Abstract
This study addresses the crowding situations caused by staff stranding in gymnasiums during emergencies. It introduces the theoretical basis for relative accidents and defines basic parameters such as crowd flow rate, group flow, marginal sizes of evacuation channels, and walking velocities across different densities. A quantitative model for the number of stranded individuals is deduced and constructed. Through a case study of a Chinese university’s multifunctional gymnasium, the research identifies potential evacuation bottlenecks and provides specific recommendations for improving stand exits to enhance safety management and emergency planning.
Introduction: The Risk of Crowd Stranding
As university gymnasiums evolve into complex, multifunctional spaces hosting diverse events, their evacuation profiles become increasingly high-risk. While structural failures or fires are often the triggers for disasters, the actual cause of most casualties is "stamping"—crushing due to extreme crowd density.
The core challenge lies in the imbalance between "inflow" from the stands and the "outflow" capacity of the exits. This creates a "stranded crowd," where density increases until the "arching effect" occurs—a physical jam where flow stops entirely. This paper moves beyond simple total evacuation time formulas to a granular, time-step-based model of crowd accumulation.
Methodology: The Stranded Crowd Number Quantitative Model
The proposed model is based on classic crowd flow theory and the "track-cross" theory of accidents. It conceptualizes the transition of crowd states as they approach a bottleneck, moving from free flow to delayed movement and eventually crowding.
Key components of the model include:
- Flow-Density-Velocity Relationship: Utilizing established functions, the model relates walking speed to density, noting that velocity drops to zero at a density of 5.38 people/m².
- Marginal Size Correction: The method subtracts "marginal sizes" (15–20 cm) from hallway widths to account for the buffer space people naturally maintain from walls and obstacles.
- Time-Step Analysis: The model calculates the "aggregation clustering number" (inflow) and "outflow clustering number" at 1.5-second intervals.
- The Stranding Formula: The "stranded number" ($N_d$) is defined as the difference between the number of people entering a waiting area ($N_i$) and those successfully exiting it ($N_e$).
Case Study: A Multifunctional University Gymnasium
The model was applied to a Chinese university gymnasium with a capacity of 1,520 mobile seats and 2,000 fixed seats. The infrastructure includes 6 main exits with stair widths of 2.3–2.5m and fire-resistant doors of 1.5m.
Key Insights and Results
- The 5-Second Warning: Analysis revealed that stranding begins just 5 seconds after evacuation starts. By 6.5 seconds, the stranding number increases rapidly, marking the critical window for potential accidents.
- The "Arching" Threshold: In a 6m² waiting area, once the crowd reaches approximately 32 people (5.38 people/m²), flow stops completely due to the physical jam.
- Bottleneck Identification: While overall evacuation took roughly 300 seconds, specific exits on the west side of the 2nd-floor stands were identified as significant hazards. At these exits, density reached 4.3 people/m² almost immediately, exceeding the "danger" threshold of 3.59 people/m².
- Design Recommendation: To mitigate these risks and meet safety requirements, the study concludes that stand exits must be widened to more than 3 meters.
Conclusion and Future Directions
While the mathematical model provides a robust framework for identifying physical bottlenecks, human factors remain a significant variable. The model currently assumes linear flow and does not fully account for psychological factors like extreme panic or irrational behavior. Future work should focus on integrating these psychological variables or utilizing real-time sensor data to refine the "stranded number" calculation during live events. This research provides essential references for the design, management, and transformation of large-scale public venues.
Summary updated