Sampling Plans Site Map
 Main Areas: Best Practices:   [ Develop Sampling Plans. ] [ Applications Area ] [ Study Area ] Study Area:    [ Latest Changes: 9/21/10 ] [ Tutorial ] [ Sequential Sampling ] [ Glossary of Terms ] [ Word Search ]
 Glossary of Terms visitors=
 Term Definition Source of Definition Acceptance sampling Sampling inspection in which decisions are made to accept or reject product.; also the science that deals with procedures by which decisions decisions to accept or reject are based on the results of the inspection of samples. Glossary and Tables for Statistical Quality Control - ASQC Comment: Typical uses of acceptance sampling in manufacturing include making acceptance decisions about incoming raw materials lots, in-process sublots, and finished product lots. Acceptance sampling software Acceptance sampling plan A specific plan that states the sample size or sizes to be used and the associated acceptance and rejection criteria. Glossary and Tables for Statistical Quality Control - ASQC Comment: Most acceptance sampling plans in use are either attributes plans and variables plans. Acceptance sampling software AOQ curve Acronym for Average Outgoing Quality. Useful to evaluate sampling plan applications where rejected lots are rectified by replacing or reworking defective items. The AOQ curve is the average quality of outgoing product as a function of the incoming quality. Examples: AOQ curve for defectives AOQ curve for defect counts AOQ curve for variables fraction non-conforming Comment: AOQ is the quality of an average outgoing lot. Therefore, you should expect half of the lots to be worse than AOQ. The AOQ calculation does not consider that the incoming quality usually varies. AOQL Acronym for Average Outgoing Quality Limit. The maximum AOQ over all possible values of incoming product quality, for a given acceptance sampling plan. Glossary and Tables for Statistical Quality Control - ASQC Comment: Maximum of the AOQ curve. See AOQ. AQL Acronym for Acceptable Quality Level. As used in the development of two-point acceptance sampling plans, the values of AQL and alpha jointly define the producers point of the operating characteristic curve. Graph showing the producers point. If the value of a quality characteristic of a particular lot is exactly equal to the AQL of it's acceptance sampling plan, the probability that the plan will accept the lot is (Pa=1-alpha). Example of specifying AQL For a discussion of common confusions about AQL, see AQL Primer. "Glossary and Tables for Statistical Quality Control" - ASQC. Comment: This definition of AQL is statistically exact and appropriate for use with two-point sampling plans, as supported by the software of H & H Servicco Corp. On the other hand, a more vague definition of AQL is typically used by documents that support one-point sampling plans. The most common of such documents are: Mil-Std-105, Mil-Std-414, ANSI/ASQC Z1.4, ANSI/ASQC Z1.9 These one-point sampling plans do not make use of the consumer's point - they do not address the issue of accepting low-quality lots. They are particularly vulnerable to this for small sample sizes. ARL curve Acronym for Average Run Length. ARL is the average number of accepted lots between rejections. The ARL curve is a plot of ARL as a function of lot quality level. Examples: ARL curve for defectives ARL curve for defect counts ARL curve for variables fraction non-conforming Comment: Use the ARL curve to assess the impact of an acceptance sampling plan on smoothness operations. ASN curve Acronym for Average Sample Number. ASN is the average number of sample units inspected per lot in reaching decisions to accept or reject. The ASN curve is a plot of ASN versus lot quality. Examples: ASN curve for defectives ASN curve for defect counts ASN curve for variables fraction non-conforming Glossary and Tables for Statistical Quality Control - ASQC Comment: Use ASN curves to evaluate sequential sampling plans to anticipate the amount of inspection that each plan will require. Audit sampling Sampling in which the goal is to estimate the value a quality characteristic but not provide a firm decision rule. The sample size n is chosen to provide a desired margin of error of the estimate. Comment: Many audit sampling situations involve more than one category, each having a  different sample size. The categories having the  smaller sample sizes will have estimates with larger margins of error. Conversely, the categories having the larger sample sizes will have estimates with smaller margins of error. Audit sampling software Margin of Error The sampling error of the estimated statistic. The margin of error is usually expressed as half the the width of a confidence interval. Examples: p' plus or minus ME Xbar plus or minus ME where ME=Margin of ErrorAudit Sample Planner develops sampling plans based on the margin of error. OC curve Acronym for Operating Characteristic Curve. A curve showing, for a given sampling plan, the probability of accepting a lot, as a function of the lot quality level. It is knowledge (by the person who designs or selects the plan) of the oc curve that makes an acceptance sampling plan statistically valid. Examples: OC curve for defectives OC curve for defect counts OC curve for variables fraction non-conforming OC curve for MTBF OC curve for reliability Glossary and Tables for Statistical Quality Control - ASQC Comment: The OC curve is fundamental to the two-point method of developing acceptance sampling plans. Two-point method RQL Acronym for Rejectable Quality Level. As used in the development of two-point acceptance sampling plans, the values of RQL and beta jointly define the consumers point of the operating characteristic curve. Graph showing the consumers point. If the value of a quality characteristic of a particular lot is exactly equal to the RQL of it's acceptance sampling plan, the probability that the plan will accept the lot is (Pa=beta). Example of specifying RQL The literature contains other terms having the same meaning as RQL. These are LQ (limiting quality) and LTPD (lot tolerance percent defective). Glossary and Tables for Statistical Quality Control - ASQC. See LTPD. Sequential Analysis, Sequential Sampling The technique by which we build up our sample one item at a time, and after inspecting each item, ask ourselves: "Can we be sure enough to accept or reject this batch on the information so far collected?" Its value is in enabling reliable conclusions to be wrung from a minimum of data. This was deemed sufficient to require that it be classified "Restricted " within the meaning of the Espionage Act during the war of 1939-45. Facts from Figures - M. J. Morony Comment: Used for lot-by-lot acceptance plans for applications needing the minimum possible average sample numbers.    Example of attribute sequential -- defectives    Example of attribute sequential -- defects    Example of variables sequential -- LISL Two general types of sequential plans are used for acceptance sampling: SPR and TSS. The SPR type is also known as Columbia sequential sampling because the methodology was developed at Columbia University. Statistically Valid An acceptance sampling plan is statistically valid when the person who designs or selects it knows the  probabilities that the plan will accept lots that were manufactured to various quality levels. These probabilities are shown by the operating charactistic curve (oc curve)