C
, but only m of the n components fail within
C
, then the remaining n � m are censored observations as it is only known that their failure time is larger than
C
, but their exact failure times are unknown.
In reliability analysis, it is assumed that censoring time
C
, is statistically inde- pendent of the failure time
T
, i.e., non-informative censoring. The observed value is the minimum of
C
and
T
. It means that censored