Pedestrian evacuation under guides in a multiple-exit room via the fuzzy logic method

Abstract

The importance of guides has been paid more attentions by managers of large buildings during the establishment of evacuation plans. Considering the inaccuracy of perceptual information by pedestrians and guides, this paper chooses the fuzzy logic theory which is good at processing qualitative knowledge and experience with unclear boundaries to investigate the problems of guide selection by informed followers and exit selection by guides. Two factors, namely the follower’s normalized distance to the guide and the normalized local pedestrian density around the guide, are taken into consideration when determining to select which guide for informed followers. Guide’s normalized distance to the exit and the normalized density around the exit are chosen as the input variables of the fuzzy inference system to assist in deciding which exit to choose for guides. On the basis of the proposed methods, pedestrian evacuation dynamics is studied through analyzing the simulation data of desired and actual velocities, trajectories, and outflow from each exit, from which the effectiveness of exit selection method proposed in this paper can be explained. Evacuation efficiency due to the limited visibility, guide quantity and exit attribute is further investigated. Simulation results indicate that evacuation efficiency does have a close relationship with the building visibility, guide quantity and exit width. To a certain extent, more guides, greater visibility and wider exit are conducive to a large proportion of pedestrians to realize safety evacuation. Moreover, the exit is suggested to design at the middle position of a wall which is better to the traffic of guided groups.