Analysis of crowd stampede risk mechanism A systems thinking perspective

• Published in: Kybernetes Vol. 48; no. 1; pp. 124 - 142
• Main Authors: Kefan, Xie, Song, Yu, Liu, Sishi, Liu, Jia
• Format: Journal Article
• Language: English
• Published: London Emerald Group Publishing Limited 14.01.2019
• Subjects: Accidents; Behavior; Computer simulation; Delay; Design; Fatalities; Risk analysis; Simulation; Systems analysis; Variables
• ISSN: 0368-492X; 1758-7883
• DOI: 10.1108/K-11-2017-0415

Purpose The purpose of this paper is to analyze the crowd stampede risk mechanism from the perspective of systems thinking. Design/methodology/approach Causal loop diagram is drawn to outline the non-linear interactions among complex factors across the whole system and dissect the contributory factors of crowd stampede accident. To systematically construct the theoretical framework and find fundamental solutions, co-word analysis with Citespace is used to get the critical data. An agent-based simulation using Pathfinder is conducted to develop a spatial model for the Shanghai Stampede Accident that happened in 2014. Findings The causal loop diagram is formed to not only illustrate the symptomatic solutions with a quick fix but also dissect the fundamental solutions through an underlying systemic analysis. The simulation shows that crowd stampede experiences an interactive process of accumulation, trigger, delay, break and diffusion of risk factors within the crowd system. A linkage effect among the multidimensional characters of individuals and the system accelerates the stampede risk deterioration. There exists delay of the result of effect from the deep-level measure. Practical implications A top-down approach is offered to policymakers for crowd stampede risk protocol design and synergic emergency control that may reduce the risk of the stampede. Originality/value In this study, SDFT paradigm is proposed as the critical solution for the crowd stampede accident. In addition, a chain effect of energy and a linkage effect within the crowd system is illustrated for in-depth understanding of crowd stampede risk.