In order to clarify the influencing factors of airport service quality and the passengers’ satisfaction of airport service quality, and improve airport service quality, the evaluation model combining technique for order preference by similarity to an ideal solution(TOPSIS) and Bayesian network were used to establish an evaluation index system for large transportation airport service quality. The bidirectional inference diagnosis model was used to evaluate the airport service quality, including the forward inference of airport satisfaction and the backward sensitivity diagnosis of influencing indicators. The passenger airport service quality satisfaction and the influencing factors were studied deeply, and the feasibility of the method was verified by a large transportation airport as an example.The results show that the probability of passengers’ satisfaction with the service quality of the airport is 0.67, and the probability of general satisfaction is 0.21. The reverse diagnosis reveals that factors with high sensitivity include the baggage claim system, the convenience of integrated transportation and urban connection in and out of the airport, and the efficiency of security check services. A theoretical basis is provided for the airport to scientifically formulate measures to improve service quality.