IES 331 Quality Control
Chapter 14
Acceptance Sampling for Attributes –
Single Sampling Plan and Military Standard
Week 13
August 30 – September 1, 2005
1
Dr. Karndee Prichanont
IES331 1/2005
Acceptance Sampling

Another area of quality control and improvement

Closely connected with inspection and testing of product

Inspection can occur at many points in a process
Acceptance Sampling: the inspection and classification of a
sample of nits selected at random from a larger batch or lot
and ultimate decision about disposition of the lot –
Lot Disposition or Lot Sentencing
Two common points of inspection
2
Dr. Karndee Prichanont
IES331 1/2005
Acceptance Sampling
3
Dr. Karndee Prichanont
IES331 1/2005
3 approaches for sentencing
1. _________________
• For low cost or low impact material on the subsequent process
• From trusted supplier
3. ___________________
• when 100% inspection is too costly
• Trusted supplier with potential risk
2. ____________________
• Inspect every item in the lot, then remove the defective units by
discarding, reworking, or returning to suppliers
• For critical process. Defective input may result high failure cost
• From new supplier
4
Dr. Karndee Prichanont
IES331 1/2005
Advantages and Disadvantages of
Sampling
Advantages


Less expensive because of
less inspection

___________________

___________________
Rejection on entire lot
motivates quality
improvement for suppliers
Disadvantages

Risk of _______________

Risk of _______________


Requires planning and
documentation
Require extensive study on
customer’s requirement
5
Dr. Karndee Prichanont
IES331 1/2005
Types of Sampling Plan
Sampling Plan for Attributes
1.
Single-sampling plan
2.
Double-sampling plan
3.
Multiple-sampling plan
4.
Sequential sampling plan
Sampling Plan for Variables
___________________________
_____________________________
6
Dr. Karndee Prichanont
IES331 1/2005
Lot Formation
1.
Lots should be ______________

produced on the same machines, by same operators, from
common raw materials, at approximately the same time
period
2.
Larger lots are better than smaller lots
3.
Lots should be conformable to the material
handling systems

___________________________________
7
Dr. Karndee Prichanont
IES331 1/2005
Random Sampling



Inspected units should be
selected at random
Inspected units should
represent all items in the
lot
Potential bad story
____________________
__________________
__________________
8
Dr. Karndee Prichanont
IES331 1/2005
Single-Sampling Plans for Attributes

A lot of size N has been submitted for inspection

Sample size n

Acceptance number of defective c

Lot sentencing is based on one sample of size n
N = 10,000
Example
N = 10,000
n= 89
c=2
n = 89
9
Dr. Karndee Prichanont
IES331 1/2005
Single-Sampling Plans for Attributes
The OC Curve

Operating characteristic (OC) curve
Curve plots the probability of accepting
the lot (Pa) versus the lot fraction
defective (p)

Also refer to Chapter 2; Page 78
10
Dr. Karndee Prichanont
IES331 1/2005
Single-Sampling Plans for Attributes
The OC Curve
11
Dr. Karndee Prichanont
IES331 1/2005
Single-Sampling Plans for Attributes
Ideal OC curve



Theoretically , it can be
achieved by
___________________
Due to sampling, ideal OC
curve cannot be perfectly
achieved.
If sample size (n) _______,
the OC curve shape will
approach the ideal OC curve
12
Dr. Karndee Prichanont
IES331 1/2005
Single-Sampling Plans for Attributes
Effects of n on OC curves

As sample size (n)
__________, the OC curve
shape will approach the
ideal OC curve

Note that c is kept in
proportional to n

____________________
13
Dr. Karndee Prichanont
IES331 1/2005
Single-Sampling Plans for Attributes
Effects of c on OC curves

Changing acceptance
number (c) does not
significantly change the
shape

_____________________
_____________________
14
Dr. Karndee Prichanont
IES331 1/2005
Single-Sampling Plans for Attributes
OC curve
Acceptable
quality level (AQL)
______________________________________


The poorest quality level for the supplier’s process that a consumer
would consider to be acceptable
A property of the supplier’s manufacturing process, not a property of
the sampling plan
______________________________________
Lot tolerance percent defective (LTPD)




The protection obtained for individual lots of poor quality
Also called rejectable quality level (RQL) and the limiting
quality level (LQL)
LTPD is a level of lot quality specified _____________, not a
characteristic of the sampling plan
Sampling plans can be designed to have specified performance at
the AQL and the LTPD points
15
Dr. Karndee Prichanont
IES331 1/2005
Single-Sampling Plans for Attributes
Type-A and Type-B OC Curves
16
Dr. Karndee Prichanont
IES331 1/2005
Designing a single-sampling plan with
a specified OC curve
To construct sampling plan such that
Probability of acceptance for lots with fraction defective p1
Probability of acceptance for lots with fraction defective p2
17
Dr. Karndee Prichanont
IES331 1/2005
Binomial
Nomograph
18
Dr. Karndee Prichanont
IES331 1/2005
Single-Sampling Plans for Attributes
Rectifying Inspection

When the lot is rejected, it requires corrective action
19
Dr. Karndee Prichanont
IES331 1/2005
Average outgoing quality - AOQ
Given the lot size of N, sample size
of n, and fraction defective of p



n items in the sample that, after
inspection, contain no defectives,
because all discovered defectives
are replaced
If N >> n, then
N – n items that, if the lot is
rejected, also contain no
defectives
N – n items that, if the lot is
accepted p(N-n) defectives
20
Dr. Karndee Prichanont
IES331 1/2005
Average outgoing quality - AOQ
21
Dr. Karndee Prichanont
IES331 1/2005
Military Standard 105E


Sampling procedure for inspection by attributes developed during
World War II and is the most widely used acceptance-sampling
system for attributes in the world today
A collection of sampling schemes; therefore an acceptancesampling system

Provides for three types of sampling: single, double, and multiple

Primary focal point is the acceptable quality level (AQL)


Different AQLs may be designated for different types of defects:
critical, major, and minor

Generally specified in contract or by authority responsible for sampling
Sample size is determined by lot size and by choice of inspection
level
22
Dr. Karndee Prichanont
IES331 1/2005
Military Standard 105E
23
Dr. Karndee Prichanont
IES331 1/2005
Military Standard 105E Procedure
1.
Choose the AQL
2.
Choose the inspection level
3.
Determine the lot size
4.
Fine the appropriate sample size code letter (from Table
14-4)
5.
Determine the appropriate type of sampling plan to use
(single)
6.
Enter the appropriate table to find the type of plan to be
used
7.
Determine the corresponding normal and reduced
inspection plans to be used when required
24
Dr. Karndee Prichanont
IES331 1/2005
Discussion of Military Standard

Several points about MIL STD 105E should be emphasized:

MIL STD 105E is AQL-oriented



Not all possible sample sizes are possible
(2,3,5,8,13,20,32,50, etc.)
Sample sizes are related to lot sizes
Switching rules are subject to criticism for both
misswitching between inspection plans and discontinuation
even though there has been no actual quality deterioration
25