Accidents frequently occur in the operation and maintenance of urban underground utility tunnels. To quantify the operation and maintenance risks of utility tunnels and analyze the key risk factors, a risk assessment method for the operation and maintenance of urban underground utility tunnels based on fault tree and fuzzy Bayesian network was proposed. A fault tree was constructed based on comprehensive consideration of the risk factors of tunnel, and it was mapped to a Bayesian network. Then, the risk factor status was classified based on ALARP criteria, fuzzy number and fuzzy subset were used to calculate the probability of occurrence of bottom events, Maximum Likelihood Estimation was used to solve the conditional probability of intermediate events, and a Bayesian network model was constructed for operation and maintenance risk of utility tunnels to assess the operation and maintenance risk of urban underground utility tunnels accurately. The results show that the probability for the assessed risk level of utility tunnels’ operation and maintenance process is High is 28%, which was close to the risk threshold of 30%. Therefore, it is necessary to control the risks of the utility tunnels timely. This method can scientifically and reasonably evaluate the risk level and determine the key factors, and can be used as a decision-making tool for the safety assurance and management of the utility tunnels’ operation and maintenance.