1、 AIIM TR34 96 m 1012348 05012LO 559 m May 13,1996 ASSOCIATION FOR INFORMATION AND IMAGE MANAGEMENT INTERNATIONAL 1100 Wayne Avenue Suite 11 O0 Silver Spring, Maryland 20910 AIIM 301-587-8202 COPYRIGHT Association for Information or for a small number of lots (or rolls) of images. Other procedures ar
2、e included for a continuing series of lots or roils of images (for example, 100 rolls) that are sufficient to allow switching rules to be applied. These procedures provide the following: - an automatic protection to the consumer if a deterioration in quality is detected (by switching to tightened in
3、spection or discontinuing inspection) - an incentive to reduce inspection costs by switching to reduced inspections if consistently good quality is achieved Clause 11 contains procedures for developing sampling plans that may be used for isolated or a small number of lots or rolls of images. For a l
4、arge number of lots or roils, clause 12 contains a discussion of - the sampling procedures used for a continuing series of lots - single plans used in IS0 2859-1 (ANSI/ASQC 21.4.) - item-by-item sequential plans using IS0 8422 2 References The following standards contain provisions that, through ref
5、erence in this text, constitute parts of this technical report. At the time of publication, the editions indicated were valid. All standards are subject to revision, and parties to agreements based on this technical report are encouraged to investigate the possibility of applying the most recent edi
6、tions of the standards listed in this clause. Members of IEC and IS0 maintain registers of currently valid international standards. 2.1 Referenced international standards IS0 2859-1:1996, Sampling procedures for inspection by attributes - Part 1: Sampling plans indexed by acceptable quality level (A
7、QL) for lot-by-lot inspection. IS0 2859-2:1985, Sampling procedures for inspection by attributes - Part 2: Sampling plans indexed by Limiting Quality (Le) for isolated lot inspection. IS0 8422:1991, Sequential sampling pians for inspection by attributes. IS0 3534-1:1993, Statistics - Vocabulary and
8、symbols - Part I: Probability and general statistical tem. 1 Association for Information and Image Management international COPYRIGHT Association for Information however, clause 9 contains a partial list of attributes that may be nonconforming enough to be classed as defects. Refer to ANSUAIIM stand
9、ards and technical reports for complete listings of such image attributes. 3.1.3 image: Unit of inspection. An image is inspected to detennine its classification as conforming or not, or to count the number of nonconformities. An image wili have a number of components or attributes that may be nonco
10、nforming. Refer to clause 9 for a list of attributes. 3.1.4 inspection by attributes: Inspection whereby either the image is classified simply as conforming or nonconforming, or the number of nonconformities in the images inspected is counted, with respect to a given list of attributes. 3.15 lot: De
11、finite quantity of images, each having uniform characteristics, collected together and submitted for examination. A lot may also be a single roll or set of rolls of images received at one time, an optical disk, a magnetic tape, or a single box of microfiche, etc. 3.1.6 lot size: Number of images in
12、a lot, disk, or roli. 3.1.7 nonconforming image: Image with one or more nonconformities. Note: Nonconforming images generally will be classified by their degree of seriousness, such as: - class A. An image that contains one or more nonconformities of class A and may also contain nonconformities of c
13、lass B and/or class C. - dacc B. An image that contains one or more nonconformities of class B and may also contain nonconformities of clas C, but contains no nonconformity of dass A. 3.1.8 nonconformities per 100 images: One hundred times the number of nonconfonnities contained in the sample (one o
14、r more nonconfonnities being possible in any image) divided by the total number of images. 2 Association for information and Image Management International COPYRIGHT Association for Information h, is the intercept of the acceptance line. 3.2.7 h,: Constant that is used to determine the rejection num
15、ber; h,is the intercept of the rejection line. 32.8 multiplier of the cumulative sample size (g): Multiplier of the cumulative sample size that is used to determine the acceptance and rejection numbers; g is the slope of the acceptance and rejection lines. 33.9 nm: Average sample size. 33.10 n, : Sa
16、mple size for a single sampling plan that has essentially the same operating characteristic as the sampling plan under consideration. 3.2.11 n,: Curtailment value of the cumulative sample size. 3.2.12 p: Lot quality level in proportion nonconforming. 3.2.13 Rejection number for sequential sampling (
17、R): Value calculated from the specified parameters of the sampling plan and the cumulative sample size. The cumulative count is compared to the rejection number after each image is inspected to determine if the lot should be rejected. 3.2.14 RI: Rejection number corresponding to the curtailed value
18、of the cumulative sample size. 3 Association for Information and Image Management International COPYRIGHT Association for Information CRQ is 0.125). That is, we wish to find a sampling plan that will accept lots that are 4% or less nonconforming 95% of the time (because PR is 5%) and accept lots tha
19、t are 12.5% nonconforming only 10% of the time (because CR is 10%). We determine ratio of CRQ to PRQ. CRQPRQ = 0.125/0.04 = 3.125 Next, we look in the CRWPRQ column of table 3, Values for single sampling plans, and locate the value nearest to our calculated value. From the CRQLPRQ column, we find th
20、at 3.21 is the number nearest to 3.125. We locate the acceptance number in table 3 by looking at the AC value that corresponds to 3.21. Our acceptance number is 6. Next, we determine the sample size by dividing n x CRQ by the CRQ and n x PRQ by the PRQ (rounding up). n = n x CRQKRQ = 10.532/0.125 =
21、85 n = n x PRWPRQ = 3.286/0.04 = 83 Finally, we choose the sampling plan. We use the highest result of the two calculations for our sampling plan. in this case, we would choose the foliowing sampling plan: n = 85 Ac=6 would sample 85 images. If no more man 6 of the images inspected are nonconfoming,
22、 the lot would be accepted. 11.13 Nomograph method for selecting a plan Another way to select a sampling plan is to use the nomograph method. Figure 1, Nomograph, can be used to select a sampling plan. The following steps describe how to use the nomograph. A. Fust, we choose a PRQ and CRQ. We will c
23、hoose the same PRQ and CRQ used in 11.1.2 -a PRQ of 4% and a CRQ of 12.5% (i.e., PRQ is 0.04; CRQ is 0.125). We wish to find a sampling plan that will accept lots that are 4% nonconforming 95% of the time and accept lots that are 12.5% nonconforming only 10% of the time. 8 Association for Informatio
24、n and Image Management International COPYRIGHT Association for Information that is, they give approximately the same protection to both the producer and the consumer. Double plans have a smaller average sample size than single plans. Multiple plans have a smaller average sample size than double plan
25、s. However, single plans are the easiest to use because they always require the same sample size. Details on the use of double and multiple plans are found in the standare however, they wiil not be discussed in this technical report. Sequential plans with approximately the same protection as the sch
26、emes in IS0 2859-1 can be found in IS0 8422. Sequentialsampling using IS0 8422 is discussed in 12.2. 12.1.1 Sample size cade letters and inspection levels The sample size code letter depends on the lot size and the inspection level. The sample size code letters are found in table 7. The inspection l
27、evel is designated by the user. The purpose of inspection levels is to allow the user to maintain greater discrimination for some suppliers of conversion services and lesser discrimination for others. Inspection level II is usually used. At each inspection level, the switching rules shall operate to
28、 require normal, tightened, and reduced inspection as specified in 12.1.3. The choice of inspection level is quite separate from these three forms of inspection intensity. Thus, the inspection level that has been specified shall be kept unchanged when switching between normal, tightened, and reduced
29、 inspection. The amount of information about the quality of a lot gained from examining samples drawn from the lot depends on the absolute size of the sample, not on the .fraction of the lot sampled, providing the lot is large relative to the sample size. In spite of this, three reasons exist for va
30、rying the sample size with the lot size. - When the risk is high, it is more important to make the correct decision. - With a large lot, a sample size can be afforded that would not be economical for a smd lot. - A tnily random selection is relatively more time consuming if the sample is too smaii a
31、 proportion of the lot (e.g., a sample of 5 images from a lot of 35,000 images). 1213 Obtaining a sampiing plan The AQL and the sample size code letter should be used to obtain the sampling plan from tables 8, 9, and 10, which are respectively the normal, tightened, and reduced plans. When no sampli
32、ng plan is available for a given combination of AQL and sample size code letter, the tables direct the user to a new letter by means of arrows. The sample size to be used is given by the new sample size code letter, not the original letter. If this procedure leads to different sample sizes for diffe
33、rent classes of nonconformities, the sample size code letter corresponding to the largest sample size can be used for ail classes of nonconformities. 18 Association for Information and Image Management international COPYRIGHT Association for Information & Imaging ManagementLicensed by Information Ha
34、ndling Services AIIM TR34 96 3032348 0503235 929 ANSIIAIRM TR34-1996 Sampling Procedures for Inspection by Attributes of Images in EIM and Micrographic Systems Table 7 - Sample size code letters 19 Association for Information and Image Management International COPYRIGHT Association for Information &
35、 Imaging ManagementLicensed by Information Handling Services AIIM TR34 96 LOl12348 050123b 865 ANSIIAIIM TR34-1996 Sampling Procedures for Inspection by Attributes of Images in EIM and Micrographic Systems Table 8 - Single sampii plans for normal inspection (Master abe) 20 Association for Informatio
36、n and Image Management International COPYRIGHT Association for Information & Imaging ManagementLicensed by Information Handling Services AIIM TR34 b = 3032348 0503237 7TL ANSUAIIM TR34-1996 Sampling Procedures for Inspection by Attributes of Images in EM and Micrographic Systems SD o. o o u rable 9
37、- Single sampling plans for tightened inspection (Master table) 21 Association for Information and Image Management International COPYRIGHT Association for Information & Imaging ManagementLicensed by Information Handling Services_ AIIM TR34 96 LO32348 0503238 638 ANSVAIIM TR34- 19% Sampling Procedur
38、es for inspection by Attributes of images in EIM and Micrographic Systems 55 - on Table 10 - Sie sampling plans for reduced inspection (Master table) 22 Association for Information and Image Management International COPYRIGHT Association for Information & Imaging ManagementLicensed by Information Ha
39、ndling ServicesAIIM TR3Y 96 W 1012348 0501239 574 ANSVAUM TR34-1996 Sampling Procedures for Inspection by Attributes of Images in EIM and Micrographic Systems Inspection Type Percent nonconforming inspection Nonconformities per 100 images inspection 12.13 Normal, tightened, and reduced inspection No
40、rmal inspection shall be carried out at the start of inspection. Normal, tightened or reduced inspection shall continue unchanged on successive lots of images until the switching procedures require a change. The switching des should be applied to each class of nonconformities independently. Table 11
41、 provides the switching mies. Sample Result Determination Number of nonconforming images found is equal to or less than the acceptance number Number of nonconforming images is equal to or greater than the rejection number Number of nonconformities found is equal to or less than the acceptance number
42、 Number of nonconformities is equal to or greater than the rejection number The lot is considered acceptable. The lot shall be considered not acceptable. The lot is considered acceptable. The lot is considered not acceptable. If the cumulative number of lots not accepted in a sequence of consecutive
43、 lots on original tightened inspection reaches five, this acceptance sampling procedure shall not be resumed until action has been taken by the supplier to improve the quality of the submitted product. Tightened inspection shall then be used as if the switching from normal to tightened inspection pr
44、ocedure had been invoked. 12.1.4 Determination of acceptability The number of sample images inspected shall be equal to the sample size given by the plan. Table 12, Inspection types and determinations, provides information about percent nonconforming inspection and nonconformities per 1 O0 images in
45、spected. Switching Procedure From normal to tightened From tightened to normal I From normal to reduced From reduced to normal Table 11 - Switching des - Rula When normal inspection is being carried out, tightened inspection shall be put into operation when 2 out of 5, or fewer, consecutive lots hav
46、e been rejected on original inspection (that is, ignoring resubmitted lots). When tightened inspection is being carried out, normal inspection shall be reverted to when 5 consecutive lots have been considered acceptable on orinal insDection. Reduced inspection is an optional procedure which should b
47、e approved by all parties prior to its use. If approved, when normal inspection is being carried out, reduced inspection may be put into operation provided that all of the following conditions are satisfied - when the acceptance number is O or 1, the preceding 15 lots have been submitted to normal i
48、nspection and all have been accepted on original inspection, or - when the acceptance number is 2 or more, the preceding 10 lots have been submitted to normal inspection and all would have been accepted on original inspection if the AQL were one step tighter. When reduced inspection is being carried out, normal inspection shall be reverted to if a lot on original inspection is not accepted. 23 Association for Information and Image Management International COPYRIGHT Association for Information & Imaging ManagementLicensed by Information Handling Services
