Characteristics analysis for travel behavior of transportation hub passengers using mobile phone data

The travel behavior of passengers from the transportation hub within the city area is critical for travel demand analysis, security monitoring, and supporting traffic facilities designing. However, the traditional methods used to study the travel behavior of the passengers inside the city are time and labor consuming. The records of the cellular communication provide a potential huge data source for this study to follow the movement of passengers. This study focuses on the passengers’ travel behavior of the Hongqiao transportation hub in Shanghai, China, utilizing the mobile phone data. First, a systematic and novel method is presented to extract the trip information from the mobile phone data. Several key travel characteristics of passengers, including passengers traveling inside the city and between cities, are analyzed and compared. The results show that the proposed method is effective to obtain the travel trajectories of mobile phone users. Besides, the travel behavior of incity passengers and external passengers are quite different. Then, the correlation analysis of the passengers’ travel trajectories is provided to research the availability of the comprehensive area. Moreover, the results of the correlation analysis further indicate that the comprehensive area of the Hongqiao hub plays a relatively important role in passengers’ daily travel.

     

求解一类线性规划问题的原始贪婪算法和对偶贪婪算法及其相互关系

给出了求解一类线性规划问题的一种贪婪算法和求解其对偶问题的一种贪婪算法,并讨论了这两种算法的若干性质及其相互关系.     

尺寸可变的装箱问题的近似算法的研究

给定物品系列,要求将所有物品装入到不同类型的箱子中,以实现从第1个箱子到最后1个箱子被使用的箱子的总尺寸最小化.用最坏情况绝对性能研究在线算法,给出了一种最坏情况绝对性能比是3的近似算法.作为这种算法的应用,给出了一种脱线算法,其最坏情况绝对性能比是2.     

路段感应式信号控制行人过街系统设计

为保障路段行人安全过街,提高行人过街效率,设计了基于感应控制的路段行人安全过街系统.该系统分别采用视频和感应线圈实现对过街行人和路段车辆的自动检测,并据此设计行人过街感应信号控制方案,辅以“减速带式”人行横道、路段行人过街通道灯和智能人行道护栏3种过街辅助设施,实现时空上的人车分离.     

基于社交网络数据的交通突发事件识别方法

为了从社交网络数据中挖掘出交通突发事件，研究了基于机器学习的文本识别方法。通过关键词和地点定位，利用网页爬虫“Beautiful Soup”爬取到原始文本。采用正则匹配、重复度计算以及“0-1”标记预处理原始文本。基于预处理后文本特征，研究了基于特征权重的特征词选取方法；其中，特征权重的计算综合了词语的出现频率和含有该词语的文本所占比例，通过将二者归一化并加权合并，获得训练集突发事件文本中各个无重复词语的特征权重；依据此值选择确定特征词，并用于后续分类器的输入。测试对比了不同的分类器以及特征词选择方法，结果表明，所提特征词选取方法与XGBoost分类器结合，在交通突发事件识别上具有最好的综合表现，精确率为0.679 6，召回率为0.648 1，F1值为0.663 5，AUC值为0.759 4。      

高速公路营运管理综合数据中心框架设计研究

为适应信息化要求,整合并优化配置信息资源,推进信息化建设,提升高速公路管理业务服务水平,在分析高速公路信息化现状和需求基础上,结合云计算、数据仓库、数据挖掘融合等技术理念,架构出1种信息化系统综合数据中心框架体系,梳理了数据中心营运管理业务的数据流程,设计数据中心平台及子系统功能,提出了数据中心建设运行的标准规范和安全管理机制。      

Mechanism Analysis and Implementation Framework for Traffic Signal Control of Over-saturated Intersection Group

For the purpose of improving traffic signal control at over-saturated intersection group, the traffic control targets and structure were optimized, and an implementation framework was proposed. After analyzing the congestion inducements, the maximum throughput and minimum average queue of critical route were selected as the optimization objectives. The critical route layer was added to optimize control structure. The implementation framework for signal timing plan optimization was also put forward considering the over-saturated traffic control characters. Detailed optimization measures for each layer were also discussed. Simulations with the traffic data of an intersection group at Guangzhou Road in Nanjing were selected to compare the performance of dynamic timing plan using proposed strategy and static timing plan. The results indicated the proposed method can be used to optimize traffic signal control of over-saturated intersections group with satisfying robustness and reliability.     

求解背包问题的一种新的近似算法

给出了求解背包问题的一种新的近似算法,这一算法是一种改进的贪婪算法,即将部分枚举法与贪婪算法相结合.从而使其具有更好的性能保证.同时,从理论上证明了这一算法的可靠性.最后,通过具体算例验证了算法的有效性.     

给定部分大小的最大有向割问题的一种近似方法

给出了求解给定部分大小的最大有向割问题的一种新的近似方法,并讨论了它的性能保证.该方法的核心是利用Pipage技术,并结合线性松驰的基本解的特性,为给定部分大小的最大有向割问题设计出了0.5-近似算法.     

Short‐term highway traffic flow prediction based on a hybrid strategy considering temporal–spatial information

Short‐term traffic flow prediction is fundamental for the intelligent transportation system and is proved to be a challenge. This paper proposed a hybrid strategy that is general and can make use of a large number of underlying machine learning or time‐series prediction models to capture the complex patterns beneath the traffic flow. With the strategy, four different combinations were implemented. To consider the spatial features of traffic phenomenon, several different state vectors including different observations were built. The performance of the proposed strategy was investigated using the traffic flow measurements from the Traffic Operation and Safety Laboratory in Wisconsin, USA. The results show the overall performance of hybrid strategy is better than a single model. Also, incorporating observations from adjacent junctions can improve prediction accuracy. Copyright © 2017 John Wiley & Sons, Ltd.