The quality of prefabricated building construction directly affects the safety and durability of prefabricated construction, with the goal of improving the quality management level of prefabricated construction, a risk evaluation model based on the fault tree-fuzzy Bayesian network method was proposed for how to effectively control the construction quality risk. Risk sources were identified and classified by combining the characteristics of prefabricated construction, the causal relationship of risk factors was modeled by using fault tree, which was mapped into a Bayesian network, fuzzy set theory and the improved similar aggregation method were introduced, and the expert opinion was transformed into the a priori probability and conditional probability of node events, and the quality risk analysis model of prefabricated construction based on fault tree-fuzzy Bayesian network was constructed. Combined with the case study: the collected data are imported into GeNIe software, parameter learning is performed to obtain the occurrence probability of each risk level. The posteriori probability is obtained by reverse inference to determine the critical factors, and the causal analysis is performed to obtain the maximum probable causal chain, and finally, the fast prediction and inference are performed by utilizing the fast prediction function of Bayesian network.