Abstract: Prediction of box-office revenue was studied, and a box-office earnings forecast method based on Bass random diffusion model was proposed. By dividing the behavior of the people to watch the film into deciding to watch the film and taking action, the volatility characteristics of the film diffusion was imitated using the Gamma process. The results show that the Gamma process is more suitable for the diffusion process of cultural creative assets. The case study proves that the established model improves the prediction precision compared with the traditional box office earnings forecast.