Behavioral,data-driven,agent-based evacuation simulation for building safety design using machine learning and discrete choice models

To improve occupant safety during building emergencies, evacuation simulations have been widely used for building safety design. Since occupant behavior is a determining factor for the outcome of building emergencies, accurately capturing how occupants make decisions and integrating occupants’ decision-making processes in evacuation simulations is important. In this study, based on the results of fire evacuation experiments in a virtual metro station, how different social (crowd flow) and environmental (visual access and vertical movement) factors would affect individuals’ wayfinding behavior was predicted using machine learning and discrete choice models. The trained models were further employed in agent-based evacuation simulations to examine crowd evacuation performance under different building design scenarios. Both the machine learning and discrete choice models could accurately predict individuals’ directional choices during emergency evacuations. Different building attributes could collectively influence occupant behavior, leading to distinct exit choices and evacuation times. While both the trained machine learning and discrete choice models generated similar results, the discrete choice model had better interpretability. Moreover, by comparing the trained models in this study with a model developed in a prior study, it was found that agents had significantly distinct responses to different building designs. Critical factors (e.g., type and size of buildings, occupants’ familiarity with the building) for the applicability of evacuation models were identified. Furthermore, recommendations were provided for future research that aims at employing evacuation simulations for building design evaluation and optimization.     

基于元胞自动机模型的人员疏散仿真研究

公共安全是经济发展和社会稳定的基础。近年来由于火灾、地震、恐怖活动等突发因素诱发的公共场所人员安全事故屡见报道,在公共场所的安全设计中,安全疏散性能已成为至关重要的因素。为解决上述问题,人员疏散的计算机仿真对提高公共场所的安全疏散性能具有重要意义。对于出口距离最近的原则建立人员疏散的二维随机元胞自动机模型,其中考虑人员绕行的影响,并利用Matlab软件编写了疏散过程的程序,对大空间疏散人群进了仿真,得到了人员疏散基本规律及疏散时间。研究表明:考虑人员绕行的模型更能体现真实的疏散情景。     

Event-driven modeling of elevator assisted evacuation in ultra high-rise buildings

In the past few decades, numerous ultra high-rise buildings have been erected in the metropolitan areas of many cities around the world. For the related building designers, building occupants and the governments, fire safety problem is certainly a major concern, especially after the collapse of World Trade Center in the 9/11 event. That disaster makes people reconsider ultra high-rise building evacuation strategies. Of the current strategies, using elevators in ultra high-rise buildings to assist evacuation seems to be promising in improving evacuation efficiency. To quantitatively evaluate elevator assisted evacuation process, an event-driven agent-based modeling approach is proposed in the present paper. This modeling approach could capture not only the movement characteristics of stair-using occupants but also the detailed elevator motion features. The combined effects of occupants’ and elevators’ parameters on the evacuation efficiency have been investigated. Results indicated that the model is helpful to reveal the dynamics of elevator assisted evacuation, and sometime, using elevators to move all occupants to ground safety point may not be an optimal solution.     

复杂条件下多层建筑人员协作疏散仿真

研究具有复杂多层协作过程条件下的人员疏散控制系统,就能够比较准确地模拟突发情况下人员的疏散情况。本文对元胞自动机进行了改进,综合人员个体特征和从众心理等各种复杂因素,对具有复杂障碍物的多层建筑中人员疏散过程进行了计算机仿真分析,并给出了人员疏散效率与人员的从众系数、障碍物及出口位置等因素的关系。该仿真能够很好地模拟大型公共场所发生突发事件时人员疏散的情况,对在复杂地理环境及人员特性条件下的多层建筑突发事件疏散策略制定具有一定的实际参考意义。     

基于ACP方法的高层建筑火灾中人员疏散策略研究

高层建筑结构规模巨大、人员众多,而疏散出口数量有限,如何有效安排疏散出口人员分布,提高疏散效率,是火灾应急管理的重要研究内容. 以人为中心的疏散系统是典型的复杂系统,具有难以真实实验分析的困难. 本文基于ACP（人工系统、计算实验、平行执行）方法,以智能体技术为核心建立了人工疏散系统,基于火灾场景,利用计算实验对疏散策略进行了验证、评估,给出了实际疏散系统与人工疏散系统的平行执行实现思想. 最后,通过案例验证了方法的可行性.     

A multi-agent based framework for the simulation of human and social behaviors during emergency evacuations

Many computational tools for the simulation and design of emergency evacuation and egress are now available. However, due to the scarcity of human and social behavioral data, these computational tools rely on assumptions that have been found inconsistent or unrealistic. This paper presents a multi-agent based framework for simulating human and social behavior during emergency evacuation. A prototype system has been developed, which is able to demonstrate some emergent behaviors, such as competitive, queuing, and herding behaviors. For illustration, an example application of the system for safe egress design is provided.     

See how they run: modeling evacuations in VR

A virtual reality application called Vegas lets users experiment in real time with evacuation scenarios. Colt Virtual Reality, Ltd., based in Hampshire, United Kingdom, developed Vegas (virtual egress analysis and simulation), which runs on a 486-based PC. Its designers say that other evacuation models lack the user-friendly visualization of Vegas, and they hope its graphic capabilities will make the data available and comprehensible to a wider audience. In particular, they hope it will allow people to see how quickly fire and smoke can spread, and how important fast and calm action can be in an emergency. Colt has used Vegas to predict egress times from an underground subway station and evacuation times from a cross-channel ferry, as well as from several other, more traditional, buildings     

Crowd evacuation simulation using active route choice model based on human characteristics

Modern buildings have become larger in scale and function, and the complexity has also increased considerably. For these reasons, there are more difficulties in evacuation and rescue when an emergency occurs, so effective evacuation methods and risk should be predicted and applied to building design, safety training, and education. We have developed an active route choice model based on human body organs and characteristics that detects risks and route conditions, communicates with neighboring occupants, determines the bottleneck point, and selects evacuation routes according to each occupant's personal characteristics. In this study, we introduce the implementation process and characteristics of the active route choice model, and by applying the model to the occupants, we compared the evacuation times according to the route condition, number of occupants, and corridor width in a virtual environment. We believe that realistic and valid results can be obtained by applying the active route choice model in crowd evacuation simulation.     

Towards a customizable immersive virtual reality serious game for earthquake emergency training

Earthquake emergencies require a variety of behavioral responses in order to ensure the safety of occupants, which is different from simply exiting a building in fire emergencies. This makes it more complex to train building occupants in order to acquire skills that align to best practices for immediate earthquake response and post-earthquake evacuation. In recent years, Immersive Virtual Reality (IVR) and Serious Games (SGs) have become popular as training tools for earthquake emergencies. IVR SGs have been introduced to train individuals for specific building layouts or settings with fixed training objectives. However, the lack of flexibility in existing IVR SGs makes it challenging to have widespread uptake as trainees require different training objectives, pedagogical strategies, context, and content. As a result, the effectiveness of IVR SGs training is jeopardized if the customization ability is limited. To overcome this limitation, this paper presents a customization framework for IVR SGs suited to earthquake emergency training, using the concept of adaptive game-based learning. Trainees can receive training in context by customizing virtual environments, storylines, and teaching methods. A case study was undertaken to validate the proposed framework. Results showed the potential to carry out the customization process with ease, to generate a customized training experience, and to deliver the customized training for optimum learning.     

基于ACP方法的应急疏散系统研究

目前地铁、建筑及煤矿等人员密集空间突发事件频发,解决人员密集空间人员安全疏散是当前社会发展亟待解决的问题. 本文以地铁为例研究了基于ACP方法的平行应急疏散系统,该系统由应急疏散人工系统、计算实验 和平行执行三部分组成. 平行应急疏散系统不仅适用于地铁,还适用于建筑及煤矿等生活和工作空间应急疏散管理,可实现人员培训、疏散演练、疏散方案优化和评估等功能. 本文初步建立了地铁枢纽站平行应急疏散系统,并进行了两种典型场景的计算实验,实验结果表明平行应急疏散系统可有效提高应急疏散效率.     

Prototyping virtual reality serious games for building earthquake preparedness: The Auckland City Hospital case study

Enhancing evacuee safety is a key factor in reducing the number of injuries and deaths that result from earthquakes. One way this can be achieved is by training occupants. Virtual Reality (VR) and Serious Games (SGs), represent novel techniques that may overcome the limitations of traditional training approaches. VR and SGs have been examined in the fire emergency context; however, their application to earthquake preparedness has not yet been extensively examined.We provide a theoretical discussion of the advantages and limitations of using VR SGs to investigate how building occupants behave during earthquake evacuations and to train building occupants to cope with such emergencies. We explore key design components for developing a VR SG framework: (a) what features constitute an earthquake event; (b) which building types can be selected and represented within the VR environment; (c) how damage to the building can be determined and represented; (d) how non-player characters (NPC) can be designed; and (e) what level of interaction there can be between NPC and the human participants. We illustrate the above by presenting the Auckland City Hospital, New Zealand as a case study, and propose a possible VR SG training tool to enhance earthquake preparedness in public buildings.     

Social influence in a virtual tunnel fire – Influence of conflicting information on evacuation behavior

Evacuation from a smoke filled tunnel requires quick decision-making and swift action from the tunnel occupants. Technical installations such as emergency signage aim to guide tunnel occupants to the closest emergency exits. However, conflicting information may come from the behavior of other tunnel occupants. We examined if and how conflicting social information may affect evacuation in terms of delayed and/or inadequate evacuation decisions and behaviors. To this end, forty participants were repeatedly situated in a virtual reality smoke filled tunnel with an emergency exit visible to one side of the participants. Four social influence conditions were realized. In the control condition participants were alone in the tunnel, while in the other three experimental conditions a virtual agent (VA) was present. In the no-conflict condition, the VA moved to the emergency exit. In the active conflict condition, the VA moved in the opposite direction of the emergency exit. In the passive conflict condition, the VA stayed passive. Participants were less likely to move to the emergency exit in the conflict conditions compared to the no-conflict condition. Pre-movement and movement times in the passive conflict condition were significantly delayed compared to all other conditions. Participants moved the longest distances in the passive conflict condition. These results support the hypothesis that social influence affects evacuation behavior, especially passive behavior of others can thwart an evacuation to safety.     

突发事件情景下地铁站人员应急疏散问题综述*

突发事件情景下,合理的进行地铁站人员应急疏散是减少人员伤亡和财产损失的有效途径。从地铁站人员应急疏散的仿真研究、不同规则下疏散模型的研究和智能算法在疏散建模中的应用三个角度对突发事件下地铁站人员应急疏散问题进行综述,并分析了当前疏散模型、求解算法以及行人特征数据的不完善之处。提出了把行人的心理行为特征、建筑物设施因素引入地铁站人员应急疏散问题中,并利用改进的蚁群算法求解最优疏散路径的重要价值。最后,展望了今后地铁站人员应急疏散问题研究的发展趋势。     

A comparative study of evacuation strategies for people with disabilities in high-rise building evacuation

This paper presents new evacuation strategies for a heterogeneous population in high-rise building environments and compares them with traditional simultaneous evacuation strategy. To do so, we first define the maximum comfortable structural capacity of the building as the number of people that can be evacuated without heavy congestion, and estimate it based on flow rate and move frequency ratio. Then we present several evacuation strategies to efficiently evacuate heterogeneous residents at the maximum comfortable structural capacity. The simulation results for a 24-story building suggest several implications for emergency planners. First, we find that a vertically phased evacuation strategy that varies delay times by physical location, is not useful for the simulated building. Second, a phased evacuation strategy that applies a fixed evacuation delay to residents with wheelchairs reduces the aggregated evacuation times, but delaying evacuations of a specific group of individuals may not be ethical or accepted. Finally, evacuation strategies that allows residents with wheelchairs to use elevators are effective, suggesting that emergency administrators should assess whether elevators in their buildings are appropriate for evacuation purposes with appropriate electric controls, electric power, and fire and smoke protection.     

A hybrid hierarchical agent-based simulation approach for buildings indoor layout evaluation based on the post-earthquake evacuation

In the aftermath of severe earthquakes, building occupants evacuation behaviour is a vital indicator of the performance of an indoor building design. However, earthquake evacuation has been systematically neglected in the current building design practice. Arguably, one of the primary reasons for this is that post-earthquake evacuation behaviour is complex and distinct from all other types of evacuation behaviours such as fire. Thus, a comprehensive approach to considering the integration of human evacuation behaviour and a building's indoor layout design, mainly focused on non-structural damage, has been consistently neglected in the literature. In this paper, a hierarchical hybrid Agent-Based Model (ABM) framework integrated with a Cellular Automata (CA) and a 2D Building Information Model (BIM) damage visualisation to consider an approximation of non-structural damage has been developed. The proposed ABM incorporates learning mechanisms and human psychological aspects influencing evacuees' utility during the navigation process. The proposed approach was verified by comparing the results to previous real-life post-earthquake evacuation data and a “model to model” comparison of results from the existing relevant studies. The model prototype was successfully tested to simulate the pedestrian evacuation process from one floor of the new engineering building at The University of Auckland, New Zealand. The proposed simulation approach has been carried out for two different internal layout design alternatives where five population sizes are evacuated through different scenarios. The outputs from this study can be used to improve the design's compatibility of the building's indoor layout with the occupants' post-earthquake evacuation behaviour.     

基于元胞自动机的普通超市火灾疏散模型的构建

通过分析已有的元胞自动机理论,结合人员疏散的特点,构建了考虑出口吸引力、火灾排斥力、摩擦力、排斥力和从众吸引力的普通超市火灾疏散模型。该模型充分考虑了多个因素对疏散过程的影响,对影响因素进行归一化处理,以综合影响因素建立的元胞转移强度作为行人移动规则。针对超市疏散人数、出口宽度及相隔距离、从众心理对疏散时间的影响进行了研究,并采用仿真疏散软件Pathfinder+FDS对模型进行验证,说明疏散模型具有一定可信性。研究表明行人疏散时间随人数呈线性正相关,人数存在临界值;出口宽度越宽或出口越多疏散时间相对越短,当达到出口阈值时对疏散时间影响不大;在陌生疏散环境或紧急情况下适当的从众行为会提高疏散效率。     

A BIM-based simulation framework for fire safety management and investigation of the critical factors affecting human evacuation performance

Fire hazards are a big threat to human life and property safety. The U.S. fire statistics reveal that, in 2017 alone, 1,319,500 fires caused 3400 deaths and 14,670 injuries, which resulted in a loss of $23 billion [1]. Effective evacuation planning in densely occupied buildings should be primarily put in place if both the number of injuries/fatalities and the level of property loss are to be minimized. However, it is not realistic, and is unethical to study human evacuation performance under a burning building. For this reason, computational tools tend to be the best approach for simulating fire growth as well as human response to fire hazards. This study aims to develop a BIM-based simulation framework that implements the Fire Dynamic Simulator (FDS) and agent-based modeling (ABM) for simulating fire growth and evacuation performance for different building layout scenarios. An experimental implementation is conducted to validate the proposed framework, which verified the benefits of (1) using BIM to offer a platform for conducting simulation design and visualizing the simulation results of (a) hazardous fire zones and (b) effective escape routes; (2) simulating fire growth using the FDS tool; (3) developing an agent-based model that accounts for the critical factors affecting evacuation performance; and (4) applying a statistical analysis for investigating the effects of influential parameters from the proposed model. As a result, the simulation outputs can be used to optimize the building design and to investigate the influential factors on human evacuation efficiency. The proposed framework contributes to building fire safety management by enabling to minimize both injuries/fatalities and property loss.     

建筑物火灾中的人群疏散研究

建筑物火灾是我国频发的安全事故,所以应研究建筑物火灾人群安全疏散问题。由于在建筑物火灾中,人群疏散时出现拥堵,存在不安全因素,造成人员伤亡。针对在现有的研究中未考虑人员行为的影响,提出了智能体(Agent)的人群行为建模技术在建筑物火灾中的人群疏散仿真中的应用方法。仿真结果显示基于Agent的行为模型可以仿真出人员特性及决策过程对人群疏散的影响,弥补现有的人群疏散模型的不足。仿真结果证明,Agent的行为建模技术具有仿真火灾全过程中人员疏散行为的功能,适用于建筑物火灾中的人群优化疏散策略。     

Integration of Smoke Effect and Blind Evacuation Strategy (SEBES) within fire evacuation simulation

Many fire evacuation models have been proposed in recent years to better simulate such as an emergency situation. However most of them do not respect a recommendation of fire evacuation experts regarding the fact that evacuees should follow the boundaries of obstacles or wall to find the exits when their visibility is limited by smoke. This paper presents an agent-based evacuation model with Smoke Effect and Blind Evacuation Strategy (SEBES) which respects that recommendation by integrating a model of smoke diffusion and its effect on the evacuee’s visibility, speed, and evacuation strategy. The implementation of this model enables us to optimise the evacuation strategies taking into account the level of visibility. The obtained simulation results on a realistic model of the Metro supermarket of Hanoi confirm the important impact of smoke effect and blind evacuation strategy on the number of casualties.     

基于建筑施工中的安全管理工作探讨

高层建筑的供水、供电等维修费用是一般建筑的3倍,而且超越消防极限,逃生几率相对小,建筑物越高,发生地震造成的伤亡就越大。权衡利弊,高层建筑的经济意义受到质疑,世界各国对高层建筑持谨慎的态度。因此应该严格履行高层建筑的论证和审批程序,适度控制标志性高层建筑,并对已有的高层建筑进行安全排查。高层建筑对节约城市土地、缩短公用设施和市政管网的开发周期、加快改善人们居住条件等方面都起到了重要作用,但也潜伏着不少危险。文章指出要根据城市发展总体规划,综合考虑人口大量集中对城市供应和交通的影响等因素,协调单体高层建筑及高层建筑群同周围环境、已有建筑、名胜古迹、城市风貌之间的分布关系。