Considering the complexity of mixed pedestrians (e.g. children, disabled people, etc.) in a hospital registration hall, once an emergency occurs in a hospital registration hall, it is important to ensure that all pedestrians can evacuate from the hall efficiently and safely. In this paper, based on the motion features of pedestrians in the hospital registration hall, we first classify the pedestrians, then propose an improved CA (cellular automaton) model to describe the motion of each type of pedestrians, and finally define some indexes to further testify the proposed model. Simulation results show that different types of pedestrians have different motion behaviors during the non-emergency evacuation in the hospital registration hall, and that these motion behaviors have different impacts on the evacuation process (especially the efficiency). The results can help the administrators understand the non-emergency evacuation process of heterogeneous pedestrians and design some reasonable evacuation strategies for unexpected emergencies in the hospital registration hall.

考虑到医院挂号大厅混合行人（如儿童、残疾人等）的复杂性，一旦医院挂号大厅发生紧急情况，如何保证所有行人能够高效、安全地撤离大厅就显得尤为重要。本文基于医院挂号大厅行人的运动特征，首先对行人进行分类，然后提出一种改进的CA（元胞自动机）模型来描述每一类行人的运动，最后定义一些指标以进一步证明所提出的模型。仿真结果表明，在医院挂号大厅非紧急疏散过程中，不同类型的行人有不同的运动行为，这些运动行为对疏散过程（尤其是效率）的影响不同。