Designing and construction of tightened-normal-tightened variables sampling scheme
The tightened-normal-tightened (TNT) attributes sampling scheme was devised by Calvin (1977). In this paper, a TNT Scheme with variables sampling plan as the reference plan, designated as TNTVSS (n σ ; k T , k N ) is introduced, where n σ is the sample size under the reference
plan, and k T and k N are the acceptance constants corresponding to tightened and normal plans respectively. The behaviour of OC curves of the TNTVSS (n σ ; k T , k N ) is studied. The efficiency of TNTVSS (n σ ; k T , k N ) with respect to smaller sample sizes has been established
over the attributes scheme. The TNTVSS is matched with the TNT (n; c N , c T ) of Vijayaraghavan and Soundararajan (1996), for the specified points on the OC curves, namely (p 1 , α) and (p 2 , β) and it is shown that the sample size of the variables scheme is much smaller than that
of the attributes scheme. The TNT scheme with an unknown σ variables plan as the reference plan is also introduced along with the procedure of selection of the parameters. The method of designing the scheme based on the given AQL (Acceptable Quality level), α (producer's risk), LQL
(Limiting Quality Level) and β (consumer's risk) is indicated. Among the class of TNTVSS which exists, for a given (p 1 ,α) and (p 2 , β), a scheme, which will have a more steeper OC curve than that of any other scheme, is identified and given.
Keywords: AQL; LQL; Variables plan; consumer's risk; producer's risk; switching rules; tightened-normal-tightened plan
Document Type: Research Article
Affiliations: 1: Sri Krishna Arts & Science College, Coimbatore, India 2: Department of Statistics, PSG College of Arts and Science, Coimbatore, India
Publication date: 01 January 2006
- Access Key
- Free content
- Partial Free content
- New content
- Open access content
- Partial Open access content
- Subscribed content
- Partial Subscribed content
- Free trial content