Building Evacuation Simulation in Dynamic Environments to assess Strategies and Building floor plan designs

Abstract

Accidental fires in public and large buildings pose significant risks, demanding effective evacuation strategies to ensure the safety of occupants and minimize property loss. This study aims to address the challenges associated with building evacuations by integrating spatial, temporal, and path planning approaches specifically for 2D building plans. By incorporating a geospatial framework, the study enables the evaluation of dynamic changes in the building environment and their impact on evacuation outcomes. The current studies do not show how the cascading effect of the person in the queue and space constraints, static or dynamic, affect the evacuation times and can exacerbate the disaster. Also, the presence of multiple exits, if any, can change the outcomes. This research studies how spatial arrangement and floor plans play an important part in evacuation strategies and advocates for their integration into building planning and design processes. The study delves into the impact of obstacles on evacuation processes, by introducing a subspace model that significantly improves computation time in various geometric spaces. The findings emphasize the hindrance of obstacles, either static or dynamic, on evacuation efficiency and propose efficient path generation to mitigate increased evacuation times. The primary objective of this study is to evaluate dynamic changes in the building environment during evacuations and assess their impact on evacuation outcomes. It also addresses the challenges associated with building evacuations during accidental fires in public and large buildings. The focus is on ensuring occupant safety and minimizing property loss by developing effective evacuation strategies. The technical approach adopted in this study involves the integration of spatial, temporal, and path planning methodologies for 2D building plans within a geospatial framework. The study employs a computational approach that combines occupancy-based path planning. This allows for a complete evaluation of dynamic changes in the building environment, highlighting the interactions between individuals and their chosen paths toward their respective exits. Notably, the study focuses on time-dependent path planning to recognize the evolving nature of emergencies, incorporating dynamic elements to enhance the accuracy of evaluating the building environment in emergency scenarios. The study explores the effectiveness of integrating floor plans into path generation and people flow analysis, recognizing the influence of spatial arrangement on evacuation strategies. The proposed method employs a graph-based path generation approach using the subspace model for an effective methodology for evacuating occupants by adapting to changing paths based on a 2D structural plan. The path planning accounts for not just current position to the main exit, but incorporates other conditions like crowding and lag on some path segments, time delays in utilizing alternate exits like windows, and finally reaching to the Evacuation point. The study shows that in comparison to the base scenario, the adverse situation resulted in a more than 12x increase in evacuation time for the 89 building occupants. This underscores the impact of changed dynamics and emphasizes the necessity for well-planned alternative exits with sufficient capacity. The study gives the influence of main exit capacity and alternate exit access time on overall evacuation time, calling for comprehensive evaluation and appropriate definition of capacities for doors and windows. The subspaces offer a clear line of sight to the exit and are integral to the evacuation path. In scenarios with obstacles, the approach, where the main door served as the primary exit, resulted in only a marginal increase of 215 seconds for 70 occupants compared to clear spaces. However, when obstacles led to disconnection of the network, alternate exits like windows were used, causing an almost 4x rise in evacuation time to 996 seconds in Case 2. In cases with critical node blockage, the last person took around 1295 seconds to exit, a 6x increase compared to the base scenario with obstacles. This gives the need for strategic planning of alternate exits to reduce delays and optimize people flow rates. The study called for further work to seamlessly integrate the proposed method with static and dynamic components of indoor spaces, emphasizing the development of an occupant-friendly information dissemination model. The use of an agent-based model facilitated dynamic decision-making and adaptation to changing graph networks. The study's technical approach thus encompasses a holistic evaluation of evacuation strategies, considering spatial path planning aspects, with practical implications for building design and emergency planning. In conclusion, this study significantly advances the field of evacuation planning and modeling by addressing the complexities associated with building evacuations during accidental fires. The outcomes clearly indicate the improved effectiveness of evacuation plans, highlighting the importance of cascading effect due to the presence of many occupants at one place. This happens due to the holding capacity of the edge. Due to this cascading effect the evacuation time increases to more than 6 fold in several cases. Moreover, the research's real-world applications are evident in the practical implications it provides for building design and emergency planning. By integrating these innovative approaches into emergency systems, accurate and timely information can be provided to evacuees, thereby improving overall safety. In summary, this research significantly enhances our understanding of fire evacuation strategies, providing practical solutions and considerations that can be applied to real-life scenarios, ultimately contributing to the advancement of safety measures in public and large buildings.