Acceptance Sampling by Variables under Measurement Uncertainty

Melgaard, Henrik; Thyregod, Poul

doi:10.1007/978-3-642-57590-7_4

Cited by 8 publications

(11 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When apparent measurements are imprecise, the condition ¤ 1 holds. Assuming that Y and Z are normally distributed, Mei et al (1975), Owen and Chou (1983), Basnet and Case (1992), Fang and Zhang (1995), Wilrich (2000), and Melgaard and Thyregod (2001) studied the single sampling variables plans. Mei et al (1975), Owen and Chou (1983), and Wilrich (2000) suggested an increase in sample size to mitigate the effect of inaccuracy in measurements.…”

Section: Measurement Error Correction For Variables Inspectionmentioning

confidence: 99%

Fractional Acceptance Numbers for Lot Quality Assurance

Govindaraju

Jones

2015

Frontiers in Statistical Quality Control 11

View full text Add to dashboard Cite

Fractional acceptance numbers such as one-quarter fraction nonconforming are used in the acceptance sampling literature. The concept of fractional acceptance number is particularly useful for short-run food manufacturing processes involving a measurable quality characteristic such as the percentage sugar or fat content. Attribute method of inspection is desirable for small sample sizes because of the difficulty in identifying the underlying distribution. Analytical testing of fat content, etc. also involves considerable measurement uncertainty, often up to half of the observed variation. However the distribution of the measurement errors can be fairly well ascertained using past calibration studies. An observed measurement X is classified with certainty as conforming or not for given specification limits only when there are no measurement errors. Measurement error uncertainty results only in an estimated probability of conformance of a unit. The probability of nonconformance of an individual unit based on the error-prone measurement is defined as the 'fractional' nonconforming unit. A new fractional acceptance number sampling plan, which is a mix of attribute and variables methods, is introduced. The operating characteristics of the proposed plan are evaluated using common error distributions and the incomplete beta function. The fractional acceptance numbers also extend to broken acceptance numbers such as 1.6.

show abstract

Section: Measurement Error Correction For Variables Inspectionmentioning

confidence: 99%

Fractional Acceptance Numbers for Lot Quality Assurance

Govindaraju

Jones

2015

Frontiers in Statistical Quality Control 11

View full text Add to dashboard Cite

show abstract

“…In the last edition of the standard ISO 3951-2 [8] an informative annex has been introduced for the purpose of taking into accout a non-negligible random measurement error. The case of a non-negligible residual systematic measurement error is analysed in [11] (a reference which appears among those listed in the bibliography of [8]). It is worth comparing the frequentist result in [11] with the Bayesian one obtained here and reported in table 1, both valid for zero mean, normal systematic measurement error.…”

Section: Frequentist Approachmentioning

confidence: 99%

“…The case of a non-negligible residual systematic measurement error is analysed in [11] (a reference which appears among those listed in the bibliography of [8]). It is worth comparing the frequentist result in [11] with the Bayesian one obtained here and reported in table 1, both valid for zero mean, normal systematic measurement error. From section 3 of [11] we have 24) is substituted in (33) when assuming normal or rectangular residual systematic error, respectively.…”

Section: Frequentist Approachmentioning

confidence: 99%

Bayesian conformity assessment in presence of systematic measurement errors

Carobbi

Pennecchi

2016

Metrologia

View full text Add to dashboard Cite

Conformity assessment of the distribution of the values of a quantity is investigated by using a Bayesian approach. The effect of systematic, non-negligible measurement errors is taken into account. The analysis is general, in the sense that the probability distribution of the quantity can be of any kind, that is even different from the ubiquitous normal distribution, and the measurement model function, linking the measurand with the observable and non-observable influence quantities, can be nonlinear. Further, any joint probability density function can be used to model the available knowledge about the systematic errors. It is demonstrated that the result of the Bayesian analysis here developed reduces to the standard result (obtained through a frequentistic approach) when the systematic measurement errors are negligible. A consolidated frequentistic extension of such standard result, aimed at including the effect of a systematic measurement error, is directly compared with the Bayesian result, whose superiority is demonstrated. Application of the results here obtained to the derivation of the operating characteristic curves used for sampling plans for inspection by variables is also introduced.

show abstract

“…Measurement errors in the design of single sampling variables plans is considered by the authors in , and . The apparent measurements ( X ′ ) subject to errors are assumed to follow N ( ϵ , τ ) instead of the standardized N (0,1) distribution for the true measurements X .…”

Section: Risk Analysismentioning

confidence: 99%

“…The impact of measurement and model uncertainties on the combined attributes-variables plan is discussed next. Measurement errors in the design of single sampling variables plans is considered by the authors in [32][33][34][35][36][37], and [38]. The apparent measurements (X ′ ) subject to errors are assumed to follow N( , ) instead of the standardized N(0, 1) distribution for the true measurements X.…”

Section: Risk Analysismentioning

confidence: 99%

A combined attributes–variables plan

Govindaraju

Kissling

2014

Appl Stoch Models Bus & Ind

View full text Add to dashboard Cite

The design of single sampling plans in which the lot acceptance decision is based on both variables and attribute measurement of quality is discussed. A new plan, called the combined attributes-variables plan, is proposed incorporating an acceptance number to the regular variables plan for consumer protection. A design approach for the new plan is also developed for food manufacturing applications in which the sample size cannot be predetermined because of short production lengths and other analytical testing issues.

show abstract

Acceptance Sampling by Variables under Measurement Uncertainty

Cited by 8 publications

References 6 publications

Fractional Acceptance Numbers for Lot Quality Assurance

Fractional Acceptance Numbers for Lot Quality Assurance

Bayesian conformity assessment in presence of systematic measurement errors

A combined attributes–variables plan

Contact Info

Product

Resources

About