CHAPTER IV

STATISTICAL PROCESS CONTROL

This chapter covers two topics that are
increasingly important in business
organizations: quality control and continuous
improvement.

 Quality control focuses on the conversion of

inputs into outputs.
 The purpose of quality control is to assure
that processes are performing in an
acceptable manner.

 This is accomplished by monitoring process

output using statistical techniques.

 If the results are acceptable, no further action

is required; unacceptable results call for
corrective action.
 In the best companies, the emphasis is on
designing quality into the process, thereby
greatly reducing the need for inspection or
control efforts.

-As you might expect, different business

organizations are in different stages of
this revolutionary process:
 The least progressive rely heavily on
inspection.
 Many occupy a middle ground that involves
some inspection and a great deal of process
control.

And the most progressive have achieved

an inherent level of quality that is high
enough that they are able to avoid
wholesale inspection activities as well as
process control activities. That is the
ultimate goal.
Figure 4-1 illustrates these phases of quality
assurance.

Figure 4-1 Phases of quality assurance

Inspection Corrective Quality built

before/after action during into the
production production process
process
Acceptance Process Continuous
sampling control improvement

The least The most

progressive progressive

 Quality assurance that relies primarily on

inspection after production is referred to as
acceptance sampling.
 Quality assurance efforts that occur during
production are referred to as process control.

 Continuous improvement is an approach

that views quality improvement as a never
ending quest.
Introduction
 In 1924, W. Shewhart of Bell Telephone
Laboratories introduced statistical control
charts that could be used to monitor
production.
 Then around 1930, H.F. Dodge and H.G.
Romig, also of Bell Labs, introduced tables
for acceptance sampling.
 Nevertheless, statistical quality control
procedures were not widely used until World
War II, when the U.S. government began to
require vendors to use them.
- Just after the war, the American Society
for Quality Control was founded.
- Over the years, the society has promoted
quality control with its publications,
conferences, and training programs.
 During the 1950s and 1960s, W. Edwards
Deming is credited with introducing statistical
quality control methods to Japanese
manufacturers.
 Prior to that time, much of the output by the
Japanese was of inferior quality.
 However, by the late 1970s the Japanese
began to acquire a reputation for high quality
in automobiles, appliances, steel, and
electronics, some of which can be attributed to
quality control efforts.

- Now, Japanese manufacturers are

formidable competitors in the world
marketplace.
- One consequence is that manufacturers in
other industrialized countries have
intensified their efforts to improve the
quality of their products.
 - During the 1970s and 1980s quality
assurance methods spread to service
organizations such as banks, government
agencies, health care, and hotels.

The Control Process

 Quality control is concerned with the quality
of conformance of a process: Does the
output of a process conform to the intent of
design?
 Toward that end, statistical process control is
used to evaluate the output of a process to
determine its acceptability.
- To do this, periodic samples are taken
from the process and compared to a
predetermined standard.
 - If the sample results are not acceptable,
the process is stopped and corrective
action is taken.
- If the sample results are acceptable, the
process is allowed to continue.

4.1 Variations and Control

 Processes that provide goods and services
usually exhibit some variation in their
output.
 This variation can be caused by many
factors, some of which we can control and
others that are inherent in the process.

 Variation that is inherent in the process is

often referred to as chance or random
variation.
 It is also sometimes referred to as common
variation.
 -The variations are created by the
combined influences of countless minor
factors, each one so unimportant that
even if it could be identified and
eliminated, the decrease in process
variability would be negligible.
-In effect, this variability is inherent in
the process.

 Variation that is caused by factors that can

be clearly identified and possibly even
managed is called assignable variation.
Sometimes this is called special variation.

- Unlike natural variation, the main

sources of assignable variation can
usually be identified (assigned to a
specific cause) and eliminated.
-Such factors as tool wear, equipment
that needs adjustment, defective
 materials, and human factors
(carelessness, fatigue, failure to follow
correct procedures, and so on) are
typical sources of assignable variation.

4.2 Control Charts

 A control chart is a graphic presentation of
process data over time. It is used to
distinguish between random variability and
nonrandom variability.

 We know that 95.5% of the sample values

will be within ± 2 standard deviations of the
mean of the distribution. Therefore, we could decide
to draw the line, so to speak, at values that represent ± 2 standard
deviations from the mean, and conclude that any value that was
further away than these limits would be judged as a nonrandom
variation. In effect, then, these limits are control limits.

- Control limits are the dividing lines

between random and nonrandom
 deviations from the mean of the
distribution.

 Control charts have two limits that separate

what will be regarded as random variation and
non random variation.

 Figure 4-2 illustrates the components of a control chart. In effect, each

sample is represented by a single value on the chart. Moreover, each
value is compared to the extremes of the sampling distribution (the
control limits) to judge if it is within the acceptable (random) range.

Figure 4-2 Abnormal

Out of variation due to
Example of a control chart assignable
control
sources
UCL

Normal
Mean
variation due to
chance

LCL
Abnormal
variation due
to assignable
sources
 1 2 3 4 5 6 7 8 9 10 11 12

Sample Number

 There are a number of different types of

control charts to choose from, depending on
the type of data involved.
 These fall into one of two categories: control
charts for variables or control charts for
attributes.
 Attribute data are counted (e.g., the number of
defective parts in a sample, or the number of
complaints); variables data are measured,
usually on a continuous scale (e.g., amount of
time needed to complete a task, length or
width of a part).

Control Charts for Variables

 Mean and range charts are used to monitor
variables. Control charts for means monitor
 the central tendency of a process, and range
charts monitor the dispersion of a process.
 Variables are characteristics that have continuous dimensions. They
have an infinite number of possibilities. Examples are weight, speed,
length or strength.
__
- Control charts for the mean, X , and the range, R, are used to monitor
processes that have continuous dimensions.

 In attribute sampling, we determine whether

something is good or bad, fits or doesn't fitit
is a go/no-go situation.
 In variables sampling, however, we measure
the actual weight, volume, number of inches,
or other variable measurements, and we
develop control charts to determine the
acceptability or rejection of the process based
on those measurements.
 For example, in attribute sampling, we might
decide that if something is over 10 pounds we
 will reject it and under 10 pounds we will
accept it.
 In variable sampling, we measure a sample
and may record weights of 9.8 pounds or 10.2
pounds.
 These values are used to create or modify
control charts and to see whether they fall
within the acceptable limits.

 There are four main issues to address in

creating a control chart: the size of the
samples, number of samples, frequency of
samples, and control limits.

Size of Samples
 For industrial applications in process control
involving the measurement of variables, it is
preferable to keep the sample size small.
 There are two main reasons.
 First, the sample needs to be taken within a
reasonable length of time; otherwise, the
process might change while the samples are
taken.
 Second, the larger the sample, the more it
costs to take.

- Sample sizes of four or five units seem to

be the preferred numbers.

Number of Samples

 Once the chart has been set up, each sample

taken can be compared to the chart and a
decision can be made about whether the
process is acceptable.
 To set up charts, however, prudence and
statistics suggest that 25 or so samples be
taken.
Frequency of Samples
 How often to take a sample is a tradeoff
between the cost of sampling (along with the
cost of the unit if it is destroyed as part of the
test) and the benefit of adjusting the system.
 Usually, it is best to start off with frequent
sampling of a process and taper off as
confidence in the process builds.
 For example, one might start with a sample of
five units every half hour and end up feeling
that one sample per day is adequate.

Control limits
 Standard practice in statistical process control
for variables is to set control limits three
 standard deviations above the mean and three
standard deviations below.

- This means that 99.7 percent of the

sample means are expected to fall within
these control limits (that is, within a 99.7
percent confidence interval).
- Thus, if one sample mean falls outside this
obviously wide band, we have strong
evidence that the process is out of control.

Mean Charts
 A mean control chart, sometimes referred to as (“X-bar”) charts, can be
constructed in one of two ways. If we know, through past data, the
standard deviation of the process population, σ , we can set upper and
lower control limits by these formulas:

Upper control limit (UCL) = x + z σ x

Lower control limit (LCL) = x - z σ x
= mean of the sample means or a target value set for the process
Where X
z= number of normal standard deviations (2 for 95.5% confidence,
3 for 99.73%)

σ x = standard deviation of the sample means = σ √ n

σ = Population (process) standard deviation
n= sample size
Example- The following example shows how to
set control limits for sample means using
standard deviations.
The weights of boxes of Oat Flakes within a
large production lot are sampled each hour. To
set control limits that include 99.73% of the
sample means, samples of nine boxes are
randomly selected and weighed. Here are the
nine boxes chosen for Hour 1:
17oz, 13oz, 16oz, 18oz, 17oz, 16oz, 15oz, 17oz,
16oz.

17+13+16+18+17+16+15+17+16
The average weight in the first sample= 9
= 16.1oz.

Also the population standard deviation (σ) is

known to be 1 ounce. We do not show each of
the boxes sampled in hours 2 through 12, but
here are the rest of the results:
Weight of sample
Hour (Avg.of 9 Boxes)
1______________ 16.1
2______________ 16.8
3______________ 15.5
4______________ 16.5
5______________ 16.5
6______________ 16.4
7______________ 15.2
8______________ 16.4
9_______________ 16.3
10______________ 14.8
11______________ 14.2
12______________ 17.3

The average mean of the 12 samples is calculated to be exactly 16

ounces. We therefore have x =16 ounces, σ =1 ounce, n=9, and z=3. The
control limits are:

1
( ) 1
)
UCL x = x +z σ x =16+3 √9 =16+3( 3 =17 ounces
1
( ) 1
)
LCL x = x -z σ x =16-3 √9 =16-3( 3 =15 ounces
The 12 samples are then plotted on the control chart shown below.
Because the means of recent sample averages fall outside the upper and
 lower control limits of 17 and 15, we can conclude that the process is
becoming erratic and not in control.
Figure 4-3
Control chart for samples of 9 boxes

Out of control
Variation due to
UCL=17 assignable causes

Variation due to
Mean=16 natural causes

LCL=15 ________________
Variation due to
assignable causes

Out of control
| | | | | | | | | | | |
1 2 3 4 5 6 7 8 9 10 11 12
Sample number

 A second approach to construct a mean control chart is to use the

sample range as a measure of process variability.

- Because process standard deviations are either not available or

`difficult to compute, we usually calculate control limits based on the
average range values rather than on standard deviations. The range
is defined as the difference between the largest and smallest items in
one sample. For example, the heaviest box of Oat Flakes in Hour 1 of
the previous example was 18 ounces and the lightest was 13 ounces,
so the range for that hour is 5 ounces. The appropriate formulas for
control limits are:

UCL x = x +A2 R
LCL x = x -A2 R

Where R =average range of the samples

A2=Value found in Table 4-1
x =mean of the sample means
Table 4-1 Factors for Computing Control Chart Limits (3 sigma) for x and
R charts
Number of Factors for R chart
observations in Factor for x chart, Lower control limit, D3 Upper control
subgroup (sample limit, D4
A2
size), n
2 ____________________ 1.88_______________ 0_________________ 3.27
3 ____________________ 1.02_______________ 0_________________ 2. 57
4 ____________________ 0.73_______________ 0_________________ 2.28
5 ____________________ 0.58_______________ 0_________________ 2.11
6 ____________________ 0.48_______________ 0_________________ 2.00
7 ____________________ 0.42_______________ 0.80______________ 1.92
8 ____________________ 0.37_______________ 0.14______________ 1.86
9 ____________________ 0.34_______________ 0.18______________ 1.82
10____________________ 0.31_______________ 0.22______________ 1.78
11____________________ 0.29_______________ 0.26______________ 1.74
12____________________ 0.27_______________ 0.28______________ 1.72
13____________________ 0.25_______________ 0.31______________ 1.69
14____________________ 0.24_______________ 0.33______________ 1.67
15____________________ 0.22_______________ 0.35______________ 1.65
16____________________ 0.21_______________ 0.36______________ 1.64
17____________________ 0.20_______________ 0.38______________ 1.62
18____________________ 0.19_______________ 0.39______________ 1.61
19____________________ 0.19_______________ 0.40______________ 1.60
20____________________ 0.18_______________ 0.41______________ 1.59

Example: Super Cola bottles soft drinks labeled “net weight 16 ounces.” An
overall process average of 16.01 ounces has been found by taking several
batches of samples, in which each sample contained 5 bottles. The average
range of the process is 0.25 ounce. Determine the upper and lower control
limits for averages in this process.
  Looking in table 4-1 for a sample size of 5 in the mean factor A 2
column, we find the number 0.58. Thus, the upper and lower
control chart limits are:

UCL x = x + A2 R
=16.01+ (0.58) (0.25)
=16.15 ounces

LCL x = x - A2 R
= 16.01-(0.58) (0.25)
=15.86 ounces

Range Charts
 In addition to being concerned with the process average, operations
managers are interested in the process dispersion, or range. Even
though the process average is under control, the dispersion of the
process may not be.

 For example, something may have worked itself loose in a piece of

equipment that fills boxes of Oat Flakes. As a result, the average of
the samples may remain the same, but the variation within the
samples could be entirely too large. For this reason, operations
managers use control charts for ranges in order to monitor the
process variability, as well as control charts for ranges is the same
as that for process average control charts.
  Limits are established that contain ±3 standard deviations of the
distribution for the average range R . We can use the following
formulas to set the upper and lower control limits for ranges:

UCL R =D4 R
LCL R =D3 R
Where values for D3 and D4 are obtained from table 4.1

Example: The average range of a process for loading trucks is 5.3 pounds.
If the sample size is 5, determine the upper and lower control
chart limits.

Looking in Table 4-1 for a sample size of 5, we find that

D4=2.11 and D3=0. The range control limits are:

UCLR=D4 R = (2.11) (5.3 pounds) =11.2 Pounds

LCLR=D3 R = (0) (5.3 pounds) =0

Using Mean and Range Charts

 Mean control charts and range control charts provide different

perspectives on a process. Mean charts are sensitive to shifts in the
process mean whereas range charts are sensitive to changes in
process dispersion. Because of this difference in perspective, both
types of charts might be used to monitor the same process. Use of
both charts provides more complete information than either chart
alone.
 Once control charts have been set up, they can serve as a basis for
deciding when to interrupt a process and search for assignable causes
of variation.

 Steps to follow when using control charts: There are five steps that are

generally followed in using mean ( x ) - and range (R)-charts:

1. Collect 20 to 25 samples of n=4 or n=5 each from a sample

process and compute the mean and range of each.

2. Compute the overall means ( X and R ), set appropriate control

limits, usually at the 99.73% level, and calculate the preliminary
upper and lower control limits.

3. Graph the sample means and ranges on their respective control

charts and determine whether they fall outside the acceptable
limits.
4. Investigate points or patterns that indicate the process is out of
control. Try to assign causes for the variation and then resume
the process

5. Collect additional samples and, if necessary, revalidate the

control limits using the new data.

Control Charts for Attributes

Control charts for attributes are used when
the process characteristic is counted rather
than measured. For example, the number of
defective items in a sample would be counted,
whereas the length of each item would be
measured. There are two types of attribute control charts, one for
the percentage of defective items in a sample (a p-chart) and one for the

number of defects (called a c-chart). A p-chart is appropriate

when both the defectives and non-defectives can be counted. For
instance, if glass bottles are inspected for chipping and cracking, both
the good bottles and the defective ones can be counted. However, one
can count the number of accidents that occur during a given period of
time but not the number of accidents that did not occur.

Similarly, one can count the number of scratches on a polished surface,

the number of bacteria present in a water sample, and the number of
crimes committed during the month of August, but one cannot count
the number of nonoccurrence. In such cases, a c-chart is appropriate.

P-Chart
A p-chart is used to monitor the proportion of
defectives generated by a process. The theoretical
base for a P-chart is the binomial distribution, although for large
sample sizes, the normal distribution provides a good approximation to
it. Conceptually, a p-chart is constructed and used in much the same
way as a mean chart.
 The center line on a P-chart is the average fraction defective in the
population, P. The standard deviation of the sampling distribution
when P is known is
 σ p= √
P ( 1− p )
n

Control limits are computed using the formulas

UCLp = p + z σ p
LCLp = p- z σ p

If p is unknown, it can be estimated from samples. That estimate, p ,

replaces p in the preceding formulas, as illustrated in the following
example.

Example Using the following information, construct a control chart that will
describe 95.5 percent of the chance variation in the process when
the process is in control. Each sample contains 100 items.

Sample Number of Sample Number of

defectives defectives
1----------------------------14 11------------------------- 8
2----------------------------10 12-------------------------12
3----------------------------12 13------------------------- 9
4----------------------------13 14-------------------------10
5---------------------------- 9 15-------------------------11
6----------------------------11 16-------------------------10
7----------------------------10 17------------------------- 8
8----------------------------12 18-------------------------12
9----------------------------13 19-------------------------10
10--------------------------10 20-------------------------16
220
_____________________________________________________________________
 Solution
z for 95.5 percent is 2.00
220___
Total number of defectives
20(100)
p =_________________________ = = .11
Total number of observations

σp=
p(1− p )
n √
.11(1−.11)
= 100 √
=.03

Control limits are

UCLp= p +z (σ p) =.11+2.00(.03) =.17

LCLp= p -z (σ p) =.11-2.00(.03) =.05

Plotting the control limits and the sample percentages, we can see that the
process is initially in control, although the last point is close to the upper
limits.

.18
18 UCL= ·17
.16
16 _ 
.14 _
14
.12
12 _
Fraction defective

.10
10 _ 
.08
08 _
.06
06 _
.04 LCL=.05
04 _
.02
02 _
.00
00 | | | | | | | | | |
2 4 6 8 10 12 14 16 18 20
Sample number
 Note: If the computed LCL is negative use zero as the lower limit.

C Chart
When the goal is to control the number of
defects per unit, a c-chart is used. Units might
be automobiles, hotel rooms, typed pages, or
rolls of carpet. The underlying sampling
distribution is the Poisson distribution. Use of the
Poisson distribution assumes that defects occur over some continuous region
and that the probability of more than one defect at any particular spot is
negligible. The mean number of defects per unit is c, and the standard

deviation is √ c . For practical reasons, the normal approximation to the Poisson

is used. The control limits are:

UCLc = c +z √ c

LCLc = c -z √ c
If the process average is unknown, it can be estimated from sample data, using
c = Number of defects/Number of samples.

Example

Rolls of coiled wire are monitored using a c-chart. Eighteen rolls have been
examined, and the number of defects per roll has been recorded in the
following table. Is the process in control? Plot the values on a control chart that
uses Z=2.
 Sample
Number of Sample Number of
defects defects
1----------------------------3 10------------------------- 1
2----------------------------2 11------------------------- 3
3----------------------------4 12------------------------- 4
4----------------------------5 13------------------------- 2
5----------------------------1 14------------------------- 4
6----------------------------2 15------------------------- 2
7----------------------------4 16------------------------- 1
8----------------------------1 17------------------------- 3
9----------------------------2 18------------------------- 1
45

Solution
c =45/18 = 2.5
UCL c = c +2√ c = 2.5+2 √ 2.5 =5.66
LCL = c -2 √ c = 2.5-2 √ 2.5 =-0.660

8_
Defects per unit

6_ UCL=5.66

4_   
  c =2.5
2_ 
   
_ LCL=0     
0 | | | | | | | | |
2 4 6 8 10 12 14 16 18
Sample number

When the computed lower control limit is

negative, the effective lower limit is zero, since
 negative observations cannot occur. (The calculation
sometimes produces a negative lower limit due to the use of the normal
distribution to approximate the Poisson distribution: the normal is
symmetrical whereas the Poisson is not symmetrical when c is close to
zero.)

4.3 ACCEPTANCE SAMPLING

 Acceptance sampling is a form of inspection

that is applied to lots or batches of items
either before or after a process instead of
during the process.
 In the majority of cases, the lots represent
incoming purchased items or final products
awaiting shipment to warehouse or
customers.

 The purpose of acceptance sampling is to

decide if a lot satisfies predetermined
standards.
 Lots that are judged to satisfy standards are
passed or accepted; those that do not are
rejected.
 Rejected lots may be subjected to 100 percent
inspection, or purchased goods may be
returned to the supplier for credit or
replacement (especially if destructive testing
is involved).
 Acceptance sampling procedures are most
useful when one or more of the following
conditions exist:
1. A large number of items must be
processed in a short time.
2. The cost consequences of passing
defectives are low.
3. Destructive testing is required.
 4. Fatigue or boredom caused by inspecting
large numbers of items leads to
inspection errors.
 Acceptance sampling procedures can be applied to both attribute and
variable inspection. However, inspection of attributes is perhaps more
widely used, so for purposes of illustration, the discussion here will
focus exclusively on attribute sampling plans.

SAMPLING PLANS
 Sampling plans specify the lot size, N; the
sample size, n; the number of samples to be
taken; and the acceptance/rejection
criteria.
 There are a variety of sampling plans in use. Some plans call for
selection of a single sample, and others call for two or more
samples, depending on the nature of the plan. Some of the different
kinds of plans are briefly described in the following paragraphs.

Single-Sampling Plan In this plan, one

random sample is drawn from each lot, and
every item in the sample is examined and
classified as either “good” or “defective.” If any
sample contains more than a specified number
of defectives, c, that lot is rejected.

Double-Sampling Plan A double-sampling

plan allows for the opportunity to take a second
sample if the results of the initial sample are
inconclusive.
- For example, if the quality of the initial
sample is high, the lot can be accepted
without need for a second sample.
- If the quality in the initial sample is poor,
sampling can also be terminated and the lot
rejected.
- For results between those two cases, a
second sample is then taken and the items
inspected, after which the lot is either
accepted or rejected on the basis of the
evidence obtained from both samples.
- A double-sampling plan specifies the lot size, the size of the initial
sample, accept/reject criteria for the initial sample (e.g., c1=2, c2=5: if
two or fewer defectives are found, accept the lot; if more than five
defectives are found, reject the lot; and if three, four, or five defectives
are found, take a second sample), the size of the second sample, and a
single acceptance number (e.g., c3 = 6: accept the lot if the total number
of defectives found in both samples is six or fewer; otherwise, reject the
lot).

Multiple-Sampling Plan A multiple-sampling

plan is very similar to a double-sampling plan,
except that more than two samples may be
required. A sampling plan will specify each
sample size and two limits for each sample.
- The values increase with the number of samples. If, for any sample,
the cumulative number of defectives found (i.e., those in the present
sample plus those found in all previous samples) exceeds the upper
limit specified for that sample, sampling is terminated and the lot is
rejected.

- If the cumulative number of defectives is less than or equal to the

lower limit, sampling is terminated and the lot is passed.

- If the number is between the two limits, another sample is taken.

The process continues until the lot is either accepted or rejected.
Choosing a Plan- The cost and time required for inspection often dictate
the kind of sampling plan used. The two primary considerations are the
number of samples needed and the total number of observations
required.
- Single-sampling plans involve only a single sample, but the sample
size is large relative to the total number of observations taken
under double-or multiple-sampling plans.

- In a situation where the cost to obtain a sample is relatively high

compared to the cost to analyze the observations, a single-sampling
plan would be more desirable. For instance, if a sample of moon
soil is needed, clearly the cost of returning for a second or third
sample would far outweigh the cost of analyzing a single large
sample.

- Conversely, in situations where item inspection costs are relatively

high, such as destructive testing, it may be better to use double or
multiple sampling, because the average number of items inspected
per lot will be lower. This stems from the fact that if the lot quality
is either very good or very poor, this will often show up initially,
and sampling can be terminated.

4.4 Continuous Improvement

Continuous improvement is a philosophy that
seeks to improve any and all factors that are
related to the process of converting inputs into
outputs. It covers equipment, methods,
materials, and people.

- A key part of the continuous improvement

philosophy is belief that improvement
efforts should never cease.

 Under continuous improvement, the old adage “If it ain’t broke, don’t
fix it” gets transformed into “Just because it isn’t broke doesn’t mean
it can’t be improved.”

 The concept of continuous improvement is not new, but until recently

did not receive much interest in the United Sates, even though it
originated here. However, many Japanese companies have used it for
years, and it has become a corner stone of the Japanese approach to
production.

- The Japanese use the term Kaizen to

refer to continuous improvement. The
successes of Japanese companies have caused other companies
to reexamine many of their approaches. One result has been a
strong interest in the continuous improvement approach.

 A successful continuous improvement process can only occur if the

environment is conducive to such an approach. This places certain
requirements on management.
The Continuous Improvement Process

Process improvement activities are often

conducted by teams. The improvement process
typically involves these steps:

1. Select a process for improvement, and set

goals for improvement.
2. Study and document the current
process.
3. Seek ways to improve the process.
4. Design an improved process.
5. Implement the improved system.
6. Evaluate it.
7. Document the improved solution,
communicate it to all concerned, and
establish appropriate training in the new
system.
 Steps 4-6 may be repeated as necessary until satisfactory results have
been obtained. The process is summarized in Figure 4-6

A number of key elements in the improvement cycle further illustrate

the nature of the activities involved. These include:

1. Standardization (documentation)
2. The plan-do-check-act cycle.Methods and tools used.
3. Methods and tools used.

These are described briefly in the following sections.

Figure 4-6
The continuous improvement process

Select a
process
Document

Study/document
Evaluation

Seek ways to
improve it

Implement the
improved process
Design an
improved process
STANDARDIZATION
 Standardization is the process of carefully
documenting the best way to perform an
operation.
 It applies in two ways. One is before
attempting to improve the process.
 This involves determining the best current
method of performing the operation under
consideration, and documenting it.
 This not only provides insight into the
process, but also a basis of comparison for
any improvements that are made.
 The second application of standardization
occurs after the process has been improved:
 Document the new method so that it becomes
the new standard for the operations.
 It is then a reference for training workers, a
reference for other, related processes, and for
further improvement, perhaps at a later time.
 Standardization illustrates the disciplined
approach frequently used in continuous
improvement activities.
 It is important to carefully document
improvements; this can be difficult to adhere
to, particularly in the excitement of frequent
achievements.

THE PLAN-DO-CHECK-ACT CYCLE

The plan-do-check-act (PDCA) cycle, also
referred to as either the Shewhart cycle or the
Deming Wheel, is the conceptual basis for
continuous improvement activities.
 The cycle is illustrated in Figure 4-7.
Representing the process with a circle
underscores its continuing nature.
 Plan

Act
Do

Check

Figure 4-7 The PDCA cycle

There are four basic steps in the cycle:

Plan. Begin by studying the current process.

Standardize that process. Then collect
data in order to identify problems. Next,
analyze data, and develop a plan for
improvement. Specify measures to be used
for evaluating the plan.

Do. Implement the plan, on a small scale if

possible. Document any changes that are
made during this phase. Collect data
systematically for evaluation.
Check. Evaluate the data collection during the
do phase. Check to see how closely the
results match the original goals of the
plan phase.

 Act. If the results are successful,

standardize the new method and
communicate the new method to all people
associated with the process.
 Implement training for the new method.
 Consider opportunities for replicating the
same changes in other, similar process.
 If the results are not successful, either
revise the plan and repeat the process, or
cease this project.
ISO 9000
 The International Organization for
Standardization (ISO) is an international
organization whose purpose is to promote
worldwide standards that will improve
operating efficiency, improve productivity,
and reduce costs.

 ISO 9000 is a series of standards that outline

the requirements for Quality Management
Systems (QMS).

 There are essentially five standards

associated with the ISO 9000 Series:
ISO 9000
ISO 9001
ISO 9002
ISO 9003
ISO 9004

 The first standard, ISO 9000, is a set of

guidelines for the selection and use of
 standards ISO 9001, 9002, and 9003, which
relate to quality assurance.
 ISO 9001 pertains to companies involved in
the design, development, production,
installation, and servicing of products or
services.
 ISO 9002 has coverage similar to ISO 9001,
but excludes companies involved in design
and development (i.e., it pertains to
companies involved in the production,
installation, and servicing of products or
services).

 ISO 9003 pertains to companies involved in

final inspection and test, such as distributors
or value-added contractors.
 ISO 9004 is employed as a guideline for the
application of the elements of the Quality
Management System.

 The ISO 9000 series is an essential factor for

companies that want to deal with the
European Community (EC). They must go
through a process that involves documenting
quality procedures and onsite assessment.

- The purpose of this is to achieve certification

as a company that meets ISO 9000
standards. The process often takes one to one and a half

years. With certification comes registration in

an ISO directory that companies seeking
suppliers can refer to for a list of certified
companies.

 In addition to the obvious benefits of

certification for companies who want to deal
with the EC, the ISO 9000 certification and
registration process is particularly helpful for
companies that do not currently have a
quality management system; it provides
guidelines for establishing the system and
making it effective.

ISO 9000 CERTIFICATION

 Why is it important to become ISO 9000

certified? For one reason, it is essential from a
purely competitive standpoint.
 Consider the situation where you need to
purchase parts for your firm and several
suppliers offer similar parts at similar prices.
 Assume that one of these firms has been ISO
9000 certified and the others have not.
 From whom would you purchase? There is no
doubt that the ISO 9000 certified company
 would have the inside track in your decision
making.
 Why? Because ISO 9000 specifies the way the
supplier firm operates as well as its quality
standards, delivery times, service levels, and
so on.

Figure 4-8
ISO 9000 Standards: Areas of Application in Production Flow, and
Guidelines for Use

Design/ Procurement ISO Production

9003 Installation Servicing
Development

ISO 9003

ISO 9002
ISO 9001

There are three forms of certification:

1. First party: A firm audits itself against
ISO 9000 standards.
 2. Second party: A customer audits its
supplier.
3. Third party: A “qualified” national or
international standards or certifying
agency serves as auditor.

The best certification of a firm is through a

third party. Once passed by the third-party audit a firm is
certified and may be registered and recorded as having achieved ISO
9000 status, and it becomes part of a registry of certified companies.
This third-party certification also has legal advantages in the European
Community. For example, a manufacturer is liable for injury to a user
of the product. The firm, however, can free itself from any liability by
showing that it has used the appropriate standards in its production
process and carefully selected its suppliers as part of its purchasing

requirement. For this reason, there is strong motivation to

choose ISO 9000-certified suppliers.