TOTAL QUALITY MANAGEMENT AND

RELIABILITY

P.CHINNA RAO
ASSISTANT PROFESSOR
DEPARTMENT OF MECHANICAL ENGINEERING
RGUKT-AP
SRIKAKULAM CAMPUS
 Learning Objectives

• Concept of Acceptance Sampling

• Acceptance sampling by variables and attributes
• Design of sampling plans
 Acceptance Sampling

Accepting or Rejecting a lot (of parts, components, etc.) based

on the inspection of sample drawn from it.
 Acceptance Sampling

 Accept/Reject entire lot based on sample results

 Created by Dodge and Roming during world war II

 Not consistent with TQM of zero defects.

 Does not estimate the quality of the process of lot.

 What is Acceptance Sampling?
A lot acceptance sampling:

 A SQC technique where a random sample is taken from a lot and upon the results of the appraising the
sample, the lot will be either accept or reject.

 Procedure for sentencing incoming batches or lots of items without doing 100% inspection.

 The most widely used sampling plans are given by Military Standard (MIL-STD-105E).

Purposes:

 Determine the quality level of an incoming shipment or end of the production.

 Judge whether the quality level is within the level that has been predetermined.
 Acceptance Sampling

 Acceptance sampling:

• The inspection and classification of a sample of units selected at random from a large batch or lot
and ultimate decision about disposition/sentencing the lot.

 Another area of quality control and improvement.

 Closely connected with inspection and testing of product.

 Inspection can occur many points in a process.

 Two common points of inspection:

 When parts are received

 After production
 Acceptance Sampling

Sampling Plans:

 Sampling Plans:

• Plans that specify lot size, sample size, number of samples and acceptance/rejection criteria.

Types of Sampling Plans:

 Sampling by Attributes vs Sampling by Variables.

 Incoming vs Outgoing inspection.

 Rectifying vs Non rectifying inspection.

 Single sampling
 Double sampling
 Multiple sampling and sequential sampling
 Acceptance Sampling

Sampling Terms:

 Acceptance Quality Level (AQL):

• The percentage of defects at which consumers are willing to accept lots as GOOD.

 Lot Tolerance Percent Defective (LTPD) or Rejection Quality Level (RQL):

• The upper limit on the percentage of defects that consumer is willing to accept.

 Consumer’s Risk:

• The probability that a lot contained defectives exceeding the LTPD will be accepted.

 Producer’s Risk:

• The probability that a lot contained acceptable quality level will be rejected.
A LOT – by – LOT sampling Plan:

N n Count Number Accept or Reject

(Lot Size) (Sample Size) conforming a lot

 Specify the plan ( n, c) given N .

 For a lot size N, determine

• The sample size n.
• The acceptance number c
 Reject a lot if number of defective > c

 Specify course of action if lot is rejected.

Quality Definitions:

 Acceptance Quality Level (AQL):

• The smallest percentage of defectives that will make the lot definitely acceptable. A quality level
that is the base line requirement of the customer.

 Lot Tolerance Percent Defective (LTPD) or Rejection Quality Level (RQL):

• Quality level that is unacceptable to the customer.

 Acceptance Sampling
Out going Inspection:

Ship
PROCESS INSPECTION CUSTOMER

Receiving/Incoming Inspection:

Ship
PROCESS INSPECTION CUSTOMER

Disposition of Lots:

Accept Ship
PROCESS INSPECTION CUSTOMER
Reject

Scrap Rework
 Single Sampling Plan

Single Sampling Plan Advantages:

 Less expensive because of less inspection

o Works with single sample
o Protects both producer and consumer

 Rejection of entire lot motivates quality improvement for suppliers

Single Sampling Plan disadvantages:

 Risk of accepting a lot with poor quality

 Risk of rejecting a lot of acceptance quality

 Requires extensive study on customer’s requirements

Lot Formation:

 Lots should be such that:

• Produced on the same machines, by same operators, from common raw materials, at
approximately the same time period.

 Larger lots are better than smaller lots:

• These are more representative of overall quality.

 Lots should be conformable to the material handling systems and personnel

Single Sampling Plan:

 The most common and easiest plan to use but not most efficient in terms of average number of samples
needed.

 One sample drawn from the lot and 100% inspected

 Single sampling plan:

• N = Lot size
• n = sample size (randomized)
• c = acceptance number
• d = number of defective items in sample

 Rule: if d < c , accept lot; else reject the lot

Single Sampling Procedure:

Take a randomized
sample of size n
from the lot N

Inspect all items in the sample

Defectives found = d

YES
d≤c? Accept lot

NO

Reject lot

Return lot to Do 100%

supplier inspection
Producer’s and consumer’s risks due to mistaken sentencing of lot:

 TYPE-I ERROR = P ( Reject good lot ):

• α or Producer’s risk
• 5% is common

 TYPE-II ERROR = P ( Accept bad lot ):

• β or Consumer’s risk
• 10% is typical value
Single Sampling Procedure:

Take a randomized Probability of Acceptance, P(A):

sample of size n
from the lot of • Assuming the lot proportion non-conforming is ‘p’
quality p
• Binomial distribution can be used to find the
probability of observing x non-conforming items in a
Inspect all items in the sample sample of size n.
Defectives found = d

𝑛
𝑛 𝑑
YES 𝑃𝑎 = 𝑃 𝑑 ≤ 𝑐 = 𝑝 (1 − 𝑝)𝑛−𝑑
d≤c? Accept lot 𝑑
𝑑=0

• Lot Quality = p
NO

Reject lot

Return lot to Do 100%

supplier inspection
 Operating Characteristic Curve [OC Curve]

 It is the graph between percentage

defective/Fraction defective of the lot and
probability that lot is going to be accepted.

 From graph we can say as percentage

defective increases probability of acceptance
decreases.

 Producer risk is shown in graph which is risk

to producer that lot will get rejected even if
defects are within acceptable quality level.

 Consumer risk is shown in graph which is risk

to consumer that lot will get accepted even if
defects are more than rejection quality level.
Design of sampling plan:
Effect of Sample Size on OC curve:
Effect of Acceptance number on OC curve:
Types of OC curve:

 TYPE-A OC Curve:

• Gives the probability of acceptance for an individual lot coming from Finite Production.

• The probability that accepting a lot is calculated by Hyper Geometric Distribution.

 TYPE-B OC Curve:

• Gives the probability of acceptance for an individual lot coming from continuous process or
infinite size lot

• The probability that accepting a lot is calculated by Binomial Distribution.

• If the lot size is large and probability of non conforming is small, a poisons distribution can be
used.
Double Sampling Plan:

 Take a small initial sample:

 If # defectives < lower limit, ACCEPT

 If # defectives > upper limit, REJECT
 If # defectives between limits take second sample

 Accept or Reject lot based on two samples

 Less inspection than single sampling

Multiple/Continuous Sampling Plan:

 Advantage: Uses smaller sample sizes

 Take initial sample:

 If # defectives < lower limit, ACCEPT

 If # defectives > upper limit, REJECT
 If # defectives between limits, re sample

 Continue sampling until accept or reject lot based on all sample data
Sequential Sampling Plan:

 The ultimate extension of the multiple sampling plan is the sequential sampling plan,
which is an item-by-item inspection plan.

 After each item is inspected, a decision is made to accept the lot, reject the lot, or choose
an other item for inspection, depending on whether the observed cumulative number of
non-conforming items is less than or equal to the acceptance number, greater than or equal
to the rejection number, or in between the two.