OPEMAN130 Quality Measures: Attributes and Variables

CHAPTER 3: STATISTICAL PROCESS CONTROL (SPC) - The quality of a product or service can be evaluated
using either an attribute of the product or service or a
variable measure.
3.1 STATISCAL PROCESS CONTROL
Attributes
- A method that involves monitoring the production
- It is a product characteristic such as color, surface
process to detect and prevent poor quality.
texture, cleanliness, or perhaps smell or taste.
- A statistical procedure using control charts to see if
any part of a production process is not functioning - It can be evaluated quickly with a discrete response
properly and could cause poor quality. such as good or bad, acceptable, or not, or yes or no.
SAMPLE Variables
- It is a subset of the items produced to use for - It is a product characteristic that is measured on a
inspections. continuous scale, such as length, weight, temperature,
- It can be a single item or a group of items. or time.

Since a variable evaluation is the result of some form

of measurement, it is sometimes referred to as a
quantitative classification method. An attribute
evaluation is sometimes referred to as a qualitative
classification.

SPC Applied to Services

- Control charts have historically been used to monitor

the quality of manufacturing processes. However, SPC
is just as useful for monitoring quality in services. The
- If a sample point is outside the limits, the process
difference is the nature of the “defect” being measured
may be out of control, and the cause is sought so
and monitored.
that the problem can be corrected. - Control charts for service processes tend to use quality
- If the sample is within the control limits, the characteristics and measurements such as time and
process continues without interference but with customer satisfaction. Following is a list of several
continued monitoring. different services and the quality characteristics for
each that can be measured and monitored with control
Two Reasons why the Production Process might Vary:
charts:
• Inherent random variability of the process o Hospitals - Timeliness and quickness of care,
(equipment and machinery, system used for staff responses to requests, accuracy of lab
measurement, etc.) tests, cleanliness, courtesy.
o Grocery stores - Waiting time to check out,
• Unique and special causes that are identifiable and
quality of food items, cleanliness, customer
can be corrected (defective materials, broken
complaints, checkout register errors.
equipment, errors due to lack of training, etc.)
o Airlines - Flight delays, waiting time at ticket
SPC in Quality Management counters and check-in, agent and flight
attendant courtesy, accuracy of flight
- SPC is a tool individuals can use to monitor
production or service processes for the purpose of information.
making improvements. o Fast-food restaurants - Waiting time, customer
complaints, cleanliness, food quality, order
Steps to Correct the Problem accuracy, employee courtesy.
 Identify the cause. (Quality-control tools: o Web-order companies - Order accuracy,
Brainstorming, Pareto Charts, Histograms, Check customer service representative knowledge and
Sheets, Quality Circles, and Fishbone Diagrams)
courtesy, packaging, delivery time.
 Problem-solving initiatives of the management.
o Insurance companies - Billing accuracy, The Normal Distribution
timeliness of claims processing, agent
availability and response time.

Where to use control charts?

- It might be the most effective way to ensure the

highest quality, it is costly and time consuming. In
general, control charts are used at critical points in
The formulas for conducting upper and lower limits in
the process where historically the process has
control charts are based on a number of standard
shown a tendency to go out of control, and at deviations, “z,” from the process average (e.g., centre
points where it is particularly harmful and costly if line) according to a normal distribution. Occasionally z
the process goes out of control. is equal to 2.00, but most frequently it is 3.00. A z value
of 2.00 corresponds to an overall normal probability of
(3.2) CONTROL CHARTS
95%, and z = 3.00 corresponds to a normal probability
- Control charts are graphs that visually show if a of 99.73%.
sample is within statistical control limits. (3.3) CONTROL CHARTS FOR ATTRIBUTES
Two basic purposes: - The quality measures used in attribute control
• To establish the control limits for a process. charts are discrete values reflecting a simple
• To monitor the process to indicate when it is out decision criterion, such as good or bad. A p-chart
of control. uses the proportion of defective items in a sample
Control charts exist for attributes and variables, and as the sample statistic; a c-chart uses the actual
there are several different types of control charts number of defects per item in a sample.
within each category. P - Chart
- A control chart that uses the proportion of defective
Attributes: Variables: items in a sample.
• P-charts • mean (x) - Used when both total size and number of defects
• C-chart • range (R) can be computed.

- With a p-chart, a sample of n items is taken

Process Control Chart periodically from the production or service process,
and the proportion of defective items in the sample
is determined to see if the proportion falls within
the control limits on the chart. Although a p-chart
employs a discrete attribute measure (i.e., number
of defective items) and thus is not continuous, it is
assumed that as the sample size (n) gets larger, the
normal distribution can be used to approximate the
distribution of the proportion defective.

FORMULA: UCL = p̅ + z σp
LCL = p̅ − z σp
Where:
A line through the centre of a graph that indicates the z = the number of standard deviations from the
process average, and lines above and below the centre process average
line that represent the upper and lower limits of the p¯ = the sample proportion defective; an estimate of
process. the process average
σp = the standard deviation of the sample proportion

The sample standard deviation is calculated as

√ P ( 1− p ) is also used to monitor the effects of process
σ p= where n is the sample size improvement theories.
n
C – Chart

- A control chart that uses the number of defective (3.5) CONTROL CHART PATTERNS
items in a sample. - Even though a control chart may indicate that a
- Used to compute only the number of defects and process is in control.
cannot compute the proportion that is defective. - It is not random if it shows:
- is used when it is not possible to calculate a - The sample values display a consistent pattern, even
proportion of defective items. within the control limits, it suggests that this pattern
- the actual number of defects must be used. has a non-random cause that might warrant
FORMULA: UCL =C̅ + Z √C investigation.
LCL = C̅ − Z √C - The sample values are consistently above (or below)
the center line for an extended number of samples or
(3.4) CONTROL CHART FOR VARIABLES if they move consistently up or down, there is
X -CHART probably a reason for this behavior; that is, it is not
random.
- Mean (x -̅ ) chart indicates how sample results relate
to the process average or mean. Examples of Non-Random Control Charts Patterns
- Used to show the quality averages of the samples
Unnatural patterns on control charts are defined in the
taken from the given process.
following:
- Each time a sample of a group of items is taken
• Trends: A trend can be defined as a continuous
from the process, the mean of the sample is
movement in one direction (either upward or
calculated and plotted on the chart.
downward).
It is constructed in two ways: • Shifting patterns: A shift may be defined as a sudden
- The standard deviation is known. or abrupt change in the average of the process.
- The standard deviation is not known. • Systematic variation: One of the characteristics of a
natural pattern is that the point-to-point fluctuations
When Standard Deviation is are unsystematic or unpredictable. In systematic
When Standard Deviation is
known: variations, a low point is always followed by a high
unknown:
one or vice versa.
UCL = x̅ + z σx̅ UCL = x̅ ̅ + A2R̅
• Stratification patterns: This pattern tends to cluster
LCL = x̅ − z σx̅ LCL = x̅ ̅ − A2R̅
around the center line. It may indicate improperly
calculated control limits.
R- Chart • Cycles: Cyclic behavior of the process can be
recognized by a series of high portions or peaks
- Range (R- chart) reflects the amount of dispersion interspersed with low portions or troughs.
present in each sample. • Mixtures: This pattern is indicated when the plotted
- Used to monitor process dispersion. points tend to fall near the control limits, with
relatively few points near the center line. A mixture
FORMULA: UCL = D4R̅ pattern is caused by two or more overlapping
LCL = D3R̅ distributions generating the process output.

NON- RANDOM PATTERNS IN CONTROL CHART

USING X-BAR AND R-CHARTS TOGETHER

- An X-bar and R (range) chart is a pair of control charts

used with processes that have a subgroup size of two
or more. The standard chart for variables data, X-bar
and R charts help determine if a process is stable and
predictable. The X-bar chart shows how the mean or
average changes over time and the R chart shows
how the range of the subgroups changes over time. It
• Whereas, tolerances, are design specifications
below a designed target value-nominal value.

Process Capability refers to the natural variation of a

process relative to the variation allowed by the design
specifications.

It answers the question: "How capable is the process of

producing acceptable units according to the design
specification?"

ELEMENTS ASSOCIATED WITH PROCESS CAPABILITY

• Process Variability - natural range if variation of
the process.
• Process Centre - mean
• Design Specification
RUN PROCESS CAPABILITY FIGURE
- A pattern in a control chart is characterized by a
sequence of sample observations that display the same
characteristics.

Pattern Test

- A test that determines if the observations within the

limits of a control chart display a non-random pattern.
- 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, and 3-
sigma limits.
- The pattern of sample observations in these zones is
then used to determine if any non-random patterns
exist. Thus, to calculate the dividing lines between each
of the three zones for an x̅ -chart, we use ⅓ A2.

Sample Size Determination

- It is the act of choosing the number of observations or

replicates to include in a statistical sample.

Sample size
- The number of completed responses your survey
receives. It's called a sample because it only represents
part of the group of people (or target population) whose
opinions or behavior you care about.

(3.6) PROCESS CAPABILITY

DIFFERENCE BETWEEN CONTROL LIMITS AND

TOLERANCES

• Control limits are different from tolerances.

• Control limits provide a means for determining or

identifying natural variations in a production PROCESS CAPABILITYMEASURES
process that reflect customer requirements for a
product. Process Capability Ratio
- A measure of the capability process to meet n=the sample size (selected randomly)
design specification. c=the acceptable number of defective items in a
sample
d=the actual number of defective items in a sample
If the process capability ratio is: PRODUCER’S AND CONSUMERS RISKS
• less than 1.0, it means that the process range is • When an acceptance-sampling plan is designed,
greater than the tolerance range, making the management specifies a quality standard commonly
process not capable of producing within the referred to as the acceptable quality level (AQL).
design specification.
• equals to 1.0, the tolerance range and process • Acceptable quality level (AQL) is an indication of the
range are the same. acceptable proportion of defects in a lot to the
• greater than 1.0, the tolerance range is greater consumer. The AQL may be determined by
than the process range. management to be the level that is generally
acceptable in the marketplace and will not result in a
Process Capability Index loss of customers. Or it may be dictated by an
- A measure that indicates if the process mean individual customer as the quality level it will accept.
has shifted away from the design target, and in • The probability of rejecting a production lot that
which direction it has shifted- that is, if it is off- has an AQL is referred to as the producer’s risk,
center. the probability of rejecting a lot that has an AQL.
- Measures specifically the capability of the • Although the customer expects to receive some of
process relative to the upper and lower these lots, there is a limit to the number of
specification. defective items the customer will accept. This
SUPPLEMENT TO CHAPTER 3 upper limit is known as the lot tolerance percent
defective, or LTPD. LTPD is also generally
OPERATIONAL DECISION-MAKING TOOLS: ACCEPTANCE negotiated between the producer and consumer.
SAMPLING
• The probability of accepting a lot in which the
• Sampling is a historical approach to quality control fraction of defective items exceeds the LTPD is
based on the premise that some acceptable referred to as the consumer’s risk, the probability
number of defective items will result from the of accepting a lot in which the fraction of
production process. The producer and customer defective items exceeds LTPD.
agree on the number of acceptable defects,
normally measured as a percentage. OPERATING CHARACTERISTIC CURVE
• Sampling is simply a means of identifying products - It is a graph that shows the probability of accepting
to throw away or rework. It does nothing to a lot for different quality levels with a specific
prevent poor quality and to ensure good quality in sampling plan.
the future.
AVERAGE OUTGOING QUALITY
SINGLE-SAMPLE ATTRIBUTE PLAN
- It is the expected number of defective items that
• Acceptance sampling, the decision of accepting or will pass on to the customer with a sampling plan.
rejecting a production lot based on the number of
defects in a sample. DOUBLE-AND MULTIPLE-SAMPLING PLANS
• In acceptance sampling, a random sample of the
units produced is inspected, and the quality of this - In a double-sampling plan, a small sample is taken
sample is assumed to reflect the overall quality of first; if the quality is very good, the lot is accepted,
all items or a particular group of items, called a lot. and if the sample is very poor, the lot is rejected.
• Sampling plan that provides the guidelines for - A multiple-sampling plan, also referred to as a
accepting a lot. sequential-sampling plan, generally employs the
• A single-sample attribute plan has as its basis an smallest sample size of any of the sampling plans
attribute that can be evaluated with a simple, we have discussed.
discrete decision, such as defective or not defective
or good or bad. The plan includes the following
elements:
N=the lot size