This thesis is concerned with optimal designs of the step stress plants to the accelated life testing. Four step stress plans by way of observation scheme and the methods of accelerating are considered;
1. simple step stress plans for time-step and failure-step tests with censored data,
2. simple step stress plans for complete and censored data under periodic inspection,
3. multi-step stress plans for complete and censored data,
4. alternate step stress plans for complete and censored data.
For each of these plans, optimum plans are obtained for minimizing asymptotic variance of the maximum likelihood estimator of the mean life at a design stress are obtained.
Sensitivity analyses of the optimum plans to changes in the pre-estimators, extrapolation amount, and censoring time are presented.