Simulation of pedestrian behaviours in high-rise buildings based on Intelligence Decision P System

ABSTRACT

Modelling pedestrians evacuation behaviour has become an essential issue in improving the safety of high-rise buildings evacuation. The motivation of this paper is to extend the Intelligence Decision P System (IDPS) to describe and evaluate the behaviour selection of pedestrians. Two types of pedestrians evacuation behaviours were studied in the framework of IDPS: queuing behaviour, corresponding to queuers who adhere to the evacuation instructions and follow the front pedestrians, maximising the efficiency of global evacuation; and competitive behaviour, corresponding to competitors who elbow their way through the crowd, with frequently overtaking each other but no violence whatsoever, maximising the efficiency of individual evacuation. The IDPS also modelled obstacle avoidance behaviour at two specific fire scenarios, diffused fire scenario and non-diffused fire scenario. The simulation results showed that the cumulative effects of queuing behaviour and competitive behaviour varied with different walking speeds and different floors. When a fire broke out, the evacuation efficiency of two behaviours was negatively correlated with the density of pedestrians in front of the exit. As the fire spread to the maximum burning area, competitors suffered fewer casualties slightly than queuers. The findings of this study can be used to develop new behavioural models of evacuation simulations in high-rise buildings.

GRAPHICAL ABSTRACT

KEYWORDS:

• Membrane computing
• Intelligence Decision P System
• competitive behaviour
• queuing behaviour
• high-rise building

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [grant numbers 61872325, 61772290]; Fundamental Research Funds for the Central Universities [grant number 63192616]; and the China Scholarship Council [grant number 201806405004].