Flight procedure designers refer to the International Civil Aviation Organization（ICAO） Regulation 8168 from the perspective of flight safety and a few economic indicators. However， flight procedure designers rarely considers the needs of different users involved. Furthermore， the evaluation system of the flight procedure is mainly based on the experience of experts， which leads to a lack of the analysis of objective factors. In this paper， a novel evaluation system of flight procedure that combines neural networks with the entropy method is proposed to satisfy the four users involved in the flight procedure. Firstly， the demands of the flight procedure users， such as pilots， civil aviation controllers， military air traffic controllers， and airports and urban planning departments are analyzed. Secondly， to satisfy the needs of these mentioned users， a three-level evaluation index system for flight procedures is established. Thirdly， the subjective weights obtained by the neural network computation and the objective weights calculated by the entropy method are fused to achieve a comprehensive subjective and objective weight evaluation system by using the game theory framework. Finally， based on the departure procedure of a military and civil aviation airport in western China， the evaluation system is verified. In addition， the suggestions are proposed for the improvement of the procedure. The flight procedure that integrates the needs of multiple users will further improve the efficiency of air flights under the conditions of limited airspace resources， and reduce delays.