Evacuation systems in buildings are frequently assessed to improve emergency response processes. This paper proposes a method to evaluate the performance of different evacuation modes, and determine a rational mode for large railway stations. We developed a simulation for the evaluation of fire safety in large buildings based on an analytic hierarchy process (AHP) method. This approach includes AHP-based exploration and simulation-based refinement. We considered a typical railway station for validation, conducted a field survey to collect the data, and calculated the influencing factors based on expert opinion. The influencing factors were further processed based on the principles of a hierarchical model. The relative weights of the influencing factors were calculated through a series of pairwise comparisons using the AHP. Further, we applied factor refinement based on the evacuation simulations to determine the degree and status of influence of each factor. The influence of external factors was generally stronger than that of the internal factors. Among them, the building component characteristics and people’s physiological capabilities were the core of the evacuation assessment in large railway stations. Additionally, the exit width, seat layout, visibility, speed, and reaction capabilities were crucial to the evacuation process. The proposed method is practical as it demands limited computations to provide useful information, such as a priority ranking of each influencing factor, for the evaluation process.