Research

围绕交通安全的主题,以行人与疏散动力学为研究主线,对人员安全疏散中的基本科学问题开展了深入的研究,获得了典型建筑结构内的人员运动行为规律,发现了人群湍流运动的新特征,构建了异质行人流模型,提出了基于协同性的异质人群流动状态管控方法,为评估人员疏散过程的安全性提供了理论依据和评估工具,为人群管控策略指定提供了科学依据。主持国家级项目资助3项,其中:国家自然科学基金面上项目2项,国家自然科学基金青年基金项目1项,累计获得5项国家自然科学基金资助,1项国家重点实验室开放课题资助。另外,作为研究骨干获得教育部2011计划“轨道交通安全协同创新中心”项目“高铁事故致因机理与应急保障关键技术”资助,科技部国家重点研发计划“公共安全风险防控与应急技术装备”重点专项项目“区域综合交通基础设施安全保障技术”资助。

任现职以来,共计申请获得发明专利4项,软件著作权2项,发表高水平论文40余篇(SCI收录20篇,EI收录8篇),其中:以第一作者或者通讯作者身份发表A++类论文2篇,A+类论文4篇,A类论文7篇,B+类论文2篇。研究成果多次在交通领域著名国际会议展示,包括Annual Meeting of the Transportation Research Board会议论文2篇(均为第一作者),Traffic and Granular Flow会议论文2篇(第一作者1篇)以及智能交通领域顶级国际会议International Conference on Intelligent Transportation Systems会议论文1篇(第一作者)。目前发表的论文多次被他人引用和正面评价,例如:

在实验研究方面,本人关于楼梯疏散实验的成果发表于安全科学与工程领域内著名国际期刊Safety Science的人员疏散专刊上。该期专刊针对(译文)“人员疏散这一新兴领域”,介绍“世界范围内人员疏散研究的最新进展”。该刊从“来自16个国家和地区44篇论文中精选出6篇论文”。人员疏散领域著名学者Prof. Galea相信这6篇论文构成的专刊虽不能完全涵盖21 世纪人员疏散领域所有的挑战,但对解决这些挑战有重要贡献the special issue is impossible to cover the complete range of 21st century grand challenges, however, does make an important contribution to the grand challenge areas.。该文发表以后受到广泛关注,Prof. Galea后续开展了多次实验对比分析,并验证了本人提出的建筑梯道疏散中疲劳引起的个体运动速度变化的推测。此外,关于人群湍流运动状态的论文最近被北京交通大学高自友教授在管理学类顶级国际期刊“Operations Research”上最新发表的论文中正面引述。

在理论研究方面,本人提出的确定人群运动状态由自由流向拥塞流转变、由拥塞流向自由流转变的密集人群中异常流动状态监测报警方法发表于国际火灾安全研究领域内重要国际会议“9th Asia-Oceania Symposium on Fire Science and Technology”,该项研究工作受到会议Technical Committee的一致推荐,入选本届会议最佳论文(评选意见译文:“作者们展示了一种探测人群流动状态的自动聚类分析方法,并用一些实验和测试视频对这一方法进行了校验,给出了对于分析人员行为的一些很好、很有意义的结果。这篇文章对于行人动力学分析,特别是人群中的分组行为有重要意义。”);申请人提出的人员疏散模型在国际知名期刊Physica A”和火灾安全领域顶级国际期刊“Fire Safety Journal”一经发表即被广泛关注,分别入选所在期刊当季最受关注的25 篇热点论文之一。

截至目前,本人研究成果已被SCI引用235次,H指数9Google Scholar引用532次,H指数13,在行人交通和疏散动力学领域产生了一定的影响,因此受邀担任“Collective Dynamics”、“Modern Transportation”等国际期刊编委,担任IEEE Transactions on Intelligent Transportation SystemsTransportmetrica ATransport PolicyPhysica APhysics Letters ASimulationInternational Journal of Modern Physics CJournal of Harbin Institute of Technology等国际期刊以及Transportation Research Board Annual MeetingPedestrian and Evacuation Dynamics等系列国际会议审稿人、国家自然科学基金通讯评议人、教育部学位中心抽检学位论文通讯评议专家、National Center of Science and Technology Evaluation of the Republic of Kazakhstan基金通讯评议人,第一及第二届International Conference on Mass Transit Rail Facilities Design and Management2012 Fire Science and Engineering Conference2013 Fire Science and Engineering Conference2014年中国消防协会科学技术年会青年学者论坛、2014 International Conference on Rail Development and Management for Sustainable Urbanism2015 7th International Conference on Performance-based Fire and Fire Protection Engineering等国际会议程序委员会委员,中国建筑学会建筑防火综合技术分会火灾风险评估专业委员。

此外,在2014年、2015IEEE国际智能交通系统会议(IEEE Conference on Intelligent Transportation System, IEEE ITSC)框架内,组织了行人交通与疏散动力学(Pedestrian traffic and Evacuation Dynamics)分会场;协助申请并共同组织了人员疏散领域的著名国际会议“Pedestrian and Evacuation Dynamics 2016,这是该系列会议将于首次在亚洲国家举办。相信这些工作对于提高我院在相关领域的学术声誉、确立我院在该领域内的国际地位有正面的影响。

 

Studies on Pedestrian Traffic and Evacuation Dynamics

Sep, 2015

With the increase of built environment complexity, pedestrian movement in public space such as modern transportation centers, airport terminals, shopping malls and so on undergoes continuously growing number of events including fire, terrorism, and massive crowd gathering, which induced in tens to hundreds of victims, such as the 2013 Allah railway station disaster in India, the 2014 Conakry beach concert stampede in the capital of Guinea and the 2015 New Year’s Eve trampling disaster in Shanghai China. Facing these serious disasters and problems, a better understanding of the origin and development process of crowd disaster would provide a better solution to guarantee a safe, efficient, and comfortable traveling experience for pedestrians. Thus, during the past few years, researches on the topic of pedestrian traffic and evacuation dynamics in public pace including various transportation systems have been performed. The following works have been done,

1, Well-structured experiments were designed and performed to capture the moving characteristics of pedestrians in complex built environment. For typical corridor, pedestrians’ moving trajectories were first extracted with digital image processing and were then mapped into real space coordinates by adopting a direct linear transformation approach. Moving characteristics of single pedestrians, and interaction between pedestrians and the corridor, as well as interaction between pair pedestrians were analyzed. For stairwell, single pedestrian down-ward moving speed and crowd evacuation characteristics including merging flow features have been found. These research findings provide solid foundation to validate and verify pedestrian models.

2, Based on the experimental findings, two kinds of pedestrian models were established. The first kind relates to pedestrian movement on horizontal plane. A metric distance free model, namely, k-Nearest-Neighbor (kNN) model was proposed. In the kNN model, an agent’s direction choice is only influenced by a fixed number of surrounding people, but not all the people in a given metric distance. The 2nd model relates to pedestrian evacuation in stairwell. A new evacuation model which could simulate the movement features of both pedestrian and lift in combined stairwell and lift evacuation was proposed. It was found the use of lift may not necessary benefit the evacuation process. Based on the simulation results of the proposed model, a first principle was proposed when designing combined evacuation of stair and lift to ensure a safety evacuation process.

3, Pedestrian traffic simulation tool was developed to evaluate the safety and comfort level of built environments. A pedestrian space analysis model is proposed and incorporated with the agent-based pedestrian model on a Geographic Information System (GIS) platform. A bi-layer mesh method is proposed to divide the pedestrian space in complex building into cells, and later used to realize origin-destination assignment of the pedestrians. Different designs of train platform and a complex building have been modeled to evaluate the pedestrian flow patterns. The results indicated that the simulation platform on GIS can serve as a useful tool to evaluate pedestrian movement in transportation facilities.

4, Crowd trampling disasters have been investigated. Video recordings right before the trampling in 2010 Love Parade disaster have been quantitatively analyzed to explore the bursts of unusual crowd movement patterns, crowd-quakes. A special phase transition from stopped pedestrians to turbulent flow was found. It was also found there were different kinds of contacts among pedestrians. These contacts mode may be the origin of the pressure release. Thus, based on the discovered contacts mode, we further built heterogeneous contact model for massive crowd. Numerical simulation results of the model can qualitatively describe the statistical properties of pedestrian movement behavior in crowd quakes.

Related publications:
  •  Ma J, Chen J, Lo SM, Xu SM, Song WG, Evaluating bottleneck magnitude of massive crowd flow networks, 2015 IEEE 18th International Conference on Intelligent Transportation Systems, in press.
  •  Ma J, Song W.G., Lo S.M. and Fang Z.M., New Insights into Turbulent Pedestrian Movement Pattern in Crowd-Quakes, Journal of Statistical Mechanics: Theory and Experiment, 2013, P02028.
  •  Ma J., Lo S.M., Song W.G., Wang W.L., Zhang J. and Liao G.X., Modeling Pedestrian Space in Complex Building for Efficient Pedestrian Traffic Simulation, Automation in Construction, 2013, 30:25-36.
  •  Ma Jian, Song Weiguo and Lo Siuming, Simulation of crowd-quakes with heterogeneous contact model, Traffic and Granular Flow 2013, Julich, Sep. 25-27, 2015, pp103-110.
  •  Ma Jian, Lo Siuming, and Song Weiguo, Cellular automata modeling approach for optimum ultra high-rise building evacuation design, Fire Safety Journal, 2012, 54:57-66.
  •  Ma Jian, Song Weiguo, Tian Wei, Lo Siuming and Liao Guangxuan, Experimental study of ultra high-rise building evacuation in China, Safety science,2012,50(8):1665-1674.
  •  Ma Jian, Song Weiguo and Liao Guangxuan, Multi-grid simulation of Pedestrian Counter Flow with topological Interaction, Chinese Physics B, 2010, 19(12):128901.
  •  Ma Jian, Song Weiguo, Fang Zhiming, Lo Siuming and Liao Guangxuan, Experimental Study on Microscopic Moving Characteristics of Pedestrians in Built Corridor Based on Digital Image Processing, Building and Environment, 2010, 45(10):2160-2169.
  •  Ma Jian, Song Wei-guo, Zhang Jun, Lo Siuming and Liao Guangxuan, k-nearest-neighbor interaction induced self-organized pedestrian counter flow, Physica A, 2010, 389:2101-2117.