ON RELIABILITY ESTIMATION IN THE WEIBULL CASE Hidekazu Tanaka* The•@ asymptotic mean integrated squared errors of the naive estimator, the maximum likelihood estimator, the uniformly minimum variance unbiased estimator, and the invariantly optimal estimator of a reliability function are calculated in the Weibull distribution case. These estimators are also compared using the concept of the asymptotic deficiency. Further, it is shown that, after bias-adjustment, the later three estimators have the same asymptotic mean inte grated squared error up to the order o(n-2), where n is a size of sample.