-​ Build quality into the process from

MODULE 1
start to finish.
-​ Don't just find what you did wrong –
eliminate the "wrongs" altogether.
-​ Use statistical control methods – not
physical inspections alone – to prove
that the process is working.

Point 4: Stop Making Decisions Purely on

Point 1: Create a Vision and Demonstrate
the Basis of Cost
Commitment
-​ The supplier and manufacturer must
-​ An organization must define its
be considered as a macro
values, mission, and vision of the
organization.
future to provide long-term direction
-​ Deming urged businesses to
for its management and employees.
establish long-term relationships
-​ Businesses should not exist simply
with fewer suppliers, leading to
for profit; they are social entities
loyalty and opportunities for mutual
whose basic purpose is to serve their
improvement.
customers and employees.

Point 5: Improve Constantly and Forever

Point 2: Learn the New Philosophy
-​ Improved design of goods and
-​ Companies must take a services comes from understanding
customer-driven approach based on customer needs and continual
mutual cooperation between labor market surveys and other sources of
and management and a never-ending feedback, and from understanding
cycle of improvement. the manufacturing and service
delivery process.
-​ Improvements in operations are
Point 3: Stop depending on inspections
achieved by reducing the causes and
-​ Inspections are costly and unreliable impacts of variation, and engaging
– and they don't improve quality, all employees to innovate and seek
they merely find a lack of quality.
 ways of doing their jobs more fear of relinquishing control, and
efficiently and effectively. fear of change.
-​ Deming chain reaction: When -​ Fear encourages short-term thinking
quality improves, productivity -​ Fear is a cultural issue for all
improves and costs decrease. organizations
-​ Continuous improvement

Point 9: Break down barriers between

Point 6: Institute Training departments.

-​ Training -​ Build the "internal customer"

-​ results in improvements in quality concept – recognize that each
and productivity department or function serves other
-​ adds to worker morale departments that use their output.
-​ demonstrates to workers that the -​ Build a shared vision.
company is dedicated to helping -​ Use cross-functional teamwork to
them and investing in the future build understanding and reduce
adversarial relationships.

Point 7: Institute Leadership -​ Focus on collaboration and

consensus instead of compromise
-​ The job of management is
leadership, not supervision.
Point 10: Get rid of unclear slogans.
Supervision
-​ Let people know exactly what you
-​ simply overseeing and directing
want – don't make them guess.
work
"Excellence in service" is short and
Leadership memorable, but what does it mean?
-​ providing guidance to help How is it achieved? The message is
employees do their hobs with less clearer in a slogan like "You can do
effort. better if you try."
-​ Don't let words and nice-sounding
phrases replace effective leadership.
Point 8: Drive Out Fear
Outline your expectations, and then
-​ Fear is manifasted in many ways: praise people face-to-face for doing
fear of failure, fear of the unknown, good work.
 -​ Organizations must invest in their
Point 11: Enumerate Numerical Quotas and people at all levels to ensure success
Management by Objective (MBO) in the long term

-​ Many organizations manage by the -​ Developing the worth of the

numbers. individual is a powerful motivation

-​ Goals are useful, but numerical method

goals set for others without

incorporating a method to reach the Point 14: Take Action
goal generate frustration and
-​ Any culture change begins with top
resentment. management and includes everyone
-​ Management must understand the -​ Team-based approach
system and continually try to
improve it, rather than focus on
short-term goals.

MODULE 2
Point 12: Remove Barriers to Pride in
Workmanship

-​ Allow everyone to take pride in their QUALITY MANAGEMENT

work without being rated or -​ In today’s business environment,
quality can be broadly defined as
compared.
the extent to which a product or
-​ Treat workers the same, and don't service meets or exceeds a
make them compete with other customer’s expectations.
-​ For any product, the quality of
workers for monetary or other
that product is determined by
rewards. Over time, the quality two primary factors.
system will naturally raise the level -​ These are DESIGN quality and
PROCESS quality.
of everyone's work to an equally
high level. TWO PRIMARY FACTORS
1.​ Design Quality
2.​ Process Quality
Point 13: Encourage Education and
Design quality
Self-Improvement
-​ can be described as the quality
-​ Continuing, broad education for that a product has in terms of the
self-improvement
 actual characteristics of the -​ He was an American engineer,
product. statistician, professor and author.
-​ Dr. Deming has an extensive list
of published works
-​ most well known for Deming’s
14 points and the Deming Cycle.

DEMING CYCLE
-​ -​ The Deming Cycle or Deming
Wheel is also known as PDCA,
Process quality or “Plan, Do, Check, Act.”
-​ refers to the ability of the -​ It is a version of continuous
organization to produce the good improvement that emphasizes the
or service having perfect quality continuous nature of process
at each stage of the process, or in improvement.
other words, manufacturing
defect- free products.

QUALITY GURUS

1.​ WALTER SHEWMART

(1891-1967)
-​ Dr. Shewhart was an American
physicist, engineer and
statistician. 3.​ JOSEPH JURAN (1904-2008)
-​ He is known as the father of -​ Joseph Juran also came upon the
statistical quality control and work of Vilfredo Pareto (1848-
spent much of his career 1923) and made the Pareto
researching variation and is Principle, also known as the
credited with the creation of the 80/20 rule, well known today as
first control chart. a tool for problem solving and
-​ His work focused around the continuous improvement.
need to reduce variation in order -​ Other examples of the 80/20 rule
to improve quality.
• 80% of a company’s profits are
2.​ EDWARD DEMING generated by 20% of the products or
(1900-1993) services sold.
-​ Dr. Deming is likely the most • 80% of the continuous improvement
well known of the Quality Gurus. ideas are generated by 20% of the
employees
 -​ Examples include quality
improvement initiatives,
4.​ PHILIP CROSBY (1926-2001) employee training, upgrading of
-​ Crosby was an American equipment, implementing quality
businessman and author. procedures and making proactive
-​ He published Quality is Free in design changes.
1979. He believed that the costs
of quality are often understated. 2.​ Appraisal Cost
-​ He coined the phrase zero defects -​ All money spent in checking and
and felt that there was no reason testing of product during the
for any errors. production process would be
-​ He taught that it is less considered Appraisal costs.
expensive to do it right the first -​ Wages of inspectors when
time rather than to pay for extra defined as part of the process,
inspection, scrap, rework and testing labs and equipment,
repairs. gauging, and process control,
would be included in this
5.​ ARMAND FIEGENBAUM category.
(1920-2014)
-​ Dr. Feigenbaum was an 3.​ Internal Failure Cost
American quality engineer and -​ Once a defect has been produced,
businessman. with any luck the organization
-​ He was the Director of will detect the error before it
Manufacturing Operations at leaves the building and is sent to
General Electric from the customer.
1958-1968. -​ Often, defective products can be
-​ He devised the concept of total repaired, but all of the extra time
quality control, which later spent on the rework is considered
became total quality management internal failure costs.
(TQM).
4.​ External Failure
COST OF QUALITY -​ Once a defective product has
The various costs of quality can be been shipped to the customer,
broken down into the following four the costs then become external
categories: prevention costs, appraisal failure costs.
costs, and failure costs, which are further -​ Replacement product, expedited
classified as internal failure costs and shipping, potential law suits,
external failure costs. product recalls, and of course
loss of future business are all
1.​ Prevention Cost external failure costs.
-​ Prevention costs include all the
funds spent to prevent the QUALITY SYSTEMS
occurrence of defects.
Three common quality systems that identify quality, efficiency, and other
many organizations use in work-related problems, to propose
order to manage their quality are: solutions, and to work with management
• Total Quality Management (TQM) in implementing their recommendations.
• International Organization for
Standardization (ISO) 3. Continuous Improvement
• Six Sigma An integral part of TQM is continuous
improvement: the commitment to
1.​ TOTAL QUALITY making constant improvements in the
MANAGEMENT design, production, and delivery of
-​ To compete today, companies goods and services. Improvements can
must deliver quality goods and almost always be made to increase
services that satisfy customers’ efficiency, reduce costs, and improve
needs. customer service and satisfaction.
-​ This is the objective of quality Everyone in the organization is
management. constantly on the lookout for ways to do
-​ Total quality management things better
(TQM), or quality assurance,
includes all the steps that a 2.​ ISO / The International
company takes to ensure that its Organization for
goods or services meet or Standardization (ISO)
exceed the customers defined -​ The International Organization
specifications and are of for Standardization (ISO) is an
sufficiently high quality to meet international standard- setting
customers’ needs. body composed of
representatives from various
Generally speaking, a company standards organizations.
adheres to TQM principles by -​ Founded on 23 February 1947,
focusing on three tasks: the organization promotes
worldwide proprietary, industrial
1. Costumer Satisfaction and commercial standards.
Companies that are committed to TQM -​ It is headquartered in Geneva,
understand that the purpose of a business Switzerland, and works in 164
is to generate a profit by satisfying countries.
customer needs. They encourage
customers to tell them how to make the -​ ISO is an independent, non-
right products, both goods and services, governmental organization and is
that work the right way. the largest developer of voluntary
international standards. Use of
2. Employee Involvement these standards assists
In many companies, employees who organization to create products
perform similar jobs work as teams, and services that are safe, reliable
sometimes called quality circles, to and of good quality.
 -​ The standards help businesses 5.​ Control the improved process
increase productivity while and future process performance.
minimizing errors and waste.

3.​ Six Sigma

-​ is a set of techniques and tools TOOLS FOR QUALITY
for process improvement. IMPROVEMENT
-​ It was introduced by engineer
Check sheets Control Charts ​
Bill Smith while working at
Motorola in 1980. Histograms ​ Scatter Diagrams
-​ It is a comprehensive quality Pareto Charts​ Cause and
system for achieving business Effect Diagrams
success by minimizing variation
in business processes. 1.​ CHECK SHEETS
-​ Unique to Six Sigma is the -​ This is a custom designed form
DMAIC Improvement Model. used to record the number of
-​ In Six Sigma, improvements are occurrences of a particular
made by the use of project teams. outcome of interest.
A team may be assembled in -​ It may collect basic information
order to fix a quality issue, to such as how many incidents
streamline an old process or to occurred, the timing, or the
develop a new process. measurement that was
non-conforming.
The team will use the DMAIC model,
which is an acronym that stands for 2.​ HISTOGRAMS
Define, Measure, Analyze Improve and -​ Raw data from a check sheet may
Control. be put into a histogram.
-​ Data that is continuous in
nature can be put into a
DMAIC Histogram that contains ranges of
1.​ Define the problem, the data.
improvement activity, -​ It will show an accurate
opportunity for improvement, the representation of the distribution
project goals, and customer of the data
(internal and external)
requirements. 3.​ CONTROL CHARTS
2.​ Measure process performance. -​ In order to monitor the
3.​ Analyze the process to determine performance of a process over
root causes of variation and poor time, a control chart is the
performance (defects). appropriate tool.
4.​ Improve process performance by -​ A Control Chart includes an
addressing and eliminating the Upper Control limit and a lower
root causes. Control limit, which are used to
 control the quality dimension
that is measured.

4.​ PARETO CHART

-​ A special type of bar chart that
shows the number of occurrences
of a particular characteristic,
ordered from highest to lowest.
-​ A Pareto analysis helps us focus
our attention on the defects that
occur the most frequently and to
allocate the resources
accordingly.

5.​ SCATTER DIAGRAM

-​ A simple diagram helps to figure
out if there is a relationship
between two variables.

6.​ CAUSE & EFFECT

DIAGRAM
-​ Also known as a Fishbone
diagram, it was developed by Dr.
Ishikawa to help identify the
causes of a problem.
Module 3: STATISTICAL PROCESS -​ SPC tools and procedures can
CONTROL help you monitor process
behavior, discover issues in
internal systems, and find
Module 3 introduces statistical process solutions for production issues.

control, an important topic in statistical Statistical process control is often

used interchangeably with
quality control.
statistical quality control (SQC).
Statistical process control (SPC)

-​ is a useful statistical procedure

to achieve high-quality products
and services.
-​ With the use of control charts,
you will be able to

The Basics of Statistical Process

SPC Control

-​ is an essential part of quality

Control Charts for Attributes
management.
-​ Many companies, committed to
Control Charts for Variables
quality management, provide
extensive and continual training
Control Chart Patterns
in SPC methods to nearly all their
workers.
SPC with Excel and OM Tools
-​ They make extensive use of SPC
for their continuous process
Process Capability
improvement.

SPC TOOLS

Statistical process control (SPC)

A popular SPC tool

-​ is defined as the use of statistical

-​ is the control chart, originally
techniques to control a process or
developed by Walter Shewhart in
production method.
the early 1920s.
A control chart (independent variables). Although both
terms are often used interchangeably,
-​ helps one record data and lets
SQC includes acceptance sampling
you see when an unusual event,
where SPC does not.
such as a very high or low
observation compared with
"typical" process performance, THE 7 QUALITY CONTROL (7-QC)
occurs. TOOLS

Control charts attempt to distinguish

In 1974, Dr. Kaoru Ishikawa brought
between two types of process variation:
together a collection of process

​ Common cause variation, improvement tools in his text Guide to

which is intrinsic to the Quality Control. Known around the

process and will always be world as the seven quality control

present (7-QC) tools, they are:

​ Special cause variation, which ​ Cause-and-effect diagram (also

stems from external sources called Ishikawa diagram or
and indicates that the process fishbone diagram)
is out of statistical control
​ Check sheet

Various tests can help determine when ​ Control chart

an out-of-control event has occurred. ​ Histogram

However, as more tests are employed, ​ Pareto chart
the probability of a false alarm also ​ Scatter diagram
increases. ​ Stratification

SQC VERSUS SPC THE 7 SUPPLEMENTAL (7-SUPP)

TOOLS
Statistical quality control (SQC) is
defined as the application of the 14 In addition to the basic 7-QC tools, there
statistical and analytical tools (7-QC and are also some additional statistical
7-SUPP) to monitor process outputs quality tools known as the seven
(dependent variables). Statistical process supplemental (7-SUPP) tools:
control (SPC) is the application of the
same 14 tools to control process inputs ​ Data stratification
 ​ Defect maps plotted point represents the
​ Events logs algebraic sum of the previous

​ Process flowcharts ordinate and the most recent

​ Progress centers deviations from the target.

​ Randomization ​ Exponentially Weighted Moving

Average (EWMA) charts: Each
​ Sample size determination
chart point represents the
HISTORY OF SPC weighted average of current and
all previous subgroup values,
A marked increase in the use of control
giving more weight to recent
charts occurred during World War II in
process history and decreasing
the United States to ensure the quality of
weights for older data.
munitions and other strategically
important products. The use of SPC
methods diminished somewhat after the
war, though was subsequently taken up
with great effect in Japan and continues
to the present day. (For more
information, see the History of Quality.)

Many SPC techniques have been

adopted by organizations throughout the
globe in recent years, especially as a
component of quality improvement
initiatives like Six Sigma. The
widespread use of control charting
procedures has been greatly assisted by
statistical software packages and
sophisticated data collection systems.

Additional process-monitoring tools

include:

​ Cumulative Sum (CUSUM)

charts: The ordinate of each
CHAPTER 3 process control (SPC), to monitor and
STATISTICAL PROCESS CONTROL improve manufacturing processes in
After reading this chapter, you will be order to remove and prevent defects.
able to:
❑Explain when and how to use A major topic in statistical quality
statistical process control to ensure the control is statistical process control.
quality of products and services. Statistical process control (SPC) is a
❑Discuss the rationale and procedure statistical procedure using control charts
for constructing control charts. to see if any part of a production process
❑Utilize attribute control charts to is not functioning properly and could
determine if a process is in control. cause poor quality. SPC is used to
❑Utilize variable control charts to inspect and measure the production
determine if a process is in control. process to see if it is varying from what
❑Identify control chart patterns and it is supposed to be doing.
describe appropriate data collection.
❑Develop control charts using Excel THE BASICS OF STATISTICAL
and OM Tools. PROCESS CONTROL
❑Assess the process capability of a
process. Process control is achieved by taking
periodic samples from the process and
QUALITY CONTROL IN plotting these sample points on a chart,
SMARTPHONE MANUFACTURING to see if the process is within statistical
control limits. A sample can be a single
Because of the importance of quality to item or a group of items. If a sample
smartphone purchasers—often more point is outside the limits, the process
important than price—process control may be out of control, and the cause is
techniques are employed throughout the sought so that the problem can be
smartphone supply chain. All corrected. If the sample is within the
smartphone manufacturers, and control limits, the process continues
especially Samsung and Apple, have without interference but with continued
well-developed and extensive quality monitoring. In this way, SPC prevents
management systems that make use of a quality problems by correcting the
host of quantitative techniques and tools, process before it starts producing
including, very prominently, statistical defects.
 Quality Measures: Attributes and
SPC in Quality Management Variables
❑An attribute is a product
Companies use SPC to see if their characteristic such as color, surface
processes are in control—working texture, cleanliness, or perhaps smell or
properly. This requires that companies taste. Attributes can be evaluated quickly
provide SPC training on a continuing with a discrete response such as good or
basis that stresses that SPC is a tool bad, acceptable or not, or yes or no.
individuals can use to monitor Even if quality specifications are
production or service process for the complex and extensive, a simple
purpose of making improvements. attribute test might be used to determine
Through the use of statistical process whether or not a product or service is
control, employees can be made defective
responsible for quality in their area: to
identify problems and either correct ❑A variable measure is a product
them or seek help in correcting them. By characteristic that is measured on a
continually monitoring the production continuous scale such as length, weight,
process and making improvements, the temperature, or time.
employee contributes to the goal of
continuous improvement and few or no SPC Applied to Services
defects.
Control charts have historically been
A statistical control chart is a graph to used to monitor the quality of
monitor a production process. By manufacturing processes. However, SPC
monitoring a process with a control is just as useful for monitoring quality in
chart, the employee and management services as shown in photo. The
can detect problems quickly and prevent difference is the nature of the “defect”
poor-quality items from proceeding being measured and monitored. Using
through the remainder of the process and Motorola’s definition—a failure to meet
ending up as defective products that customer requirements in any product or
must be thrown away or reworked, thus service—a defect can be an empty soap
wasting time and resources. dispenser in a restroom or an error with a
phone catalog order, as well as a blemish
on a piece of cloth or a faulty tray on a
DVD player. Control charts for service ❑INSURANCE COMPANIES:
processes tend to use quality Billing accuracy, timeliness of claims
characteristics and measurements such processing, agent availability and
as time and customer satisfaction response time.
​
Following is a list of several different Where to Use Control Charts
services and the quality characteristics
for each that can be measured and Most companies do not use control
monitored with control charts. charts for every step in a process.
❑HOSPITALS: Timeliness and Although that might be the most
quickness of care, staff responses to effective way to ensure the highest
requests, accuracy of lab tests, quality, it is costly and time consuming.
cleanliness, courtesy, accuracy of In most manufacturing and service
paperwork, speed of admittance and processes, there are clearly identifiable
checkouts. points where control charts should be
❑GROCERY STORES: Waiting time used. In general, control charts are used
to check out, frequency of out-of-stock at critical points in the process where
items, quality of food items, cleanliness, historically the process has shown a
customer complaints, checkout register tendency to go out of control, and at
errors. points where if the process goes out of
❑AIRLINES: Flight delays, lost control it is particularly harmful and
luggage and luggage handling, waiting costly.
time at ticket counters and check-in, Control chart are graphs that visually
agent and flight attendant courtesy, show if a sample is within statistical
accurate flight information, passenger control limits. They have two basic
cabin cleanliness and maintenance. purposes: to establish the control limits
❑FAST-FOOD RESTAURANTS: for a process and then to monitor the
Waiting time for service, customer process to indicate when it is out of
complaints, cleanliness, food quality, control. Control charts exist for
order accuracy, employee courtesy. attributes and variables; within each
❑CATALOG-ORDER COMPANIES: category there are several different types
Order accuracy, operator knowledge and of control charts.
courtesy, packaging, delivery time,
phone order waiting time.
Process Control Chart intervals), it is not possible to compute a
proportion. In this case a c-chart is
required.

❑p-chart Uses the proportion

defective in a sample.

The Normal Distribution

CONTROL CHARTS FOR ❑c-chart Uses the number of defects

ATTRIBUTES in a sample.

The quality measures used in attribute

control charts are discrete values
reflecting a simple decision criterion
With a p-chart a sample of n items is
such as good or bad. A p-chart uses the
taken periodically from the production
proportion of defective items in a sample
or service process, and the proportion of
as the sample statistic; a c-chart uses the
defective items in the sample is
actual number of defects per item in a
determined to see if the proportion falls
sample. A p-chart can be used when it is
within the control limits on the chart.
possible to distinguish between defective
Although a p-chart employs a discrete
and non-defective items and to state the
attribute measure (i.e., number of
number of defectives as a percentage of
defective items) and thus is not
the whole. In some processes, the
continuous, it is assumed that as the
proportion defective cannot be
sample size (n) gets larger, the normal
determined. For example, when counting
distribution can be used to approximate
the number of blemishes on a roll of
the distribution of the proportion
upholstery material (at periodic
defective. This enables us to use the tend to be small, usually around 4 or 5.
following formulas based on the normal The center line of the control chart is the
distribution to compute the upper control overall process average, that is, the mean
limit (UCL) and lower control limit of the sample means.
(LCL) of a p-chart:
The x¯- chart is based on the normal
A c-chart is used when it is not possible distribution. It can be constructed in two
to compute a proportion defective and ways depending on the information that
the actual number of defects must be is available about the distribution. If the
used. For example, when automobiles standard deviation of the distribution is
are inspected, the number of blemishes known from past experience or historical
(i.e., defects) in the paint job can be data, then formulas using the standard
counted for each car, but a proportion deviation can be used to compute the
cannot be computed, since the total upper and lower control limits. If the
number of possible blemishes is not standard deviation is not known, then a
known. In this case a single car is the table of values based on sample ranges is
sample. Since the number of defects per available to develop the upper and lower
sample is assumed to derive from some control limits. We will first look at how
extremely large population, the to construct an x¯-chart when the
probability of a single defect is very standard deviation is known.
small. As with the p-chart, the normal
distribution can be sed to approximate A quality control inspector tests samples
the distribution of defects. of chocolate candy. The sample results
can be plotted on a control chart to see if
CONTROL CHARTS FOR the production process is in control. If
VARIABLES not, it will be corrected before a large
number of defective candies are
Mean (x¯-) Chart produced, thereby preventing costly
In a mean (or x¯-) control chart, each waste.
time a sample of a group of items is
taken from the process, the mean of the
sample is computed and plotted on the
chart. Each sample mean (x¯) is a point
on the control chart. The samples taken
 investigated to see if it is sustainable. If
so, new control limits would need to be
developed. Sometimes an x¯-chart is
used alone to see if a process is
improving, perhaps toward a specific
performance goal.

In other situations, a company might

have studied and collected data for a
- process for a long time and already
R- chart knows what the mean and standard
deviation of the process are; all it wants
to do is monitor the process average by
taking periodic samples. In this case it
would be appropriate to use the mean
chart where the process standard
deviation is already known,
Using X¯- and R-Charts Together

CONTROL CHART PATTERNS

Even if a control chart indicates that a

process is in control, it is possible that
the sample variations within the control
It is also possible for an R-chart to limits are not random. If the sample
exhibit a distinct downward trend in the values display a consistent pattern, even
range values, indicating that the ranges within the control limits, it suggests that
are getting narrower and there is less this pattern has a nonrandom cause that
variation. This would be reflected on the might warrant investigation. We expect
x¯-chart by mean values closer to the the sample values to “bounce around”
center line. Although this occurrence above and below the center line,
does not indicate that the process is out reflecting the natural random variation in
of control, it does suggest that some the process that will be present.
nonrandom cause is reducing process However, if the sample values are
variation. This cause must be consistently above (or below) the center
line for an extended number of samples Run A sequence of sample values that
or if they move consistently up or down, display the same characteristic.
there is probably a reason for this
behavior; that is, it is not random.

3 Control Chart Patterns

One type of pattern test divides the

control chart into three “zones” on each
side of the center line, where each zone
is one standard deviation wide. These
are often referred to as 1-sigma, 2-sigma,
A pattern in a control chart is and 3-sigma limits. The pattern of
characterized by a sequence of sample sample observations in these zones is
observations that display the same then used to determine if any nonrandom
characteristics—also called a run. One patterns exist.
type of pattern is a sequence of
observations either above or below the Recall that the formula for computing an
center line. For example, three values x¯-chart uses A2 from Table 3.1, which
above the center line followed by two assumes 3 standard deviation control
values below the line represent two runs limits (or 3-sigma limits). Thus, to
of a pattern. Another type of pattern is a compute the dividing lines between each
sequence of sample values that of the three zones for an x¯-chart, we
consistently go up or go down within the use 1/3A2. The formulas to compute these
control limits. Several tests are available zone boundaries are shown in Figure
to determine if a pattern is nonrandom or 3.4.
random.
Pattern test Determines if the
observations within the limits of a
control chart display a nonrandom SPC WITH EXCEL AND OM TOOLS
pattern. Computer software and spreadsheet
packages are available that perform
There are five general guidelines statistical quality control analysis,
associated with the zones for identifying including the development of process
patterns in a control chart where none of control charts. We will demonstrate how
the observations are beyond the control to develop a statistical process control
limits: chart on the computer using Excel and
OM Tools. The Excel spreadsheet in
1.Eight consecutive points on one side of Exhibit 3.1 shows the data for Example
the center line 3.1 in which we constructed a p-chart to
2.Eight consecutive points up or down monitor the production process for
3.Fourteen points alternating up or down denim jeans at the Western Jeans
4.Two out of three consecutive points in Company. The values for p¯, UCL, and
zone A (on one side of the center line) LCL have been computed using
5.Four out of five consecutive points in formulas embedded in cells I4, I5, and
zone A or B on one side of the center I6. To construct our control chart, it is
line necessary to enter the values for the
control chart mean (p¯) and upper and
If any of these guidelines applied to the lower control limits (UCL and LCL) in
sample observations in a control chart, it columns B, C, D, and E for all 20 sample
would imply that a nonrandom pattern points. This will provide the data points
exists and the cause should be to construct UCL and LCL lines on the
investigated. control chart.
Sample size determination
Exhibit 3.1
Exhibit 3.2

PROCESS CAPABILITY
❑Control limits are occasionally
mistaken for tolerances; however, they
are quite different things. Control limits
provide a means for determining natural
variation in a production process. They
The three main elements associated with
are statistical results based on sampling.
process capability are process variability
Tolerances are design specifications
(the natural range of variation of the
reflecting customer requirements for a
process), the process center (mean), and
product. They specify a range of values
the design specifications. Figure 3.5
above and below a designed target value
shows four possible situations with
(also referred to as the nominal value)
different configurations of these
within which product units must fall to
elements that can occur when we
be acceptable.
consider process capability.
❑Process capability refers to the
Process Capability Measures
natural variation of a process relative to
the variation allowed by the design
specifications. In other words, how
capable is the process of producing
acceptable units according to the design
specifications? Process control charts are If Cp is less than 1.0, the process range is

used for process capability to determine greater than the tolerance range, and the

if an existing process is capable of process is not capable of producing

meeting design specifications. within the design specifications all the

time. This is the situation depicted in Exhibit 3.3 shows the Excel solution
screen for the computation of the process
capability ratio and the process
capability index for Examples 3.7 and
3.8. The formula for the process
capability index in cell D16 is shown on
the formula bar at the top of the screen.
Exhibit 3.4 shows the computation of
the process capability ratio and the
process capability index for Example 3.7
and 3.8 using the “quality control”
module from the OM Tools software.
Figure 3.5a. If Cp equals 1.0, the
tolerance range and the process range are
virtually the same—the situation shown
in Figure 3.5b. If Cp is greater than 1.0,
the tolerance range is greater than the
process range—the situation depicted in
Figure 3.5c. Thus, companies would
logically desire a Cp equal to 1.0 or
greater, since this would indicate that the
process is capable of meeting
specifications.

Process Capability with Excel and OM

Tools