The life of oil-filled power transformers depends on the long-term life of the paper - oil insulation system throughout continuous operation. The study of reliability assessment of oil–paper insulation is used to evaluate the reliability level of power transformers accurately. In this paper, the life of oil–paper insulation is proven to obey the Weibull probability distribution under ten different parameters, set by the Weibull distributed reliability; the degradation mechanisms of the different parameters are highly consistent. The breakdown voltage, tensile strength, degree of polymerization of paper, furfural, oil spectral response and five other characteristic parameters of oil were chosen to reflect the reliability of paper-oil insulation. In this work, the reliability assessment is carried out using entropy weight method. An experiment is conducted to validate the proposed theoretical method. Thus, the present work puts forward a simple and effective method for the integrated oil–paper insulation reliability assessment of power transformers under different parameters at different life stages.