Circuit packs were manufactured to the same design specification, but by two different vendors. The trial ran for 10,000 hours to determine which vendor's circuit packs were more reliable. The 4993 circuit packs from Vendor 1 came straight from production. The 4993 circuit packs from Vendor 2 had already seen 1000 hours of burn-in testing at the manufacturing plant under operating conditions similar to those in the field trial. The circuit packs manufactured by Vendor 2 were sold at a higher price because field reliability was supposed to have been improved by the burn-in screening of circuit packs containing defective components. Failures during the first 1000 hours of burn-in were not recorded. This is the reason for the unknown entries in the table and for having information out to 11,000 hours for Vendor 2. The data are for the first failure in a position. Information on circuit packs replaced after initial failure in a position was not part of the study.
A data.frame
with 27 rows and 7 variables:
[, 1] | lower | Start of an observation interval | Numeric |
[, 2] | upper | End of an observation interval | Numeric |
[, 3] | event | Event observed in the interval (failure/right-censored/left-censored/interval-censored) | Categoric |
[, 4] | truntype | Truncation type (left-truncated/right-truncated/interval-truncated) | Categoric |
[, 5] | truntime | Truncation time | Numeric |
[, 6] | count | Number of events observed in the interval | Numeric |
[, 7] | vendor | Producing vendor | Categoric |
Meeker, W. Q. and Escobar, L. A. (1998), Statistical Methods for Reliability Data, New York, NY; Wiley-Interscience
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.