The prevention and control of infectious disease epidemic (IDE) is an important task for every country and region. Risk assessment is significant for the prevention and control of IDE. Fuzzy Bayesian networks (FBN) can capture complex causality and uncertainty. The study developed a novel FBN model, integrating grounded theory, interpretive structural model, and expert weight determination algorithm for the risk assessment of IDE. The algorithm is proposed by the authors for expert weighting in fuzzy environment. The proposed FBN model comprehensively takes into account the risk factors and the interaction among them, and quantifies the uncertainty of IDE risk assessment, so as to make the assessment results more reliable. Taking the epidemic situation of COVID‐19 in Wuhan as a case, the application of the proposed model is illustrated. And sensitivity analysis is performed to identify the important risk factors of IDE. Moreover, the effectiveness of the model is checked by the three‐criterion‐based quantitative validation method including variation connection, consistent effect, and cumulative limitation. Results show that the probability of the outbreak of COVID‐19 in Wuhan is as high as 82.26%, which is well‐matched with the actual situation. “Information transfer mechanism,” “coordination and cooperation among various personnel,” “population flow,” and “ability of quarantine” are key risk factors. The constructed model meets the above three criteria. The application potential and effectiveness of the developed FBN model are demonstrated. The study provides decision support for preventing and controlling IDE.