IES 331 Quality Control

Chapter 5
Control Charts for Variables

Week 4-5
June 28 – July 7, 2005
1
 Dr. Karndee Prichanont IES331 1/2005

Intro to Control Charts for Variables

† Many quality characteristics can be expressed in terms of a
numerical measurement
† Quality characteristic in a continuous scale – variable
† Examples: dimension, weight, volume, density
† It is important to maintain control over both
„ Process mean
„ Process variability

2
 Dr. Karndee Prichanont IES331 1/2005

Control charts for x and R

3
 Dr. Karndee Prichanont IES331 1/2005

Subgroup Data with Unknown µ and σ

4
Dr. Karndee Prichanont IES331 1/2005

5
Dr. Karndee Prichanont IES331 1/2005

6
 Dr. Karndee Prichanont IES331 1/2005

R chart

7
Dr. Karndee Prichanont IES331 1/2005

8
 Dr. Karndee Prichanont IES331 1/2005

Phase I Application of x and R Charts

Equations 5-4 and 5-5 are trial control limits
† Determined from m initial samples
„ Typically 20-25 subgroups of size n between 3 and 5
† Any out-of-control points should be examined for assignable causes
„ If assignable causes are found, discard points from calculations and
revise the trial control limits
„ Continue examination until all points plot in control
„ Adopt resulting trial control limits for use
„ If no assignable cause is found, there are two options
1. Eliminate point as if an assignable cause were found and revise
limits
2. Retain point and consider limits appropriate for control
† If there are many out-of-control points they should be examined for
patterns that may identify underlying process problems 9
 Dr. Karndee Prichanont IES331 1/2005

Example
5-1

10
 Dr. Karndee Prichanont IES331 1/2005

Example
5-1

11
 Dr. Karndee Prichanont IES331 1/2005

Estimating Process Capability

† X-bar and R charts provide info about
capability of the process
„ Probability that the product produced is outside the
specifications

† Example: Hardbake process. Determine the

process standard deviation and probability
(percentage) of the wafers produced will be
outside of the specifications.

12
 Dr. Karndee Prichanont IES331 1/2005

Process capability ratio (PCR Cp)

† Cp implies whether the natural tolerance limits (3-sigma above
and below the mean) in the process are inside the lower and
upper specification limits

† It can also be interpreted as the percentage of the specification

band that process uses up

† Example: Hardbake process

13
 Dr. Karndee Prichanont IES331 1/2005

Process capability ratio (PCR Cp)

14
 Dr. Karndee Prichanont IES331 1/2005

Revision of Control Limits

and Center Lines
† Effective use of control charts requires periodic
review and revision of control limits and center
lines
† Sometimes users replace the center line on the x
chart with a target value
† When R chart is out of control, out-of-control
points are often eliminated to recompute a R
revised value of which is used to determine new
limits and center line on R chart and new limits
on x chart
15
 Dr. Karndee Prichanont IES331 1/2005

Phase II Operation of Charts

† Use of control chart for monitoring future
production, once a set of reliable limits are
established, is called phase II of control chart
usage (Figure 5-4)

† A run chart showing individuals observations in

each sample, called a tolerance chart or tier
diagram (Figure 5-5), may reveal patterns or
unusual observations in the data

16
Dr. Karndee Prichanont IES331 1/2005

17
Dr. Karndee Prichanont IES331 1/2005

18
Dr. Karndee Prichanont IES331 1/2005

19
 Dr. Karndee Prichanont IES331 1/2005

Control vs. Specification Limits

† Control limits are derived
from natural process
variability, or the natural
tolerance limits of a
process

† Specification limits are

determined externally, for
example by customers or
designers

† There is no mathematical or
statistical relationship
between the control limits
and the specification limits
20
 Dr. Karndee Prichanont IES331 1/2005

Rational Subgroups
† x charts monitor between-sample variability
† R charts measure within-sample variability
† Standard deviation estimate of σ used to construct control
limits is calculated from within-sample variability
† It is not correct to estimate σ using

21
 Dr. Karndee Prichanont IES331 1/2005

Guidelines for Control Chart Design

† Control chart design requires specification of sample size,
control limit width, and sampling frequency
„ Exact solution requires detailed information on statistical
characteristics as well as economic factors
„ The problem of choosing sample size and sampling
frequency is one of allocating sampling effort
† For x chart, choose as small a sample size is consistent
with magnitude of process shift one is trying to detect. For
moderate to large shifts, relatively small samples are
effective. For small shifts, larger samples are needed.
† For small samples, R chart is relatively insensitive to
changes in process standard deviation. For larger samples
(n > 10 or 12), s or s2 charts are better choices.

22
 Dr. Karndee Prichanont IES331 1/2005

Charts Based on Standard Values

23
 Dr. Karndee Prichanont IES331 1/2005

Interpretation of x and R charts

24
 Dr. Karndee Prichanont IES331 1/2005

Interpretation of x and R charts

25
 Dr. Karndee Prichanont IES331 1/2005

Effect of Nonnormality on
x and R charts
† An assumption in performance properties is that the underlying
distribution of quality characteristic is normal
„ If underlying distribution is not normal, sampling distributions can be
derived and exact probability limits obtained

† Burr (1967) notes the usual normal theory control limits are very
robust to normality assumption
† Schilling and Nelson (1976) indicate that in most cases, samples of
size 4 or 5 are sufficient to ensure reasonable robustness to normality
assumption
x for chart
† Sampling distribution of R is not symmetric, thus symmetric 3-sigma
limits are an approximation and α-risk is not 0.0027. R chart is more
x
sensitive to departures from normality than chart.
† Assumptions of normality and independence are not a primary
concern in phase I
26
 Dr. Karndee Prichanont IES331 1/2005

The Operating Characteristics Function

† Operating characteristic (OC) curve: The ability of the
x-bar and R charts to detect shifts in process quality
† For Phase II monitoring of a process

27
 Dr. Karndee Prichanont IES331 1/2005

Average Run Length (ARL)

† ARL: The expected number of samples taken before the
shift is detected

28
Dr. Karndee Prichanont IES331 1/2005

29
 Dr. Karndee Prichanont IES331 1/2005

Average Run Length for x and R charts

For the in-control ARL

For the out of control ARL

30
 Dr. Karndee Prichanont IES331 1/2005

The average time to signal (ATS): the number of time

periods that occur until a signal is generated on the control
chart

Expected number of individual units sampled (I)

31
Dr. Karndee Prichanont IES331 1/2005

32
 Dr. Karndee Prichanont IES331 1/2005

Control charts for x and S

† Occasionally, the direct estimation of process standard
deviation is preferred
† This leads to control charts for x-bar and s, especially
when
1. The sample size n is large; n > 10 or 12
2. The sample size n is variable

s chart parameters with a standard value for s given:

33
 Dr. Karndee Prichanont IES331 1/2005

If no standard is given for s, it must be estimated by

analyzing the past data

m samples with each of size n

The average of the m

standard deviations:

s chart parameters with an estimated s:

34
Dr. Karndee Prichanont IES331 1/2005

35
 Dr. Karndee Prichanont IES331 1/2005

Additional Example – Problem 5-5

See also example 5-3 (pg 225-226)
The fill volume of soft-drink beverage bottles is an important quality
characteristic. The volume is measured by placing a gauge over
the crown and comparing the height of the liquid in the neck of the
bottle against a coded scale. On this scale, a reading of zero
corresponds to the correct fill height.
(a) Set up x-bar and s charts and comment
(b) Set up an R chart, and compare with the s chart
Samples 1 2 3 4 5 6 7 8 9 10
1 2.5 0.5 2 -1 1 -1 0.5 1.5 0.5 -1.5
2 0 0 0.5 1 1.5 1 -1 1 1.5 -1
3 1.5 1 1 -1 0 -1.5 -1 -1 1 -1
4 0 0.5 -2 0 -1 1.5 -1.5 0 -2 -1.5
5 0 0 0 -0.5 0.5 1 -0.5 -0.5 0 0
6 1 -0.5 0 0 0 0.5 -1 1 -2 1
7 1 -1 -1 -1 0 1.5 0 1 0 0
8 0 -1.5 -0.5 1.5 0 0 0 -1 0.5 -0.5
9 -2 -1.5 1.5 1.5 0 0 0.5 1 0 1
10 -0.5 3.5 0 -1 -1.5 -1.5 -1 -1 1 0.5
11 0 1.5 0 0 2 -1.5 0.5 -0.5 2 -1
12 0 -2 -0.5 0 -0.5 2 1.5 0 0.5 -1
13 -1 -0.5 -0.5 -1 0 0.5 0.5 -1.5 -1 -1
14 0.5 1 -1 -0.5 -2 -1 -1.5 0 1.5 1.5
15 1 0 1.5 1.5 1 -1 0 1 -2 -1.5 36
 Dr. Karndee Prichanont IES331 1/2005

Control charts for x and S

with variable sample size

37
 Dr. Karndee Prichanont IES331 1/2005

Example
5-4

38
Dr. Karndee Prichanont IES331 1/2005

39
 Dr. Karndee Prichanont IES331 1/2005

The Shewhart Control Chart for

Individual Measurements
† Sample consists of an individual unit (n = 1)
† Examples are:
„ The use of automated inspection and measurement technology
„ Low production rate. Long interval between observations
„ In some chemical processes
„ Multiple measurements are taken on the same unit of product
„ In some continuous process where the deviation is much too small

† Use moving range of 2 successive observations as a basis of

estimating the process variability

Moving Range

40
Dr. Karndee Prichanont IES331 1/2005

Example 5-5

41
Dr. Karndee Prichanont IES331 1/2005

42
 Dr. Karndee Prichanont IES331 1/2005

Additional Example –
Problem 5-55 (pg 259)
Test Viscosity Test Viscosity
The viscosity of a polymer is
measured hourly.
1 2838 11 3174
Measurements for the last 20 2 2785 12 3102
hours are shown as follows:
3 3058 13 2762
4 3064 14 2975
(a) Set up a control chart on
viscosity. Does the process 5 2996 15 2719
exhibit statistical control
6 2882 16 2861
(b) Estimate the process mean and
standard deviation
7 2878 17 2797
8 2920 18 3078
9 3050 19 2964
10 2870 20 2805
43
 Dr. Karndee Prichanont IES331 1/2005

Phase II Operation and

Interpretation of the Charts

44
Dr. Karndee Prichanont IES331 1/2005

45
Dr. Karndee Prichanont IES331 1/2005

46
 Dr. Karndee Prichanont IES331 1/2005

Average Run Lengths

† Crowder (1987b) showed that
ARL0 of combined individuals and
moving-range chart with
conventional 3-sigma limits is
generally much less than ARL0
(= 370) of standard Shewhart
control chart

† Ability of individuals chart to

detect small shifts is very poor
„ Rather than narrowing the 3-
sigma limits, correct approach to
detecting small shifts is a
cumulative-sum or exponentially
weighted moving-average control
chart (Chapter 8) 47
 Dr. Karndee Prichanont IES331 1/2005

Normality
† Borrer, Montgomery, and
Runger (1999) found in-
control ARL is dramatically
affected by nonnormal data

† One approach for

nonnormal data is to
determine control limits for
individuals control chart
based on percentiles of
correct underlying
distribution
„ Requires at least 100 and
preferably 200 observations
48
 Dr. Karndee Prichanont IES331 1/2005

Example 5-6

49
Dr. Karndee Prichanont IES331 1/2005

50
Dr. Karndee Prichanont IES331 1/2005

51