1960
DOI: 10.1093/biomet/47.3-4.363
|View full text |Cite
|
Sign up to set email alerts
|

Sampling inspection of continuous processes with no autocorrelation between successive results

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

3
21
0
7

Year Published

1999
1999
2009
2009

Publication Types

Select...
5
4

Relationship

0
9

Authors

Journals

citations
Cited by 178 publications
(31 citation statements)
references
References 8 publications
3
21
0
7
Order By: Relevance
“…11 Specifically, when monthly baseline rates of 0.22, 0.39, and 8 were simulated, mean times to a false alarm, 41.1, 41.5, and 43.3 years, respectively, were comparable to the 41.7 years reported previously (Table 2). However, additional simulations of baseline rates and ''optimal'' k and h combinations, recommended but not reported, 11 produced a wide range (357-978 months) of mean time to first false alarm compared with the expected time of 500 months. These findings emphasize the need to closely evaluate the performance of the cusum technique prior to the adoption of ''optimal'' parameters.…”
Section: Resultssupporting
confidence: 73%
See 1 more Smart Citation
“…11 Specifically, when monthly baseline rates of 0.22, 0.39, and 8 were simulated, mean times to a false alarm, 41.1, 41.5, and 43.3 years, respectively, were comparable to the 41.7 years reported previously (Table 2). However, additional simulations of baseline rates and ''optimal'' k and h combinations, recommended but not reported, 11 produced a wide range (357-978 months) of mean time to first false alarm compared with the expected time of 500 months. These findings emphasize the need to closely evaluate the performance of the cusum technique prior to the adoption of ''optimal'' parameters.…”
Section: Resultssupporting
confidence: 73%
“…''Optimal'' parameter values for the cusum technique were initially selected to have an associated mean time to a false alarm of 500 months (41.7 years). 11 Alternatively, 500 hours, days, or weeks could also have been used. In the current study, simulations were performed to evaluate a 300-year (3,600-month) time period.…”
Section: Resultsmentioning
confidence: 99%
“…1)/2, which after our standardization reduces to IJ./2. The chain of references (if any) usually ends with Ewan and Kemp (1960), where it is concluded from a nomogram that this value seems to be good. The likelihood ratio method for C={'t=i} gives alarm for…”
Section: 5cusummentioning
confidence: 99%
“…Além do mais, como já assinalamos anteriormente, também é passível de crítica a construção de intervalos de confiança baseados no pressuposto da validade da teoria Normal de distribuição de probabilidades, usando um conjunto de observações oriundas da coleta de dados de vigilância como amostra aleatória. Shore & Quade (1989) fazem um estudo comparativo entre o esquema de memória curta e o procedimento Cusum, partindo de dados tabelados por Ewan & Kemp (1960) para o ARL de uma variável Poisson. Concluem que o procedimento Cusum é mais eficiente quando a amplitude da memória é pequena, mas, mesmo em condições assintóticas para a amplitude da memória, o procedimento Cusum parece ser mais eficiente.…”
Section: O Procedimento Cusumunclassified