ASTM B697-88(2021)

This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the
Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.
Designation: B697 − 88 (Reapproved 2021)
Standard Guide for
Selection of Sampling Plans for Inspection of
Electrodeposited Metallic and Inorganic Coatings
This standard is issued under the fixed designation B697; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (´) indicates an editorial change since the last revision or reapproval.
1. Scope procedures. Such plans, when applicable, can reduce inspec-
tion cost and increase quality protection. Information on
1.1 This guide gives guidance in the selection of sampling
variables plans is given in Test Method B762, MIL-STD-414,
plans to be used in the inspection of electrodeposited and
ANSI/ASQC Z1.9-1979, and Refs (1-2).
related coatings on products for the purpose of deciding
1.6 This international standard was developed in accor-
whether submitted lots of coated products comply with the
dance with internationally recognized principles on standard-
specificationsapplicabletothecoatings.ThissupplementsTest
ization established in the Decision on Principles for the
Method B602 by giving more information on sampling inspec-
Development of International Standards, Guides and Recom-
tion and by providing additional sampling plans for the user
mendations issued by the World Trade Organization Technical
who finds the limited choice of plans in Test Method B602 to
Barriers to Trade (TBT) Committee.
be inadequate.
1.2 When using a sampling plan, a relatively small part of
2. Referenced Documents
thearticlesinaninspectionlotisselectedandinspected.Based
2.1 ASTM Standards:
on the results, a decision is made that the inspection lot either
B602 Test Method for Attribute Sampling of Metallic and
does or does not satisfactorily conform to the specification.
Inorganic Coatings
1.3 This guide also contains several sampling plans. The
B762 Test Method of Variables Sampling of Metallic and
plans are attribute plans, that is, in the application of the plans
Inorganic Coatings
each inspected article is classified as either conforming or
2.2 ANSI Standard:
nonconforming to each of the coating requirements. The
ANSI/ASQC Z1.9-1979 Sampling Procedures and Tables
number of nonconforming articles is compared to a maximum
for Inspection by Variables for Percent Nonconformance
allowable number. The plans are simple and relatively few.
2.3 Military Standards:
Additional plans and more complex plans that cover more
MIL-STD-105 Sampling Procedures and Tables for Inspec-
situations are given in the Refs (1-7) at the end of this guide
tion by Attributes
and in MIL-STD-105.
MIL-STD-414 Sampling Procedures and Tables for Inspec-
1.4 Acceptance sampling plans are used:
tion by Variables for Percent Defective
1.4.1 When the cost of inspection is high and the conse-
quences of accepting a nonconforming article are not serious. 3. General
1.4.2 When 100 % inspection is fatiguing and boring and,
3.1 Procedure—The use of acceptance sampling consists of
therefore, likely to result in errors. In these cases a sampling
a series of decisions and actions. These are listed in order
plan may provide greater protection than 100 % inspection.
below and are discussed in this standard.
1.4.3 When inspection requires a destructive test. Here,
3.1.1 Select characteristics to be inspected,
sampling inspection must be used.
3.1.2 Select type of sampling plan,
1.5 Anothergeneraltypeofacceptancesamplingplanthatis 3.1.3 Select quality level,
not covered in these guidelines is the variables plan in which
3.1.4 Define inspection lot,
measured values of characteristics are analyzed by statistical
For referenced ASTM standards, visit the ASTM website, www.astm.org, or
contact ASTM Customer Service at service@astm.org. For Annual Book of ASTM
This guide is under the jurisdiction of ASTM Committee B08 on Metallic and Standards volume information, refer to the standard’s Document Summary page on
Inorganic Coatings and is the direct responsibility of Subcommittee B08.10 on Test the ASTM website.
Methods. Available fromAmerican National Standards Institute (ANSI), 25 W. 43rd St.,
Current edition approved Oct. 1, 2021. Published October 2021. Originally 4th Floor, New York, NY 10036, http://www.ansi.org.
approved in 1981. Last previous edition approved in 2016 as B697 – 88 (2016). Available from Standardization Documents Order Desk, DODSSP, Bldg. 4,
DOI: 10.1520/B0697-88R21. Section D, 700 Robbins Ave., Philadelphia, PA 19111-5098.
Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States
B697 − 88 (2021)
3.1.5 Select sample, will be rejected, and the percentage of nonconforming articles
3.1.6 Inspect sample, in the accepted inspection lots will be the same as in the
3.1.7 Classify inspection lot, and rejected lots. In other words, acceptance sampling does not, by
3.1.8 Dispose of inspection lot. itself, result in higher quality. Rectifying inspection (3.11) will
result in higher average quality in the product leaving inspec-
3.2 The need for acceptance sampling arises when a deci-
tion.
sion must be made about what to do with a quantity of articles.
This quantity (called the inspection lot in this guide) may be a 3.7 Because acceptance sampling plans permit the accep-
shipment from a supplier, may be articles that are ready for a tance of inspection lots that contain nonconforming articles,
subsequent manufacturing operation, or may be articles ready basic to the selection of a sampling plan is a decision about the
for shipment to a customer. percentage of nonconforming articles that is acceptable. If the
function of the article is so important that no nonconformers
3.3 When acceptance sampling is done, several of the
can be tolerated, acceptance sampling cannot be used. In these
articlesintheinspectionlotareselectedatrandom(seeSection
cases, every article must be inspected, and, to guard against
7). These articles constitute the sample. Each article in the
error, may have to be inspected twice.
sample is inspected for conformance to the requirements
placed on it. If an article meets a requirement, it is classified as 3.8 The protection that an attributes sampling plan provides
conforming. If not, it is classified as nonconforming. If the against accepting an undesirable number of nonconforming
number of nonconforming articles in the sample is no more articles is determined by the size of the sample and by the
than a predetermined number (called the acceptance number), acceptance number. The protection provided by a plan is
the inspection lot is accepted. If it exceeds the acceptance usually expressed in the form of an operating characteristic
number, the inspection lot is rejected. (OC) curve. Fig. 1 is the OC curve for the plan that calls for a
sampleof55articlesandanacceptancenumberoftwo.Plotted
3.4 The disposition of rejected inspection lots is beyond the
alongthehorizontalaxisisthequalitylevelofaninspectionlot
scope of this guide because, depending on the circumstances,
expressed as the percentage of the articles in the lot that are
lots may be returned to the supplier, kept and used, put to a
nonconforming (Note 1).The vertical axis is the probability, as
different use, scrapped, reworked, or dealt with in some other
apercentage,thataninspectionlotwillbeacceptedbytheplan
way. An exception is rectifying inspection (3.11) in which
(Note 4). Inspection lots with zero percent nonconforming
rejected lots are screened and used.
articles will be accepted 100 % of the time (Note 2). As the
3.5 Because the decision about the disposition of an inspec-
percentage of nonconforming articles in the inspection lot
tion lot is based on the inspection of a sample, and because
increases, the probability of acceptance decreases. For
there is a chance that a sample will not be representative of an
example,asshowninFig.1,aninspectionlotcontaining1.5 %
inspection lot, some inspection lots that have the desired
nonconforming articles has a 95 % chance of being accepted,
quality level (Note 1) will be rejected and some inspection lots
while one containing 9.6 % nonconforming articles has only a
that do not have the desired quality level will be accepted.
10 % chance of being accepted.
Thereareonlytwosituationsinwhichtheresultsofacceptance
NOTE4—TheverticalaxisoftheOCcurvecanhavetwomeanings.One
samplingaretotallypredictable(Note2).Oneiswhenthereare
is the probability that a particular inspection lot will be accepted. The
no nonconforming articles in the inspection lot. There, of
other meaning is the percentage of a series of lots of a given quality level
course,willbenononconformingarticlesinthesampleandthe
that will be accepted. The latter meaning is the one that is strictly correct
decision to accept the lot will always be made. The other
mathematically. The former meaning is also correct, as long as the
inspection lot is at least ten times bigger than the sample.
situation is when no article in the inspection lot conforms. All
of the articles in the sample will be nonconforming and the
decision to reject the lot will always be made (Note 3).
NOTE 1—In this guide the term “quality level” means the percentage of
nonconforming articles in an inspection lot or it means the average
percentage of nonconforming articles in a series of inspection lots
received from a single source. Terms such as high quality, increased
quality, and better quality mean a relatively smaller percentage of
nonconforming articles, while terms such as low quality, decreased
quality, and poorer quality mean a relatively larger percentage of
nonconforming articles.
NOTE 2—In this discussion and elsewhere in this guide, it is assumed
that no errors are made.
NOTE 3—To be strictly correct, lots that contain no more nonconform-
ing articles than the acceptance number will always be accepted, and lots
that contain fewer conforming articles than the sample size minus the
acceptance number will always be rejected.
3.6 The discussion in 3.5 leads to two important points: (1)
acceptance sampling plans will permit the acceptance of
inspection lots that contain nonconforming articles, and (2)in
a series of inspection lots, each containing the same percentage
FIG. 1 Operating Characteristic Curve for Single Sample, Attri-
of nonconforming articles, some will be accepted and some butes Sampling Plan, Sample Size = 55, Acceptance Number = 2
B697 − 88 (2021)
3.9 The characteristics of a sampling plan are often ex-
pressed in terms of the Acceptable Quality Level (AQL) and
theLimitingQualityLevel(LQL).TheAQListhequalitylevel
that will result in the acceptance of a high percentage of
incoming inspection lots; usually it is the quality level that will
result in the acceptance of 95 % of the incoming inspection
lots. In Fig. 1, theAQL is 1.5 %. The LQL is the quality level
thatwillresultintherejectionofahighpercentageofincoming
inspection lots; usually it is the quality level that will result in
the rejection of 90 % of the incoming inspection lots. In Fig. 1
the LQL is 9.6 %. In this standard, AQL and LQL are defined
FIG. 2 Average Outgoing Quality of Rectifying Inspection with
as the quality levels that will be accepted 95 and rejected 90 %
Single Sample, Attributes Plan, Sample Size = 55, Acceptance
of the time, respectively.
Number = 2. Rejected Lots Are 100% Inspected with Nonconform-
ing Articles Removed and Replaced with Conforming Articles
3.10 Another characteristic of sampling plans that is used in
this standard is the 50/50 point. This is the quality level that
will result in the acceptance of half of the incoming inspection
16 of Ref (4) discusses this point.
lots. In Fig. 1 the 50/50 point is 4.8 %.
3.11.3 Use of rectifying inspection will assure that with a
3.11 Rectifying Inspection:
continuous series of inspection lots the average quality level of
3.11.1 As stated in 3.4, one of the options when an inspec- all the accepted articles, considered as a whole, will not be
tion lot is rejected is screening of the lot. In this procedure, worse than the AOQL of the sampling plan used. However,
called rectifying inspection, all of the articles in a rejected lot
rectifying inspection can significantly increase inspection costs
are inspected and the nonconforming ones are removed and since every rejected inspection lot is 100 % inspected. The
replacedwithconformingarticles.Thenow100 %-conforming
lower the quality of incoming lots, the more of them that will
inspection lot is accepted and is passed along with the berejectedandthen100 %inspected.Fig.3showshow,forthe
inspection lots that were accepted on the basis of acceptance
sampling plan of Fig. 1 and lots of 550, the average number of
sampling. The addition of these 100 %-conforming inspection articles inspected per inspection lot increases as the quality
lotsimprovestheaveragequalitylevelofalltheinspectionlots
levels of incoming lots decrease. In lots containing up to about
takentogether.Theamountthequalitylevelisimprovedcanbe 1.5 % nonconforming articles the increase in inspection is
calculated if the average quality level of incoming inspection
moderate. Beyond that point the average amount of inspection
lots is known. The calculations reveal that if the incoming increases rapidly. At an incoming quality level of 2.1 % the
quality level is high, few inspection lots will be rejected and
amount of inspection is doubled. And with incoming quality
screened and so the average quality of the outgoing lots will be levels of 15 % virtually every inspection lot is 100 % in-
only slightly improved over the incoming. If the quality level spected.
of the incoming inspection lots is low, many of the inspection 3.11.4 Because the cost of inspection using rectifying in-
lots will be rejected and screened. The addition of this large spection plans is so greatly influenced by the quality level of
number of 100 %-conforming lots will result in a high outgo- incoming inspection lots, past information of that level is
ing quality level. At intermediate incoming quality levels, the necessary before choosing an AOQL. The AOQL plans in
outgoing quality will be poorer than these two extremes, and Table1givetherangeofincomingqualitylevelforwhicheach
there will be a particular incoming quality level for which the plan is recommended. The cost of the inspection is also
outgoing level will be the poorest.
3.11.2 When rectifying inspection is used the average qual-
ity level of a series of outgoing lots is called the Average
Outgoing Quality (AOQ) and the worst possible AOQ for a
given plan is called the Average Outgoing Quality Limit
(AOQL). Fig. 2 is a plot of the AOQ for the sampling plan of
Fig. 1 (Note 5). This shows that the worstAOQ, theAOQL, is
2.5 % and occurs only if the average incoming quality level is
4.2 %. Fig. 2 also shows that when the quality level of
incoming lots is high, the improvement caused by inspection is
small. For example, if the incoming lots are of AQL quality,
1.5 %, theAOQ is 1.4 %.At lower incoming quality levels the
relative improvement is greater; for example, at an incoming
quality level of 3 %, the AOQ is 2.3 %.
NOTE 5—The AOQs and AOQLs in this guide are calculated on the
basis that when rejected lots are screened the nonconforming articles
found are replaced with conforming articles. If the discarded nonconform- FIG. 3 Average Number of Articles Inspected Per Lot with Recti-
ing articles are not replaced, a practice that is frequently done, theAOQs fying Inspection, Single Sampling, Attributes Plan with Sample
andAOQLs will be somewhat different from those in this guide. Chapter Size = 55, Acceptance Number = 2, and Lot Size = 550
B697 − 88 (2021)
TABLE 1 Constant AQL Plans
NOTE 1—The values listed in columns headedAQL, 50/50 Point, LQL,
andAOQLarethepercentagesofnonconformingarticlesintheinspection
lot.
NOTE 2—The AOQL values are calculated for inspection lots that are
very large compared to the sample. The values can be corrected for cases
where this is not true by multiplying them by:
sample size
lot size
Accept-
AQL Sample Size ance 50/50 Point LQL AOQL
Number
0.25 145 1 1.2 2.7 0.6
325 2 0.8 1.6 0.4
0.65 55 1 3.1 6.9 1.5
126 2 2.1 4.2 1.2 NOTE 1—Curve 2 is the plan shown in Fig. 1. Curve 1 is a plan with a
210 3 1.7 3.2 0.92
sample size of three and an acceptance number of zero; it has the same
303 4 1.5 2.6 0.84
AQLasCurve2andanLQLof54 %.Curve3isaplanwithasamplesize
612 7 1.3 1.9 0.73
of 23 and an acceptance number of zero; it has the same LQL as Curve 2
and an AQL of 0.22 %. The inset in the upper right is an enlargement of
1.5 24 1 7.0 15 3.5
the AQL region.
55 2 4.9 9.4 2.5
FIG. 4 Operating Characteristic Curves of Three Single Sample
92 3 4.0 7.1 2.1
Attributes Sampling Plans
132 4 3.5 6.0 1.9
174 5 3.3 5.3 1.8
365 7 2.9 4.4 1.7
number 2. Curve number 3 is the zero acceptance number plan
4.0 9 1 19 37 9.3
that has the same LQL as curve number 2. The sample size is
21 2 13 22 6.5
35 3 10 18 5.5 23. With this plan the AQL is 0.2 % as compared to 1.5 % for
50 4 9.3 15 5.1
curve number 2.
66 5 8.6 14 4.8
84 6 7.9 12 4.5
4. Selection of the Type of Sampling Plan
4.1 The sampling plans of this guide are given in Table 1,
Table 2, and Table 3.All are single sampling plans, that is, the
determined by the size of the inspection lot and by the size of
the sample. If rectifying inspection is to be used on a large
scale, it is recommended that the user refer to Ref (3).It
TABLE 2 Constant LQL Plans
contains plans that yield the lowest total inspection for each
NOTE 1—The values listed in columns headedAQL, 50/50 Point, LQL,
combination of AOQL, incoming quality level, and inspection
andAOQLarethepercentagesofnonconformingarticlesintheinspection
lot size.
lot.
3.11.5 Whetherthe100 %inspectionofrejectedlotsisdone
NOTE 2—The AOQL values are calculated for inspection lots that are
by the purchaser or the supplier is a business decision of the
very large compared to the sample. The values can be corrected for cases
purchaser. Having the supplier do the inspection provides an
where this is not true by multiplying them by:
incentive to improve the quality of future lots. However, if the
sample size
supplier does the 100 % inspection, the purchaser may want to 12
lot size
do sampling inspection of screened lots. This adds even more
Accept-
50/50
to the cost of inspection.
LQL Sample Size ance AQL AOQL
Point
Number
3.11.6 Rectifying inspection, of course, cannot be used if
5.0 76 1 0.47 2.2 1.1
the inspection methods destroy the inspected articles.
105 2 0.78 2.5 1.3
3.12 AQL and LQL plans with an acceptance number of
130 3 1.1 2.8 1.5
zero are not included in this guide because their operating
10 37 1 0.97 4.5 2.3
characteristics are different from plans that have acceptance
52 2 1.6 5.1 2.6
numbers of one or more. They can provide LQL protection 65 3 2.1 5.6 3.0
78 4 2.6 6.0 3.4
against the bad inspection lot but only at the cost of rejection
of a large number of good lots. Or, if they are selected on a
15 24 1 1.5 7.0 3.5
basis of the AQL, they will allow the acceptance of a large 34 2 2.4 7.8 4.0
43 3 3.2 8.5 4.5
number of bad lots. Fig. 4 illustrates this. The OC curves of
51 4 3.9 9.2 5.0
three plans are shown. Curve number 2 is the plan shown in
Fig. 1, a sample of 55 and an acceptance number of two. Curve 20 18 1 2.0 9.3 4.7
25 2 3.4 11 5.5
number 1 is the zero acceptance number plan that has the same
31 3 4.5 12 6.2
AQL, 1.5 %, as curve number 2. The sample size is 3. But this
38 4 5.2 12 6.7
plan has an LQL of 54 % as compared to 9.7 % for curve
B697 − 88 (2021)
TABLE 3 Constant AOQL Plans
5. Selection of a Specific Plan
NOTE 1—The values listed in columns headedAQL, 50/50 Point, LQL, 5.1 After the type of plan (AQL, LQL, or AOQL) is
andAOQLarethepercentagesofnonconformingarticlesintheinspection
selected, a specific plan must be selected. Several factors need
lot.
be taken into consideration.
NOTE 2—The AOQL values are calculated for inspection lots that are
5.2 The cost of inspection should be balanced against the
very large compared to the sample. The values can be corrected for cases
cost that results from the acceptance of nonconforming prod-
where this is not true by multiplying them by:
uct. In the absence of other considerations the plan chosen
sample size
should be the one that yields the lowest total cost. Often a
lot size
smaller sample will be used when a test method is destructive
Quality Accept-
Sample 50/50 b
...