基于时空图卷积网络的机场地铁短时客流预测

机场地铁短时客流预测是实现机场旅客快速疏解、航站楼现场运力资源指挥调度的关键。考虑到机场复杂的空间结构与航班波动的影响，建立基于图卷积神经网络（GCN）和组合门控卷积（GLU）的机场地铁短时客流预测模型。通过图卷积神经网络融合机场空间路径点与地铁口的空间结构关系，同时，设计一种组合门控卷积模块挖掘航班波动下地铁客流的时变特征，有效地捕捉地铁客流的波动性。基于首都机场T3航站楼真实客流数据对模型的有效性进行检验，经多次实验结果表明，提出的时空图卷积短时客流预测模型在均方根误差、平均绝对误差和平均绝对百分比误差均小于传统ARIMA预测模型与深度学习中LSTM、STGCN模型，该模型能捕捉地铁客流与航班客流的波动变化关系，具有较高的预测精度，提高了模型预测的鲁棒性。     

灰色模型客流预测

灰色预测模型是数据分析的重要模型,已经应用到了众多领域。天津滨海快速交通发展有限公司经过多年的运营积累了大量的客流数据。该文运用灰色预测模型对滨海快速塘沽站的客流进行了分析,为运营决策提供了一定的依据。     

多尺度时序特征融合的地铁短时客流预测方法

面向城市地铁进出站客流数据,本文提出了一种融合多尺度时序特征的地铁短时客流预测方法。首先,通过数据分析,构建地铁客流的长期周期性和短期波动性等多尺度时序特征融合机制。其次,提出地铁短时客流预测方法模型来有效提高客流预测精度,既利用了长期预测模型捕捉客流的长期周期性规律,又融合了短期预测模型获取客流的短期波动性趋势。最后,在2020年9月厦门地铁53个站点的客流数据上验证模型的有效性。实验结果表明,与LSTM、GRU等基线方法相比,本文提出的方法在地铁客流预测任务上具有预测精度更高、收敛速率更快等优势。     

基于WiFi数据的地铁车站客流预警模型

为了降低地铁车站因客流增加或突发事件产生的客流安全风险,提出了基于WiFi探针检测数据的地铁车站客流预警模型。基于WiFi探针客流数据采集原理、数据属性特征和探针网络化布设方案,实现了对WiFi探针原始数据的预处理。同时建立了基于时间序列的地铁车站短时客流预测模型,并与线性回归模型进行了比较,通过计算地铁车站客流承载能力,构建了车站客流预警指标和分级预警模型。最后以上海地铁江苏路站为例进行模型验证,结果表明,基于WiFi探针技术的地铁车站客流检测和预测模型具备可行性和有效性,且预警模型对客流预警应用和研究有一定的参考意义。     

数据挖掘在铁路客流分析预测中的应用①

应用数据仓库和数据挖掘技术,以铁路客票发售和预订系统为研究主体,将有效的数据挖掘技术应用于铁路客流分析,采用神经网络思想,建立了一个基于BP神经网络模型的客流分析预测模型,为客运部门合理安排运能、科学组织管理提供了准确的决策信息和先进的预测手段。     

基于客流分配的高铁票价调整策略

针对目前高铁票价单一、客运收益率低、区段客流不均衡等问题,提出基于客流分配的高铁票价调整策略。首先,分析影响旅客出行选择行为的相关因素,构建包含经济性、快速性、便捷性和舒适性四项指标的广义出行费用函数;然后,建立兼顾高铁客运管理部门收益最大化和旅客出行费用最小化的双层规划模型,其中上层规划通过制定票价调整策略实现高铁客运收益最大化,下层规划以旅客广义出行费用最小为目标,利用区段不同车次间的竞合关系构建随机用户均衡（SUE）分配模型,同时采用基于改进Logit分配模型的相继平均法（MSA）进行求解;最后,结合案例验证了所提票价调整策略能够有效地平衡区段客流,降低旅客出行成本并在一定程度上提高客运收益。结果分析表明该票价调整策略能够为铁路客运管理部门优化票价体系、制定票价调整方案提供决策支持与方法指导。     

IBM全球商业服务部智慧商业服务线咨询经理、前上品折扣信息总监张鹏： 基于单品管理的全员营销

首先感谢能有机会参加本次零售CIO考察交流沙龙。今天跟大家分享的主题是：基于单品管理的全员营销。 大家都知道,营销是以客流为基础的,我们要把客流吸引到店里来,只有客流来了以后,才能通过营销手段做转化。现在受互联网的冲击很大,移动化的趋势也越来越明显,如何在现在这个信息化、移动化时代来完成营销的触达,这是我今天要讲的主要内容。     

基于神经网络的城市轨道交通短期客流预测研究

该文结合神经网络来研究城市轨道交通中短期客流预测问题。设计出了基于自回归神经网络的轨道交通客流预测模型、模型描述及其模型训练算法。通过matlab仿真实验来验证预测模型的性能,优于将最小二乘支持向量机与离散一维Daub4小波分析结合起来预测效果。     

基于神经网络的城市轨道交通短期客流预测研究

该文结合神经网络来研究城市轨道交通中短期客流预测问题。设计出了基于自回归神经网络的轨道交通客流预测模型、模型描述及其模型训练算法。通过matlab仿真实验来验证预测模型的性能,优于将最小二乘支持向量机与离散一维Daub4小波分析结合起来预测效果。     

短时公交客流的弦不变量预测方法

短时公交客流预测是智能公交系统动态调度的基础.文中根据短时公交客流数据特性,提出基于弦理论的短时公交客流预测方法,模拟弦结构建立弦不变量客流预测模型(SI-PFPM),并采用遗传算法优化SI-PFPM中各参数.提出基于动态时间弯曲距离的仿射传播(AP)聚类算法,对短时公交客流时间序列进行聚类分析.利用SI-PFPM预测聚类子集数据,并分析预测残差,验证SI-PFPM可以预测短时公交客流的假设成立.最后将SI-PFPM的预测性能与现有方法进行对比分析,验证SI-PFPM对短时公交客流预测的有效性.     

公交车乘客流量自动计数的P2DHMM技术研究

我国一些大中城市公交车运营管理采用了无人售票的刷卡系统管理,但是自动刷卡系统不能够满足车辆调度、运营规划等公共交通管理的要求,为了获取乘客流量的实际数据,开发高精度自动记数模型成为当前的热点和技术难点。我国现有乘客流量统计系统的精度在80%左右,造成精度问题的主要原因是在乘车高峰时段拥挤的状态背景条件下,一般模式识别算法和模型设计难以准确将单个人体识别和分割出来,为此,在充分调研的基础上,设计了采用普通数码摄像机和模板匹配方案,选择实验了先进的P2DHMM模型,模拟实验结果证实,该方法具有稳定的识别性能,为真实条件下的实验奠定了良好的基础。     

基于增强学习的网格化出租车调度方法

高度信息化的网格化城市管理可以为出租车运营优化提供新的实时动态乘客需求信息和车辆位置信息。以此为契机,针对城市出租车空驶率高和司乘匹配率低的问题,提出了一种网格化的出租车实时动态调度的增强学习控制方法。通过为出租车提供空驶巡游的动态最佳路线,新的控制方法旨在提高出租车的服务效率,并降低乘客的等待时间。首先,以城市单元网格为基础,明确出租车调度的关键问题;其次,以空驶路线的动态调整为控制手段,建立调度的增强学习模型;最后,给出求解模型的Q学习算法,并通过算例验证新调度方法的有效性。研究表明新方法可以有效提高司乘匹配率、增加总的出租车运营收入、减少乘客平均等车时间和减少总的出租车空驶时间。     

A model of check-in system management to reduce the security checkpoint variability

This article presents a model of dynamic management of the check-in desk system at an airport. The elaborated model enables the obtainment of time limits ensuring a uniformly distributed stream of passengers reporting to the security checkpoint on the basis of a schedule of check-in desk operations. The model also allows for the creation of a check-in desk operation schedule for an assumption of the longest acceptable time in queue and the maximum acceptable length of a queue. The model of dynamic management of the check-in desk system at an airport was implemented in Flexsim software. A significant advantage of the developed algorithms is that they consider the stream of passengers report for security control, which is generated by the check-in operating system. The developed model was verified and implemented for management of the check-in system at the Wrocław Airport. For input data obtained in the course of research conducted in 2014 (characteristics of the system and the stream of passengers reports - flight schedule), system efficiency measures were determined: the average queuing time for the check-in, number of work-hours of check-in operators and characteristics of the number of passengers reports for the security control. The developed model can be used in existing check-in systems, and it can also be used to evaluate the operation of a system being designed.     

基于时间序列的机场短时段值机客流量预测

在机场向数字化运营的转型过程中,为了实现航站楼内高效的运作以及资源的合理分配,从而对短时段的值机客流量的预测提出了更高的要求。通过对机场历史数据的统计和分析,结合航班的DOW特性,分析值机客流量的相关影响因素,以每小时的值机客流量为研究对象,构建基于时间序列的动态回归ARIMAX模型。实验结果表明,上述模型相对于传统的预测模型,预测精度更高,拟合效果更精确,有效地预测了航站楼内短时段的值机旅客人数,为航站楼内资源的动态分配和优化提供了不可或缺的决策支持。     

基于模糊控制的全数字键盘电梯群控系统

信息的完整决定着控制的优良,该文介绍了一种全部采用数字键盘的电梯群控模型,针对该模型,在乘客两种不同需求的基础上,推导了乘客的最长候梯时间和最大冲击数的计算公式,采用规则模糊推理,实现层站召唤的合理分配,仿真实验对常规电梯群控系统和全数字群控系统做了比较,结果表明,在正常客流模式下,乘客的乘梯时间和经受的冲击数较常规的群控系统有降低,特别是强度较轻的客流模式效果更为明显,为研究符合乘客利益的电梯群控系统奠定了基础.     

Detection of passenger flow on and off buses based on video images and YOLO algorithm

Bus passenger flow information is very important as a reference data for bus company line optimization, schedule scheduling basis, and passenger travel mode arrangement. With the development of image processing technology, it has become a current research trend to count passenger flow with the help of surveillance video of passengers getting on and off the bus. The specific research contents of this paper based on video image detection and statistics of passengers are as follows:(1) Collect head target image samples through a variety of ways, including 3960 positive head target samples and 4150 negative head target samples, which together constitute the head target feature database. (2) Established a head target detection model based on deep learning. First, the labeling of the head target training data set is completed. Then, after 15,000 iterations of model training, the YOLOv3 head target detection network model was obtained, with a recall rate of 92.12% and an accuracy rate of 89.71%. (3) A multi-target matching tracking algorithm based on the combination of Cam-shift and YOLOv3 is proposed. First, the Cam-shift algorithm is used to track the head target. Secondly, the head target tracking data and the YOLOv3 detection data are combined to solve the problem of drift during the tracking of the Cam-shift algorithm through the data association matching method based on the minimum distance, and then combined with the time constraint, a passenger location information judgment rule is proposed. Optimize the error and missed detection in the process of head target detection and tracking, and improve the reliability of passenger trajectory tracking. (4) A statistical algorithm for the detection of passengers getting on and off the bus is proposed. First, the trajectory of passengers in the bus boarding and disembarking area is analyzed, and a process for judging passengers’ boarding and boarding behavior is proposed. At the same time, a passenger position information judgment rule is proposed according to the different situations of whether there are new passengers or missing passengers, so as to optimize the problem of wrong detection and missing detection in the process of head target detection and tracking. (5) Finally, experiments are carried out in actual bus scenes and simulation scenes. The experiment proves that the statistical algorithm for the detection of passengers getting on and off the bus proposed in this paper has good detection, tracking and statistics effects in bus scenes and simulation scenes.

     

一种基于航空旅客行为的旅客细分模型

针对当前航空公司旅客细分工作不够精准的问题,在传统旅客划分模型的基础上,提出了一种旅客价值评价模型来提高旅客细分的准确度,该模型根据旅客的行为偏好对旅客进行细分,构建模型采用的方法为层次分析法,其中在使用层次分析法确定各参数权重时的关键在于如何构造完全一致的判断矩阵,提出了一种改进判断矩阵一致性的算法,使得一致性调整过程简单且结果有效。采用改进后的层次分析法确定旅客模型各参数的权重并通过旅客购票信息对旅客进行分类,为航空公司针对有着不同行为偏好的旅客制定相应的个性化服务以及营销策略提供了良好的参考依据。     

Passenger travel behaviour on Chinese high‐speed railways using machine learning based on revealed‐preference data

Existing studies on the selection of railway travel modes almost always compare high‐speed trains and regular‐speed trains. In this study, three high‐speed trains (Revival, Harmony, and Electric multiple unites (EMU)) on the Beijing–Shanghai line, and the first‐ and second‐class seats on these three trains are selected as research objects to analyse the travel selection behaviour of Chinese high‐speed train passengers. Three methods are selected to study this behaviour: the Support Vector Machine, Nested Logit model, and Multiple Logit model. The results of these three models are calibrated using LIBSVM software and STATA software and show that age, funding source, income level, and purpose of travel are major factors that affect Chinese railway passengers' choices on high‐speed railways. Finally, it is shown that the Support Vector Machine is the most accurate of these three methods, followed by the Nested Logit model. The results of this study can complement existing research on the travel selection behaviour of Chinese railway passengers and has important implications for enabling China's high‐speed rail operators to adjust their passenger transportation products.     

基于人工蜂群算法的轨道交通列车行车间隔优化

轨道交通运营组织作为轨道交通运营企业管理的核心,在降低企业运营成本、提升服务水平和旅客出行效率方面起着非常重要的作用。提出一种基于人工蜂群（ABC）优化算法的列车行车间隔优化策略,在考虑运营企业和旅客各自利益的基础上,以列车发车间隔为决策变量,建立旅客平均候车时间最小和列车等候时间最大的双目标非线性规划模型。采用ABC算法对模型进行优化求解,结合京津城际铁路某日不同时段客流基础数据进行仿真,实例验证了所提算法和模型的有效性。     

The Dynamic Marketing–Entrepreneurship Interface: A Creative Management Approach

In recent years increasing attention has been paid to the marketing – entrepreneurship interface in the management literature. This work is valuable in pointing out the need for a multidisciplinary perspective and a more holistic approach to research in the area. The present article argues that there is a need to further extend the paradigm and study the interface in a more dynamic context, than that which is provided by traditional economic approaches to marketing and entrepreneurship. This is done by using a creative management approach, which ties together economic, organizational and psychological mechanisms, to try to understand the dynamic marketing and entrepreneurship interface in a strategic management perspective. The creative management approach should be viewed as a meta-model which includes the economic model as a way of describing the outcome of the dynamic process.