Comparing two designs when the new design has few or no failures-is the new design better than previous one?

After a component has experienced problems in the field, a new design would be introduced to eliminate/reduce the probability of the same failure. With limited experience on the new configuration in the field/test, a question, which always needs to be answered by the reliability engineer, is that: is the new one better than the old? The new component has accumulated certain operating hours in the field/test. Few failures or (most of the time) no failures have been observed. The customer cannot wait. The problem is to determine whether the reliability of the component is improved significantly with the new design. There was no analytical method existing for the studied case. Based on the Weibayes idea, a statistical approach is developed to calculate the confidence level at which the null hypothesis, that two data sets come from the same parent population, will be rejected. A case study is provided to demonstrate/illustrate the proposed approach. It is simple and accurate. The method appears to be the best available method for the studied case