Operations
Management
Statistical Process Control
Supplement 6
S61
Outline
Statistical Process Control (SPC).
Mean charts or XCharts.
Range chart or RCharts.
Control charts for attributes.
Managerial issues and control charts.
Acceptance Sampling.
S62
Statistical Process Control (SPC)
Statistical technique to identify when nonrandom variation is present in a process.
All processes are subject to variability.
Natural causes: Random variations.
Assignable causes: Correctable problems.
Machine wear, unskilled workers, poor materials.
Uses process control charts.
S63
Statistical Process Control Steps
Start
Take Sample
Produce Good
Provide Service
Take Samples
Create
Control Chart
Inspect Sample
Is process
in control?
Yes
No
Stop Process
Find Out Why
S64
Process Control Charts
Plot of Sample Data Over Time
Sample Value
80
Upper control limit
60
40
20
0
Lower control limit
13
17
Time
S65
21
Control Charts
Process is not in control if:
Sample is not between upper and lower control
limits.
A nonrandom pattern is present, even when
between upper and lower control limits.
Based on sample being normally distributed.
S66
Distribution of Sample Means
Mean of sample means x
x
Standard deviation of
x
the sample means
n
3 x 2 x 1 x
x 2 x 3 x
(mean)
95.5% of all x fall within 2 x
99.7% of all x fall within 3 x
S67
Central Limit Theorem
Central Limit Theorem
As sample size
gets
large
enough,
distribution of mean
values becomes
approximately normal
for any population
distribution.
X
X
S68
Control Chart Types
Control
Charts
Continuous
Numerical Data
Categorical or
Discrete Numerical
Data
Variables
Charts
R
Chart
Attributes
Charts
P
Chart
X
Chart
S69
C
Chart
Quality Characteristics
Variables
Attributes
Characteristics that you
measure, e.g., weight,
length.
Characteristics for which
you focus on defects.
Continuous values.
Categorical or discrete
values.
S610
Good or Bad.
# of defects.
X Chart
Shows sample means over time.
Monitors process average.
Example: Weigh samples of coffee.
Collect many samples, each of n bags.
Sample size = n.
Compute mean and range for each sample.
Compute upper and lower control limits (UCL, LCL).
Plot sample means and control limits.
S611
X Chart Control Limits 
Std. Dev. of Process Is Known
UCLx x z x
LCLx x z x
xi
x
n
i 1
n
sample mean
at time i
= known process
standard deviation
S612
X Chart  Example 1
Each sample is 4 measurements.
Process mean is 5 lbs.
Process standard deviation is 0.1 lbs.
Determine 3 control limits.
0.1
UCLx 5 3
5.15
4
0.1
LCLx 5 3
4.85
4
S613
Setting Control Limits
n=9
Hour 1
Hour Mean
Hour Mean
Sample
Weight of
1
16.1
7
15.2
Number
Oat Flakes
2
16.8
8
16.4
1
17
3
15.5
9
16.3
2
13
4
16.5
10
14.8
3
16
5
16.5
11
14.2
4
18
6
16.4
12
17.3
5
17
6
16
For 99.73% control limits, z = 3
7
15
8
17
UCLx = x + zx = 16 + 3(1/3) = 17 ozs
9
16
Mean 16.1
=
1
LCLx = x  zx
2011
Pearson
= 16  3(1/3) = 15 ozs
Setting Control Limits
Control Chart for
sample of 9
boxes
Variation due to
assignable
causes
Out of
control
17 = UCL
Variation due to
natural causes
16 = Mean
15 = LCL
           
1 2 3 4 5 6 7 8 9 10 11 12 Out of
Sample number
2011
Pearson
control
Variation due to
assignable
causes
X Chart Control Limits 
Std. Dev. of Process is Not Known
UCLx x A2 R
LCL x x A2 R
A2 is from Table S6.1
n
Ri
i
1
R
n
xi
i 1
sample range
at time i
sample mean
at time i
S616
Factors for Computing Control
Chart Limits
Sample
Size, n
2
Mean
Upper
Lower
Factor, A 2 Range, D4 Range, D3
1.880
3.268
0
1.023
2.574
0.729
2.282
0.577
2.115
0.483
2.004
0.419
1.924
0.076
0.373
1.864
0.136
0.337
1.816
0.184
10
0.308
1.777
0.223
S617
X Chart  Example 2
Each sample is 4 measurements.
Determine 3 control limits.
sample
1
2
3
4
5
mean
5.02
4.99
4.97
5.03
4.99
x 5.0
range.
.12
.08.
.13.
.18.
.14.
4.96, 5.03, 5.01, 5.08
R 0.13
UCLx 5 0.729 0.13 5.095
LCLx 5 0.729 0.13 4.905
S618
X Chart  Example 2
Sample Mean
5.1
Upper control limit
5.0
Lower control limit
4.9
Time
S619
Example 2 New Samples
sample
6
7
8
values
5.05, 5.00, 4.80, 4.95
5.00, 5.10, 5.10, 5.00
4.80, 5.20, 5.10, 5.00
Sample Mean
5.1
mean
4.95
5.05
5.025
range
0.25
0.10
0.40
Upper control limit
5.0
Lower control limit
4.9
Time
S620
Setting Control Limits
Process average x = 12 ounces
Average range R = .25
Sample size n = 5
2011
Pearson
Setting Control Limits
Process average x = 12 ounces
Average range R = .25
Sample size n = 5
UCLx
= x + A2R
= 12 + (.577)(.25)
= 12 + .144
= 12.144 ounces
From Table
S6.1
2011
Pearson
Setting Control Limits
Process average x = 12 ounces
Average range R = .25
Sample size n = 5
UCLx
LCLx
= x + A2R
= 12 + (.577)(.25)
= 12 + .144
= 12.144 ounces
UCL = 12.144
= x  A2R
= 12  .144
= 11.857 ounces
LCL = 11.857
2011
Pearson
Mean = 12
R Chart
Shows sample ranges over time.
Sample range = largest  smallest value in sample.
Monitors process variability.
Example: Weigh samples of coffee.
Collect many samples, each of n bags.
Sample size = n.
Compute range for each sample & average range.
Compute upper and lower control limits (UCL, LCL).
Plot sample ranges and control limits.
S624
R Chart Control Limits
UCL R D4 R
From Table S6.1
LCL R D3R
sample range at
time i
Ri
i 1
n
S625
R Chart  Example 2
Each sample is 4 measurements.
Determine 3 control limits.
sample
1
2
3
4
5
mean
5.02
4.99
4.97
5.03
4.99
x 5.0
range
.12
.08
.13
.18
.14
4.96, 5.03, 5.01, 5.08
R 0.13
UCLR 2.282 0.13 0.297
LCLR 0 0.13 0
S626
Sample Range
R Chart  Example 2
0.3
Upper control limit
0.2
0.1
Lower control limit
0
Time
S627
Example 2 New Samples
Sample Range
sample
6
7
8
values
5.05, 5.00, 4.80, 4.95
5.00, 5.10, 5.10, 5.00
4.80, 5.20, 5.10, 5.00
0.3
mean
4.95
5.05
5.025
range
0.25
0.10
0.40
Upper control limit
0.2
0.1
Lower control limit
0
Time
S628
Control Chart Steps
Collect 20 to 25 samples of n=4 or n=5 from a
stable process & compute the mean and range.
Compute the overall mean and average range.
Calculate upper and lower control limits.
Collect new samples, and plot the means and
ranges on their respective control charts.
S629
Control Chart Steps  Continued
Investigate points or patterns that indicate the
process is out of control. Assign causes for
the variations.
Collect additional samples and revalidate the
control limits.
S630
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 (pchart)
Number of defects (cchart)
2011
Pearson
Control Limits for pCharts
Population will be a binomial distribution, but
applying the Central Limit Theorem allows us to
assume a normal distribution for the sample
statistics
UCLp = p + zp^
^p =
LCLp = p  zp^
where
p
z
p^
n
=
=
=
=
p(1  p)
n
mean fraction defective in the sample
number of standard deviations
standard deviation of the sampling distribution
sample size
2011
Pearson
pChart for Data Entry
Sample
Number
1
2
3
4
5
6
7
8
9
10
p=
Number
of Errors
Fraction
Defective
Sample
Number
6
5
0
1
4
2
5
3
3
2
.06
.05
.00
.01
.04
.02
.05
.03
.03
.02
11
12
13
14
15
16
17
18
19
20
80
(100)(20) =
.04
^p =
2011
Pearson
Number
of Errors
6
1
8
7
5
4
11
3
0
4
Total = 80
(.04)(1  .04)
100
Fraction
Defective
.06
.01
.08
.07
.05
.04
.11
.03
.00
.04
= .02
pChart for Data Entry
UCLp = p + zp^ = .04 + 3(.02) = .10
Fraction defective
LCLp = p  zp^ = .04  3(.02) = 0
.11
.10
.09
.08
.07
.06
.05
.04
.03
.02
.01
.00
UCLp = 0.10
p = 0.04
10
12
14
16
Sample number
2011
Pearson
 LCLp = 0.00
18 20
pChart for Data Entry
Fraction defective
UCLp = p + zp^ = .04 + 3(.02) = .10
.11
.10
.09
.08
.07
.06
.05
.04
.03
.02
.01
.00
Possible
LCLp = p  zp^ = .04  3(.02) = assignable
0
causes
present
UCLp = 0.10
p = 0.04
10
12
14
16
Sample number
2011
Pearson
 LCLp = 0.00
18 20
Control Limits for cCharts
Population will be a Poisson distribution, but
applying the Central Limit Theorem allows us to
assume a normal distribution for the sample
statistics
UCLc = c + 3 c
where
LCLc = c  3 c
c = mean number defective in the sample
2011
Pearson
cChart for Cab Company
c = 54 complaints/9 days = 6 complaints/day
LCLc = c  3 c
=63 6
=0
UCLc = 13.35
14
Number defective
UCLc = c + 3 c
=6+3 6
= 13.35
12
10
8
6
c= 6
4
2
0 
1

2

3

4

5
Day
2011
Pearson

6

7

8
LCLc = 0

9
Managerial Issues and
Control Charts
Three major management decisions:
Select points in the processes that need
SPC
Determine the appropriate charting
technique
Set clear policies and procedures
2011
Pearson
Which Control Chart to Use
Variables Data
Using an xChart and RChart
1. Observations are variables
2. Collect 20  25 samples of n = 4, or n = 5, or
more, each from a stable process and
compute the mean for the xchart and
range for the Rchart
3. Track samples of n observations each.
Table S6.3
2011
Pearson
Which Control Chart to Use
Attribute Data
Using the pChart
1. Observations are attributes that can be
categorized as good or bad (or passfail,
or functionalbroken), that is, in two
states.
2. We deal with fraction, proportion, or
percent defectives.
3. There are several samples, with many
observations in each. For example, 20
samples of n = 100 observations in each.
Table S6.3
2011
Pearson
Which Control Chart to Use
Attribute Data
Using a cChart
1. Observations are attributes whose defects
per unit of output can be counted.
2. We deal with the number counted, which is
a small part of the possible occurrences.
3. Defects may be: number of blemishes on a
desk; complaints in a day; crimes in a
year; broken seats in a stadium; typos in a
chapter of this text; or flaws in a bolt of
cloth.
Table S6.3
2011
Pearson
Use of Control Charts
S642
Acceptance Sampling
Quality testing for incoming materials or
finished goods.
Procedure:
Take one or more samples at random from a lot
(shipment) of items.
Inspect each of the items in the sample.
Decide whether to reject the whole lot based on
the inspection results.
S643
Acceptance Sampling
Inspecting all items is too expensive.
The larger the sample inspected:
The greater the cost for inspection.
The less likely you are to accept a bad lot or to
reject a good lot.
Key questions:
How many should be inspected in each lot?
How confident are you in the accept/reject
decision?
S644