Barrel is an important part of artillery weapons, its performance state determines the service life of gun. Wiener process with random effect was used for residual lifetime prediction for gun barrel in this paper, the model parameters were estimated by maximum likelihood method, and then updated through Bayesian method in order to improve the estimation accuracy. Finally, a certain type of gun was used as an example for residual lifetime prediction, the probability density function and reliability function curves of the residual lifetime under different projectiles conditions were obtained, and the prediction accuracy was improved gradually with the increase of projectiles. The application example shows the feasibility of the present approach in this paper, and it has some reference value for gun barrel’s residual lifetime prediction.