As a method of describing and evaluating the reliability of product, analysis of reliability data has been an important component of reliability engineering. The idea is put forward that the usage environment can be transformed into the test environment to fusion the test data and usage data based on the equal damage principle and then the reliability of product life can be analyzed by using the data fusion. After the data fusion, the technique states of usage product and test product are assumed to be the same and they come from the same matrix. The parameters of distribution function can be estimated by the maximum likelihood estimate method when the product life obeys some typical distributions, then the reliability of product life is analyzed by using the data fusion. The examples indicate that the sample capacity is increased after the data fusion, and the analysis accuracy of product life reliability can be improved under the same reliability and confidence level.