Guidelines for
Acceptance Sampling for Inspections
Document Number: 04.03.0008.005, Rev. 0
Effective Date: May 23, 2017
Ernest Orlando Lawrence Berkeley National Laboratory
University of California
Berkeley, CA 94720
Prepared under the U.S. Department of Energy Contract No. DE-AC02-05CH11231.
Guidelines for Acceptance Sampling for Inspection Office of Institutional Assurance and Integrity
Document No. 04.03.008.005, Rev. 0 Page 2
Effective Date: 5/23/17
TABLE OF CONTENTS
Revision History Table ................................................................................................................3
1.0 Introduction .......................................................................................................................4
2.0 Quality and Risk ...............................................................................................................4
3.0 Inspection Levels ..............................................................................................................5
A. General Inspection Levels ...........................................................................................5
B. Special Inspection Levels ............................................................................................5
4.0 Sampling Plans and Results .............................................................................................6
5.0 Nonconforming Items/ Services ........................................................................................7
TABLE I – Sample Size Code Letters .........................................................................................8
ATTACHMENT A - SINGLE-STAGE SAMPLING NORMAL INSPECTION PLAN & RESULTS
.....................................................................................................................................................9
ATTACHMENT B - DOUBLE-STAGE SAMPLING NORMAL INSPECTION PLAN & RESULTS
...................................................................................................................................................10
Guidelines for Acceptance Sampling for Inspection Office of Institutional Assurance and Integrity
Document No. 04.03.008.005, Rev. 0 Page 3
Effective Date: 5/23/17
Revision History Table
Date Revision # Description of Revision
Guidelines for Acceptance Sampling for Inspection Office of Institutional Assurance and Integrity
Document No. 04.03.008.005, Rev. 0 Page 4
Effective Date: 5/23/17
1.0 Introduction
This guidance document provides procedures and methodologies for sampling items and/or services by
attributes (i.e. conforming or nonconforming) during inspection activities, and is applicable to
acceptance inspection activities performed by Lawrence Berkeley National Laboratory (LBNL)
personnel. The sampling guidelines are derived from standard industry quality assurance inspection
practices and ANSI/ASQCZ1.4-1993, Sampling Procedures and Tables for Inspection by Attributes,
which is based on statistical sampling principles.
Quality Control (QC) inspections should be performed of the following:
• End items
• Components and raw materials
• Operations
• Materials in process
• Supplies in storage
• Maintenance operations
• Data or records
• Administrative procedures
• Other items, as determined
Inspections may take place locally on site or at the manufacturing or testing facility. Such inspections
may be performed to validate or verify that items and/or services meet requirements at the point of
acceptance or at certain points during a process (e.g. design, manufacturing, plating, welding, etc.)
meet requirements prior to completion and/or acceptance of an item or service.
If a lot or batch of the same type of items and the total number is greater than 50 units each, it may not
be practical to inspect 100% of them. In instances where 100% receipt inspection is not practical or
achievable, a sampling procedure and methodology may be used.
The sampling procedures and methodologies outlined in this guidance document include an inherent
risk-based graded approach, which takes into consideration potential or actual risk associated with an
item or service not performing as intended.
1.1 Baseline References
• I.76 DEAR 970.5203-1, Management Controls (Jun 2007) (Deviation)
• 10 CFR 830, Nuclear Safety Management
• DOE O 414.1D, Quality Assurance
1.2 Referenced Documents
• LBNL/PUB-3111, Quality Assurance Program Description (QAPD)
• LBNL/PUB-5519, Issues Management Program
• LBNL Design and Construction Manual, Chapter 5, Suspect Counterfeit Items
• ANSI/ASQCZ1.4-1993, Sampling Procedures and Tables for Inspection by Attributes
2.0 Quality and Risk
When developing an acceptance sampling plan, two levels of quality should be considered – the Lab’s
risk and the subcontractor’s risk.
Guidelines for Acceptance Sampling for Inspection Office of Institutional Assurance and Integrity
Document No. 04.03.008.005, Rev. 0 Page 5
Effective Date: 5/23/17
The first is the acceptable quality limits (AQL), which is the quality level expected by the Lab. The AQL
is the “least tolerable level of quality for an item or service and it decides on the maximum number of
defective units, beyond which a batch is rejected.
The quality level is typically communicated in the subcontract to the subcontractor. The subcontractor
takes on the risk that the quality of a lot (i.e. a group of items) cannot be verified by the sampling plan
and the Lab rejects the lot. Typically, a supplier’s acceptable level of risk is 5% (0.05) of the quantity.
The second level of quality is the worst level of quality that the Lab can tolerate and would reject if the
level of quality cannot be validated. The Lab’s would take on risk if it accepted the lowest level of
item/service quality. Typically, a consumer’s acceptable level of risk is 10% (0.10) of the total quantity.
3.0 Inspection Levels
When determining the appropriate sample of items to inspect, the following General/Special Inspection
Levels (GILs) should be considered. Refer to Table I-Sample Size Code Letters.1
A. General Inspection Levels
Quality Control for consumer goods typically are inspected under General Inspection Levels (GILs).
There are three GILs, as follows:
Level I (L-I, Reduced Inspection)
L-I is used for suppliers of the items/services that have rarely provided the Lab with nonconforming
items, and the number of new items/services from that supplier to be inspected does not warrant
100% inspection.
Level II (L-II, Normal Inspection)
L-II is typically the default inspection level. L-II is used for suppliers of items/services that have
varied performance identified through inspection where nonconformances have been identified, but
does not indicate a historical problem with quality.
Level III (L-III, More Rigorous Inspection)
Level III is used for high-value items/services or for suppliers of items/services who have had recent
and/or historical quality problems (meaning, multiple nonconformances identified during inspection,
the number of nonconformities identified is above the AQL, etc.) and require a higher level of rigor
of inspection.
B. Special Inspection Levels
Special Inspection Levels (SIL) can be applied in cases where it is critical that a particular
acceptance criteria is met, only very few samples can be checked (e.g. destructive tests, container
packaging, etc.) and will either take a significant amount (e.g. multiple hours or days) of time or
destroy the samples.
1
American Society for Quality, ANSI/ASQCZ1.4-1993, American National Standard Sampling Procedures and Tables for Inspection by Attributes, Tables I –
Sample Size Code Letters, p.10.
Guidelines for Acceptance Sampling for Inspection Office of Institutional Assurance and Integrity
Document No. 04.03.008.005, Rev. 0 Page 6
Effective Date: 5/23/17
The four additional Special Inspection Levels (i.e. S-1, S-2, S-3 and S-4) may be used where
relatively small sample sizes are necessary and larger sampling risks can be tolerated.
4.0 Sampling Plans and Results
Based on the inspection levels, the inspector develops and implements a single-stage or double-stage
sampling plan, and documents the results of the sampling plan.
A single-stage sampling plan is a decision rule to accept or reject a lot based on the results of one
random sample from the lot.
1. Identify the acceptance criteria to which the sample size needs to meet.
2. Identify the AQL that the sample size needs to meet.
3. Take a random sample size (n), based on the inspection level, and inspect each item within the
sample.
4. If the number of defects does not exceed the AQL (or c), the lab may accept the entire lot.
5. Any defective items found in the sample are dispositioned by either being repaired or returned
to the supplier.
6. If the number of defects in the sample is greater than the AQL, the lab should
a. perform 100 percent inspection of the lot or
b. reject the entire lot and returns it to the supplier.
Double-stage sampling plans dictate that a smaller sample size (n1) is inspected, if the number of
defects in n1 is above certain limits, and more samples are selected for inspection. Double-stage
sampling specifies two sample sizes (n1 and n2) and two AQLs (c1 and c2). If the quality of the lot is
very good or very bad, the consumer can make a decision to accept or reject the lot on the basis of the
first sample, which is smaller than in the single-sampling plan.
1. Identify the acceptance criteria to which the sample size needs to meet.
2. Identify the AQL that the sample size needs to meet.
3. Specify two sample sizes (n1 and n2) and two AQLs (c1 and c2).
4. Take a random sample of size n1 based on the inspection level selected.
5. If the number of defects is less than or equal to c1, the lab may accept the lot.
6. If the number of defects is greater than c2, the lab should reject the lot.
7. If the number of defects is between c1 and c2, take a second sample of size n2.
8. If the combined number of defects in the two samples is less than or equal to c2, the lab should
accept the lot.
9. If the combined number of defects in the two samples is greater than or equal to c2, the lab
should reject the lot.
A single document may be used for both the sampling plan and associated results provided the
following information is identified. Standard Sampling Plans may be developed in advance of, and used
for, the inspection activities. Refer to Tables II – III2 and the example Sampling Plan & Results for
Receipt Inspection to develop, or determine which, the sampling plan to be used. Results of inspections
to the sampling plans are an indication of a supplier’s performance and capability to meet quality and
other technical requirements, and should be documented and maintained as part of the project files.
Consistent poor performance of a supplier’s items/services should be brought to the attention of the
Project Management team and the Procurement representative in order for adverse performance by a
supplier can be adequately managed.
Sampling plans should identify the following:
2
American Society for Quality, ANSI/ASQCZ1.4-1993, American National Standard Sampling Procedures and Tables for Inspection by Attributes, Tables II-
III, Single and Double Sampling Plans, p.10-16.
Guidelines for Acceptance Sampling for Inspection Office of Institutional Assurance and Integrity
Document No. 04.03.008.005, Rev. 0 Page 7
Effective Date: 5/23/17
1. Items/services to be inspected,
2. PO associated with items/services to be inspected,
3. Subcontractor/Supplier that provided the items/services,
4. Acceptance criteria that the Items/services must meet (e.g. the specifications/requirements that
the items/services are required to meet),
5. Lot or batch number,
6. Lot or batch size,
7. Inspection level (L-I, L-II, L-III/ S-1, S-2, S-3, S-4)
8. Type of plan (Single-stage or Double-stage)
9. Sample size, and
10. AQL (i.e. the number of acceptable nonconformances for the lot).
Sampling Results should identify the following:
1. Person who inspected items/services to the plan,
2. Date the inspection took place, and
3. Results of the inspection (i.e. acceptance criteria is met or not, AQL is met or not):
a. Which acceptance criteria was met or not met,
b. The total number of nonconformances identified for the lot/batch,
c. What the disposition is for the nonconforming items/services,
d. Did the sample pass or fail, and
e. If a second sample is needed (if Double-Stage sampling plan was used).
5.0 Nonconforming Items/ Services
Items identified as nonconforming, regardless of whether or not the AQL is met, will need to be
dispositioned in accordance with Design and Construction Manual, Chapter 5, Suspect Counterfeit
Items. Services/performance identified as nonconforming will need to be dispositioned in accordance
with LBNL/PUB-5519, Issues Management Program.
Guidelines for Acceptance Sampling for Inspection Office of Institutional Assurance and Integrity
Document No. 04.03.008.005, Rev. 0 Page 8
Effective Date: 5/23/17
TABLE I – Sample Size Code Letters
Guidelines for Acceptance Sampling for Inspection Office of Institutional Assurance and Integrity
Document No. 04.03.008.005, Rev. 0 Page 9
Effective Date: 5/23/17
ATTACHMENT A - SINGLE-STAGE SAMPLING NORMAL INSPECTION PLAN & RESULTS
Guidelines for Acceptance Sampling for Inspection Office of Institutional Assurance and Integrity
Document No. 04.03.008.005, Rev. 0 Page 10
Effective Date: 5/23/17
ATTACHMENT B - DOUBLE-STAGE SAMPLING NORMAL INSPECTION PLAN & RESULTS
