This thesis is concerned with an economic design of life testing under type I censoring. Lindley's average amount of information is adopted as the decision criterion and the decision variables are the sample size and the type I censoring point at which the experiment terminates.
The average amounts of information about the parameters of Weibull and Pareto distribution are derived when censored data are contained in the sample.
Methods for determining the decision variables optimally are suggested. The goal is to maximize the average amount of information under the experimental budget constraints.
When the data are collected periodically, the sample size and the inter-observation time are determined in a similar manner.
Some numerical examples are also given.