New Agent-Based Egress Model Allowing for Social Relationships

Authored by J Fang, S El-Tawil, B Aguirre

Date Published: 2016

DOI: 10.1061/(asce)cp.1943-5487.0000532

Sponsors: United States National Science Foundation (NSF)

Platforms: No platforms listed

Model Documentation: Other Narrative Pseudocode

Model Code URLs: Model code not found

Abstract

Many of the models developed to date for human evacuation lack realism because of their inability to adequately simulate social relationships. To begin to address this shortcoming, an agent-based model is proposed based upon a new technique termed scalar field method (SFM). The new method draws on an analogy to a charged particle in an electromagnetic field. In the new model, virtual potential energies (VPEs), that can be made to represent both human will and social relationships, simulate the interactions that occur between an agent and its surrounding entities. Each agent has stochastic characteristics, is independent, and makes autonomous decisions regarding behavior by minimizing the VPE. The proposed model supersedes the pattern-based format of many current ones and provides the potential for greater emergence and variability. Preliminary tests show the ability of the new model to mimic reasonable evacuation behavior, and its potential for exploring the significance of social relationships during egress.

Tags

Simulation Crowd dynamics collective behavior Evacuation Fire Flow Counterflow