06-Mar-15

Outline
Attribute
Control Charts for Nonconforming Units
Control Charts for Count of
Nonconformities
A Quality Rating System

Photo by brizzle born and bred - Creative Commons Attribution-NonCommercial-ShareAlike License https://www.flickr.com/photos/20654194@N07 Created with Haiku Deck

Learning Objectives Learning Objectives contd.

When you have completed this chapter you should: When you have completed this chapter you
Know limitations of variable control charts and the should:
different types of attribute charts. Know how to minimize the effect of variable
Know the objectives of the p chart group and the subgroup size.
applicable distribution. Know the applications of the c chart group, the
Be able to construct a: applicable distribution and two conditions.
Fraction defective chart- fixed subgroup size Be able to construct a c chart and a u chart and
Fraction defective chart-variable subgroup size know the difference between them.
Percent defective chart
Know the three classes of defect severity
Number defective chart

Attribute Attribute

The term Attribute refers to those quality A nonconformity is a departure of a quality

characteristics that conform to characteristic from its intended level or
specifications or do not conform to state that occurs with a severity sufficient
specifications. to cause an associated product or service
not to meet a specification requirement.
Attribute are used:
1. Where measurements are not possible.
Defect is concerned with satisfying
intended normal, or reasonably
2. Where measurements can be made but are not
made because of time, cost, or need. foreseeable, usage requirement.

1
 06-Mar-15

Attribute Attribute

Defect is appropriate for use when evaluation Defective =defect and is appropriate for
is in terms of usage. use when unit of product or service is
Nonconformity is appropriate for conformance evaluated in terms of usage rather than
to specifications. conformance to specifications.
The term Nonconforming Unit is used to Limitations of variable control charts:
describe a unit of product or service containing
These charts cannot be used for quality
at least one nonconformity.
characteristics which are attributes.

Attribute The P Chart

Types of Attribute Charts: The P Chart is used for data that consist of the
1. Nonconforming Units (based on the proportion of the number of occurrences of an
Binomial distribution): p chart, np chart. event to the total number of occurrences.
2. Nonconformities (based on the Poisson It is used in quality to report the fraction or
distribution): c chart, u chart. percent nonconforming in a product, quality
characteristic, or group of quality
characteristics.

The P Chart The P Chart

np It can be used to control one quality
Formula:
p characteristic, as is done with X bar and R chart,
n Or to control a group of quality characteristics
of the same type or of the same part,
The fraction nonconforming, p, is usually Or to control the entire product.
small, say, 0.10 or less. It can be established to measure the quality
Because the fraction nonconforming is very produced by a work center, by a department, by
small, the subgroup sizes must be quite large a shift, or by an entire plant.
to produce a meaningful chart.

2
 06-Mar-15

The P Chart The P Chart

It is frequently used to report the Objectives of the P Chart:
performance of an operator, group of 1. Determine the average quality level (AQL):
operators, or management as a means of This information provides the process
evaluating their quality performance. capability in terms of attributes.
The subgroup size of the P chart can be 2. Bring to the attention of management any
changes in the average.
either variable or constant.
3. Improve the product quality: Ideas for
quality improvement.

The P Chart The P Chart

Objectives of the P Chart contd: P-Chart Construction for Constant Subgroup
Size
4. Evaluate the quality performance of
1. Select the quality characteristic(s):
operating and management personnel.
a) Single quality characteristic.
5. Suggest places to use X bar and R chart: b) Group of quality characteristics.
They are more sensitive to variation. c) A part.
d) An entire product.
6. Determine acceptance criteria of a e) A number of products.
product before shipment to the customer. f) It can be established for performance control of an
operator, work center, department, shift, plant, or
corporation

The P Chart The P Chart

P Chart Construction for Constant Subgroup Size P Chart Construction for Constant Subgroup Size
contd. contd.
2. Determine the subgroup size and method: 3. Collect the data:
The size of the subgroup is a function of the
At least 25 subgroups.
proportion nonconforming.
Different sources (Check sheet).
A minimum size of 50 is suggested as a starting
For each subgroup the proportion
point.
nonconforming is calculated by the formula P =
np/n

3
 06-Mar-15

The P Chart The P Chart

P Chart Construction for Constant Subgroup Size P Chart Construction for Constant Subgroup Size
4. Calculate the trial central line and the control contd.
limits:
p
np 5. Establish the revised central line and control
n limits.
pnew p0
np np d
p (1 p )
UCL p 3 n n d
n
p0 (1 p0
p (1 p ) UCL p0 3
LCL p 3 n
n
p0 (1 p0 )
LCL p0 3
n

The P Chart The P Chart

The P Chart is most effective if it is posted
where operating and quality personnel can view
A P Chart will also indicate long-range
it. trends in quality, which will help to
evaluate changes in personnel, methods,
The control limits are usually three standard
deviations from the central value. Therefore, equipment, tooling, materials, and
approximately 99% of the plotted points, P, will inspection techniques.
fall between the upper and lower control limits. P-chart is based on the binomial
distribution.

The P Chart
The P Chart
P Chart Construction for Variable Subgroup Size
P Chart Construction for Variable Subgroup Size contd.
1. Collect the data. 2. As the subgroup size gets larger, the control
limits are closer together.
2. Determine the trial central line and control
limits: Since the subgroup size changes each 3. Establish revised central line and control limits:
day, limits must be calculated for each day.

4
 06-Mar-15

The P Chart The P Chart

P Chart Construction for Variable Subgroup Size P Chart Construction for Variable Subgroup
contd.
Size contd.
If Po is known, the process of data collection
Po is the standard or reference value of
and trial control limits is not necessary.
the proportion (fraction) nonconforming
P is the proportion (fraction) nonconforming in
based on the best estimate of PBar.
a single subgroup.
is the population proportion (fraction)
P is the average proportion (fraction)
nonconforming.
nonconforming of many subgroup.

The P Chart The P Chart

Minimizing the Effect of Variable Subgroup Size
Minimizing the Effect of Variable Subgroup Size contd.
1. Control limits for an average subgroup size: By using an
average subgroup size, one limit can be calculated and Case I: This case occurs when a point (subgroup fraction
placed on the control chart. nonconforming) falls inside the limits and its
nav
n subgroup size is smaller than the average subgroup
g size.
p0 (1 p0 Case II: This case occurs when a point (subgroup fraction
UCL p0 3
nav nonconforming) falls inside the average limits and its
p0 (1 p0 ) subgroup size is larger than the average subgroup
LCL p0 3
nav size.

The P Chart The np Chart

Minimizing the Effect of Variable Subgroup Size contd. Number Nonconforming Chart (np):
Case III: This case occurs when a point (subgroup
The np chart is easier for operating
fraction nonconforming) falls outside the limits and
personnel to understand than the p chart.
its subgroup size is larger than the average subgroup
size. The limitation that this chart has is that the
Case IV: This case occurs when a point (subgroup
subgroup size needs to be constant.
fraction nonconforming) falls outside limits and its
subgroup size is less than the average subgroup size.

5
 06-Mar-15

The np Chart The np Chart

Central Line = np0 Number Nonconforming Chart (np):
Control Limits = np0 3 np0 (1 p0 ) If the fraction nonconforming po is
unknown, then it must be determine by
collecting data, calculating trial control
limits, and obtaining the best estimate of
po.

Control Charts for Count of Non-

Process Capability conformities
For an attribute this process is much simpler.
The nonconformities chart controls the count
The process capability is the central line of of nonconformities within the product or
the control chart. service.
Management is responsible for the capability. An item is classified as a nonconforming unit
whether it has one or many nonconformities.
When the plotted point is outside the control
limit, operating personnel are usually Count of nonconformities (c) chart.
responsible. Count of nonconformities per unit (u) chart.

Control Charts for Count of Non- Control Charts for Count of Non-
conformities conformities
Since these charts are based on the Objectives:
Poisson distribution, two conditions must 1. Determine the average quality level: This
be met: information gives the initial process
1. The average count of nonconformities must be capability.
much less than the total possible count of 2. Bring to the attention of management any
nonconformities. changes in the average.
2. The occurrences are independent. 3. Improve the product quality: Ideas for
quality improvement.

6
 06-Mar-15

Control Charts for Count of Non- Control Charts for Count of Non-
conformities conformities
C Chart construction:
Objectives contd.:
1. Select the quality characteristic(s):
4. Evaluate the quality performance of a) Single quality characteristic.
operating and management personnel. b) Group of quality characteristics.
5. Suggest places to use X bar and R chart. c) A part.
6. Determine acceptance criteria of a d) An entire product.
product before shipment to the e) A number of products.
customer. f) It can be established for performance control of an:
operator, work center, department, shift, plant, or
corporation

Control Charts for Count of Non- Control Charts for Count of Non-
conformities conformities
C Chart construction contd: c-Chart Construction contd:
2. Determine the subgroup size and method:
4. Calculate the trial central line and the control
3. Collect the data: limits:
c
At least 25 subgroups. c
g
Different sources.
UCL c 3 c
LCL c 3 c

Control Charts for Count of Non- Control Charts for Count of Non-
conformities conformities
C chart construction contd: C chart construction contd:
5. Establish the revised central line and control
limits 6. Achieve the objectives: The reason for the

c new = c 0 = c - cd control chart is to achieve one or more of the

g - gd previously stated objectives.
UCL = c 0 + 3 c 0
LCL = c 0 - 3 c0

7
 06-Mar-15

Control Charts for Count of Non- Control Charts for Count of Non-
conformities/Unit conformities
Chart for Count of Nonconformities/Unit (u Chart for Count of Nonconformities/Unit
Chart) (u Chart)
u
c
u
c
n n Scale selected is continuous for the u chart.
u For the c chart is discrete.
UCL u 3
n Subgroup size for the u chart can vary. For
LCL u 3
u the c chart is 1.
n
The u chart is limited in that we do not know
the location of the nonconformities.

Control Charts for Demerits/Unit

A Quality Rating System
D wc uc wma uma wmi umi
Nonconformity Classification: when wc , wma and wmi are 9, 3 and 1
1. Critical nonconformities: Unsafe conditions for individuals D 9uc 3uma 1umi
using, maintaining, or depending upon the product.
D0 9u0 c 3u0 ma 1u0 mi
2. Major nonconformities: Result in failure or reduce materially
the usability of the product for its intended purpose.
0u 9 2 u0 c 32 u0 ma 12 u0 mi
3. Minor nonconformities: Reduce materially the usability of
the product for its intended purpose.
UCL D0 3 0 u
LCL D0 3 0 u

Control Chart Selection

In Class Exercise
Quality Characteristic
Variable Attribute Calculate the trial control limits and revised
Defective Defect
no tendency line for a non-conformities data
n>1? x and MR
(refer text book exercise) :
yes constant
Assume all out-of-control points have an
yes constant
p or sampling
sample
np unit?
n>=10?
no
x and R
size? assignable causes
yes no
no
Refer to Table 1
yes
p-chart with c u
x and s variable sample
size

8
 06-Mar-15

Thank you.