1、BRITISH STANDARD BS 6000-1:2005 Guide to the selection and usage of acceptance sampling systems for inspection of discrete items in lots Part 1: General guide to acceptance sampling ICS 03.120.30 BS 6000-1:2005 This British Standard was published under the authority of the Standards Policy and Strat
2、egy Committee on 25 July 2005 BSI 25 July 2005 First published as BS 6000 March 1972 Second edition May 1996 BS 6000-1 first edition published 25 July 2005 The following BSI references relate to the work on this British Standard: Committee reference SS/5 Draft for development 03/102301 DC ISBN 0 580
3、 46484 9 Committees responsible for this British Standard The preparation of this British Standard was entrusted to Technical Committee SS/5, Acceptance sampling schemes, upon which the following bodies were represented: Association for Road Traffic and Safety Management British Measurement and Test
4、ing Association City University Clay Pipe Development Association Ltd. Institute of Metal Finishing Institute of Quality Assurance Ministry of Defence UK Defence Standardization Co-opted members Amendments issued since publication Amd. No. Date CommentsBS 6000-1:2005 BSI 25 July 2005 i Contents Page
5、 Committees responsible Inside front cover Foreword ii Introduction 1 1S c o p e 1 2 Terms and definitions 2 3 Abuses and uses of acceptance sampling 2 4 Acceptance sampling plans, schemes and systems 5 5 Practical and economic advantages of using British Standard sampling plans 5 6 Attributes versu
6、s variables 6 7 Further considerations influencing a selection 7 8 Making a comparison of the methods for sampling inspection 20 9 Other methods sometimes adopted in practice 26 10 Relevance of market and production conditions 27 11 The final selection realism 29 Annex A (informative) Example of a s
7、imple model for profit maximization under destructive inspection by attributes 30 Bibliography 33 Figure 1 Operating characteristic curve defined by producers risk (PR) and consumers risk (CR) 14 Figure 2 Operating characteristic curve defined by acceptance quality limit (AQL) and limiting quality (
8、LQ) 15 Figure 3 Ideal operating characteristic curve 21 Figure 4 Comparison of operating characteristic curves for single, double, multiple and sequential sampling plans (code letter L, AQL 0.65 %) 24 Figure 5 Comparison of average sample sizes for single, double, multiple and sequential sampling (f
9、or the acceptance plans of Table 3) 25 Figure 6 Comparison of the probabilities of the single sampling sample size being exceeded, for the double, multiple and sequential sampling plans of Table 3 26 Table 1 Optimal sampling plans for Example 1 3 Table 2 Comparison of sample sizes in inspection by a
10、ttributes and by variables 7 Table 3 Equivalent sampling plans by attributes for code letter L, AQL 0.65 % 23 Table A.1 Optimal sampling plans for example of A.8 32BS 6000-1:2005 ii BSI 25 July 2005 Foreword This part of BS 6000 has been prepared by Technical Committee SS/5. Together with BS 6000-2
11、and BS 6000-3 it supersedes BS 6000:1996, which is withdrawn. BS 6000:2005 Guide to the selection and usage of acceptance sampling systems for inspection of discrete items in lots is in three parts: Part 1: General guide to acceptance sampling Part 2: Guide to sampling by attributes Part 3: Guide to
12、 sampling by variables The primary purpose of the BS 6000 series is to give guidance on the selection of an acceptance sampling system, scheme or plan. It does this principally by reviewing the available systems specified by various British Standards and showing ways in which these can be compared i
13、n order to assess their suitability for an intended application. The guide also indicates how prior knowledge of the manufacturing or service delivery process and quality performance might influence the choice of sampling system, scheme or plan, and likewise how the particular needs of the customer
14、affect selection. Some specific circumstances encountered in practice are described and the method of choosing a plan is explained. Some checklists or pointers and tables are provided to assist users in selecting an appropriate system, scheme or plan for their purposes. Charts are included to illust
15、rate the procedures to be followed in the selection process. This publication does not purport to include all the necessary provisions of a contract. Users are responsible for its correct application. Compliance with a British Standard does not of itself confer immunity from legal obligations. Summa
16、ry of pages This document comprises a front cover, an inside front cover, pages i and ii, pages 1 to 33 and a back cover. The BSI copyright notice displayed in this document indicates when the document was last issued.BS 6000-1:2005 BSI 25 July 2005 1 Introduction This British Standard gives guidanc
17、e on the selection of an appropriate acceptance sampling scheme for the inspection of discrete items submitted in lots from the schemes described in various national and international standards. There are many situations where products (materials, parts, components, assemblies and systems) are trans
18、ferred from one organization to another, where the organizations may be different companies or parts of a single company or even different shops within a plant. In these situations both the supplier and the customer may use acceptance sampling procedures to satisfy themselves that the product is of
19、acceptable quality. The supplier will be seeking to maintain a reputation for good quality and to reduce the likelihood of claims under warranty, but without incurring unnecessary production and supply costs. On the other hand, the customer will require adequate evidence, at minimum cost to himself,
20、 that the product he receives conforms to specification. Compared with, say, 100 % inspection, suitable sampling methods will often be beneficial in achieving these aims. Sometimes acceptance sampling methods are the only practical procedure, especially when the tests for conformance are destructive
21、. Several types of sampling systems, schemes and plans are available for these purposes. They are presented in a number of British and ISO Standards that explain how they are to be used. However, it is often difficult to decide on the most appropriate procedure for use in a particular situation. The
22、 purpose of this British Standard is to assist in that decision. The choice of sampling system, scheme or plan depends on a number of conditions and on the prevailing circumstances. In any supply situation, the first essential is that the supplier and the customer understand, and have agreed, the re
23、quirements and the basis for release and acceptance of the product, including any acceptance sampling methods to be used. Lots that are not acceptable cause difficulties for both supplier and customer. The supplier incurs additional costs in rework, scrap, increased inspection, damage to reputation
24、and possibly loss of sales. Delays in delivery and re-inspection costs are a burden to the customer. For these reasons, it is usually considered essential for the supplier to provide lots that have a very high probability of being accepted, i.e. 95 % or more. The supplier has to ensure that quality
25、control of the production or delivery process provides lots of a quality sufficient to meet this objective. A basic principle of some acceptance sampling inspection schemes is to promote the production of lots of acceptable quality. The primary purpose of these schemes is not to discriminate between
26、 acceptable and non-acceptable lots, i.e. to sort, but to keep production under control to yield an acceptable process average quality. Although all acceptance sampling plans are discriminatory to some degree, the process average quality (expressed in terms of percent nonconforming or number of nonc
27、onformities) should not be greater than half the acceptance quality limit in order to ensure a very high probability of acceptance. 1 Scope This part of BS 6000 gives general guidance on the selection of an acceptance sampling system, scheme or plan. It does this principally in the context of Britis
28、h Standards that either already exist or are presently under development. (For more detailed information about specific acceptance sampling systems, see BS 6000-2 for sampling by attributes or BS 6000-3 for sampling by variables.) The guidance in this British Standard is confined to acceptance sampl
29、ing of products that are supplied in lots and that can be classified as consisting of discrete items (i.e. discrete articles of product). It is assumed that each item in a lot can be identified and segregated from the other items in the lot and has an equal chance of being included in the sample. Ea
30、ch item of product is countable and has specific characteristics that are measurable or classifiable as being conforming or nonconforming (to a given product specification). Standards on acceptance sampling are typically generic, as a result of which they can be applied to a wide variety of inspecti
31、on situations. These include, but are not limited to, the following: a) end items, such as complete products or sub-assemblies; b) components and raw materials; c) services; d) materials in process;BS 6000-1:2005 2 BSI 25 July 2005 e) supplies in storage; f) maintenance operations; g) data or record
32、s; and h) administrative procedures. Although this British Standard is written principally in terms of manufacture and production, this should be interpreted liberally, as it is applicable to the selection of sampling systems, schemes and plans for all types of product and processes as defined in BS
33、 EN ISO 9000. 2 Terms and definitions For the purposes of this British Standard, the terms and definitions given in BS ISO 3534-1, BS ISO 3534-2 and BS EN ISO 9000 apply. 3 Abuses and uses of acceptance sampling 3.1 Abuses of acceptance sampling Acceptance sampling has become unpopular since the ear
34、ly 1980s. Some of the reasons for this (although certainly not all) are well founded, so it is important to be able to distinguish those situations where acceptance sampling should not be used from those where it may be appropriate. The chief arguments used against the use of acceptance sampling are
35、 as follows. a) When quality is generally very high, the sample sizes needed to detect a slip in quality are uneconomically large. b) Quality cannot be inspected into a product. c) It is far better to establish a robust design and to implement comprehensive process controls than to try to find and e
36、liminate nonconforming items after manufacture. d) Most acceptance sampling standards are indexed in terms of acceptable quality level (AQL). Once an AQL has been established and quality has been brought sufficiently below the AQL to achieve high probabilities of lot acceptance, there is no incentiv
37、e for the producer to try continuously to improve quality. e) Quoting an AQL is tantamount to granting a licence to produce defects. f) The only acceptable quality level is zero defects. These arguments are examined in turn in the following sub-clauses. 3.2 EXAMPLE 1 The following simplified example
38、, supplied by Baillie 1, demonstrates how the optimum sampling plan can vary according to the quality level against which it is desired to guard. A certain item is produced in lots of size 10 000, with a unit production cost of 10.00. The selling price per item is a in accepted lots and at a discoun
39、ted price of 0.50 in lots non-accepted by the acceptance procedure. Testing is destructive, and the cost of testing each item is 1.00. The downstream cost (e.g. warranty cost plus loss of goodwill) of a nonconforming item in an accepted lot is 10 000, but zero in non-accepted lots sold at a discount
40、. Historical data indicate that the process fraction nonconforming is p for 99 % of lots, but that it unaccountably and randomly slips to 100p for 1 % of the lots. A single sampling plan by attributes is to be used, i.e. a random sample of size n is to be selected from each lot, and the lot is to be
41、 considered acceptable if the sample contains no more than Ac nonconforming items. What is the optimal sampling plan, i.e. the plan that maximizes the profit per item sold? Mathematical details are provided in Annex A for information. Table 1 shows the optimal sampling plan for a range of values of
42、the process quality level p. The results are instructive. BS 6000-1:2005 BSI 25 July 2005 3 Table 1 Optimal sampling plans for Example 1 Not surprisingly, it is found that improvements in the quality level allow the selling price to be decreased while at the same time increasing the profit per item
43、sold. At first, improvements in quality levels necessitate larger sample sizes in order to be able to provide the necessary discrimination between the two quality levels. As quality levels improve, the optimal acceptance number Ac reduces and there comes a point when the sample size that is required
44、 also begins to reduce until, eventually, it becomes uneconomical to sample at all. This final state is called “indirect inspection”, as the inspection has effectively been transferred from the producer to the consumer; nonconforming items are so rare that it is more economical not to sample and ins
45、pect but to reimburse consumers on the infrequent occasions that they invoke the warranty. Thus 3.1a) is seen to be misleading for, when quality levels reach a sufficiently high level, acceptance sampling simply becomes an unnecessary overhead rather than requiring uneconomically large sample sizes.
46、 3.3 Inspecting quality into a product Inspection makes little difference to the outgoing quality if the incoming quality is more or less constant, unless the sample size is a large proportion of the lot size, in which case the inspection process is a large overhead. Either way, it is not a particul
47、arly sensible approach to improving quality levels. 3.4 Design and control The advantages of establishing a robust design and a comprehensive process control system are many. The robust design places the least possible demands on the manufacturing process and the process control system tends to prev
48、ent process parameters from straying too far from their target values, so process variation and waste is kept low and output quality is kept high. Moreover, the design and the control system of the production process can be reviewed and improved in the light of experience to provide continual qualit
49、y improvement. 3.5 AQLs The initials AQL used to stand for Acceptable Quality Level, although in reality the AQL is simply an index to a sampling plan. Standards tried to make this clear by explaining that the level was acceptable for the purposes of acceptance sampling (rather than in an absolute sense). Indeed, lot quality levels typically have to be better than half the AQL to have a very high chance of being accepted. During the late 20th century, many
