
assignable
causes present
.11 –
.10 – UCLp = 0.10
.09 –
Fraction defective
.08 –
.07 –
.06 –
.05 –
.04 – p = 0.04
.03 –
.02 –
.01 – LCLp = 0.00
| | | | | | | | | |
.00 –
2 4 6 8 10 12 14 16 18 20
Sample number
CONTROL LIMITS FOR C-CHARTS
C-CHART FOR CAB COMPANY
PATTERNS IN CONTROL CHARTS
UCL UCL
Variations
Variations
Nominal Nominal
LCL LCL
Sample number Sample number
(a) Normal – No action (b) Run – Take action
UCL UCL
Variations
Variations
Nominal Nominal
LCL LCL
Sample number Sample number
(c) Sudden change – Monitor (d) Exceeds control limits – Take action
EXERCISE 12.2
• Hometown Bank is concerned Sample Wrong Account Sample Wrong Account
about the number of wrong Number Numbers Number Numbers
customer account numbers 1 15 7 24
recorded. Each week a random 2 12 8 7
sample of 2,500 deposits is taken
3 19 9 10
and the number of incorrect
account numbers is recorded 4 2 10 17
5 19 11 15
• Using three-sigma control limits, is
the booking process out of 6 4 12 3
statistical control? Total 147
SOLUTION 12.2
Total defectives 147
p= = = 0.0049
Total number of observations 12(2,500)
σp = √𝒑𝒑�(1 – 𝒑𝒑�)/n = √0.0049(1 – 0.0049)/2,500 = 0.0014
UCLp = p + zσp = 0.0049 + 3(0.0014) = 0.0091
LCLp = p – zσp= 0.0049 – 3(0.0014) = 0.0007
Calculate the sample proportion defective and plot each
sample proportion defective on the chart.
SOLUTION 12.2
Fraction Defective
.0091 X
UCL
X X
X
X X
.0049 X Mean
X
X
X X
.0007 X LCL
| | | | | | | | | | | |
1 2 3 4 5 6 7 8 9 10 11 12
Sample
The process is NOT in statistical control.
PROCESS CAPABILITY
PROCESS CAPABILITY
When Cp = 1, the natural variation is the same as the design
tolerance as in Figure 1 (b).
When Cp < 1, a significant percentage of output will not conform to
the specifications as in Figure 1(a).
PROCESS
CAPABILITY When Cp > 1, indicates good capability as in Figure 1(c); in fact, many
firms require Cp values of 1.66 or greater from their suppliers, which
equates to a tolerance range of about 10 standard deviations.
The value of Cp does not depend on the mean of the process; thus, a
process may be off-center such as in Figure 1(d) and still show an
acceptable value of Cp.
http://elsmar.com/Cp_vs_Cpk.html
Cp > 1.33 is desirable
Cp = 1.00 process is barely capable
Cp < 1.00 process is not capable
PROCESS CAPABILITY VERSUS DESIGN
SPECIFICATIONS
PROCESS CAPABILITY INDEX
INTERPRETING
CPK
http://elsmar.com/Cp_vs_Cpk.html
Technical goal of six sigma
Ensure the variation that are present within
product process, a product, or a service to
become so small that the distance of six
SIX SIGMA LEVEL standard deviations to either side of mean
(μ) is still within the acceptance quality
OF PROCESS
range.
The acceptance quality range is
usually represented by upper and
lower tolerance limit (UTL and LTL).
Six sigma quality means the distance from
the mean value of the process to both the
LTL and the UTL is six standard deviations.
SIX SIGMA LEVEL
OF PROCESS The six sigma level of process can be
calculated as,
Process Sigma = UTL - Mean or Mean - LTL
σ σ
EXERCISE 12.3
The intensive care unit lab process has an average
turnaround time of 26.2 minutes and a standard
deviation of 1.35 minutes.
The nominal value for this service is 25 minutes + 5
minutes.
Is the lab process capable of four sigma-level
performance?
SOLUTION 12.3
The intensive care unit
lab process has an
average turnaround time Upper specification =
of 26.2 minutes and a 30 minutes
standard deviation of Lower specification
1.35 minutes. 20 minutes
The nominal value for Average service
this service is 25 26.2 minutes
minutes + 5 minutes. σ=
Is the lab process 1.35 minutes
capable of four sigma-
level performance?
SOLUTION12.3
Cpk = Minimum 𝒙𝒙 �
�– Lower specification , Upper specification – 𝒙𝒙
3σ 3σ
Cpk = Minimum 26.1 – 20 , 30 – 26.1
3 ( 1.35) 3 ( 1.35)
Cpk = 1.506
Cpk = 0.96
Process does not meets 4-sigma level of 1.33
SOLUTION12.3
Upper specification – Lower specification
Cp =
6σ
30 – 20
Cp = = 1.23
6 (1.35)
Process did not meet 4-sigma level of 1.33
Source: https://www.moresteam.com/toolbox/six-sigma-
conversion-table.cfm
THANK YOU
