This review covers the design and analysis of essentially two types of reliability study: method comparison studies and generalizability (including inter-rater reliability) experiments. Likelihood-based methods of inference (confirmatory factor analysis and REML estimation of variance components, for example) are advocated, partly because of their ease of use but, primarily as a stimulus to the use of more ambitious designs for the investigation of the quality of measurements. The more sophisticated approaches are not intended to replace the simpler traditional methods, however, but are expected to be used to supplement them.

Design and analysis of reliability studies.

A design for single sampling attribute plan based on fuzzy sets theory

Single-sampling-attribute plans which make the producer's risk less than or equal to α (that is,  α) and the consumer's risk less than or equal to β (that is, β) are called exact sampling plans. In using exact sampling plans, however, cases often arise in which large sample size is needlessly sought. A minimax criterion which relaxes the conditions on the risks for reducing the sample size was. introduced by Fujino and Okuno. This paper presents a design procedure for relaxing the conditions on the risks by the membership functions used in fuzzy methodologies. The proposed procedure is general because it includes sampling plans based on the minimax criterion and exact sampling plans in the special. cases.

Determination of single-sampling-attribute plans based on membership functions

A method for designing single sampling plans by attributes with relaxed producer’s and consumer’s risks and quality levels is suggested. Since small deviations of the parameter (risk, quality level) from its target value are very often of no importance this soft method for designing sampling plans is more flexible than the conventional one.

Acceptance Sampling Plans by Attributes with Fuzzy Risks and Quality Levels

When acceptance sampling is used to judge the quality of an ongoing production process, the quality of successive items may exhibit statistical dependence that is not accounted for in standard acceptance-sampling plans. Computing the probabilities required to design sampling plans for general dependent processes is often complex, and sometimes intractable. This paper presents an efficient method for estimating single-sampling attribute plans for any production process model that can be simulated. Numerical illustrations are given.

Estimating acceptance-sampling plans for dependent production processes

Several methods have previously been suggested for determination of the sample size and acceptance number for attribute sampling plans while sampling without replacement. A method reported by L.J. Stephens is coupled with one by A. Hald to produce integ..

Determination of Acceptance Numbers and Sample Sizes for Attribute Sampling Plans

Replicate measurements on a single sample are commonly used to obtain a measure of repeatability If conditions that are normally allowed to vary within a laboratory are held fixed while making the replicate measurements, the resulting estimate of repea

Estimating Within-Laboratory Variability From Interlaboratory Test Data

A computer program is presented that may be used to develop specific sequential variables sampling plans. The program was motivated by the following problem situation. A producer is required to make the following statement about a lot, "With 100( p..

John L. Jaech

Papers

Determination of Acceptance Numbers and Sample Sizes for Attribute Sampling Plans

Estimating Within-Laboratory Variability From Interlaboratory Test Data

A Class of Sequential Variables Sampling Plans