基于免疫进化算法的航班过站时间优化模型

针对当前航班延误在航班链中的波及的特性,通过研究航班过站时间对航班延误的影响的内在机理,在基于免疫进化算法基础上,没有增加过站时间的总数,调整过站时间来减少航班的延误波及的优化模型.在保持航班的飞行任务和机组安排不变的情况下,只对航班时刻做出了轻微的修改.根据国内某大型枢纽机场的数据进行仿真实验,调整后的计划可以很好的改善延误的情况.     

基于贝叶斯网络的航班过站时间分析与延误预测

从航班延误链式波及的角度出发,分析了影响航班过站时间的多种因素,建立了贝叶斯网络模型,模型能够清晰地反映多种因素对下游航班过站时间的影响。提出了基于贝叶斯网络参数估计的航班延误预测算法,当航班发生起飞延误时能够预测下游航班的起飞时间和延误状况。对算法进行了实现,并利用实际航班数据进行仿真,结果表明了该算法有比较高的预测准确率。     

基于状态空间模型序号编码进化算法的航班优化调度

多跑道航班进离港优化调度是将某一时间窗内进离港航班看作一个整体，对进离港航班进行统一优化排序，属于典型的组合优化问题。求解组合优化问题时，采用序号编码比二进制编码和实数编码等方式更直接、更方便。本文提出一种基于状态空间模型序号编码进化算法（OSEA），通过构造状态进化矩阵等操作来实现变异算子的功能，简化了遗传操作，并研究其在航班进离港优化调度中的应用。仿真实验表明：OSEA算法与先到先服务（FCFS）航班排序相比，航班总延误损失降低32.88％，且运算速度更快。     

时间有色佩特里网在航班进离港中应用

为了解决目前和以后越来越严重的空中交通拥挤问题,文中对终端区空中交通流量管理中航班进离港部分连续航班动态飞机排序问题进行研究.提出了基于时间有色佩特里网(TCPN)的航班进离港的调度方法,用MATLAB软件中的M函数编写了算法程序,然后进行了仿真.仿真结果表明,该方法选用的算法是合理的,而且可以明显的减少航班总的延误,优化效率高,算法简单可行,而且更符合实际的要求.该方法在一定程度上可以帮助空管人员进行调度决策,从而更加快速有序地对飞机进离港进行调度,对目前空管优化调度,减少延误等情况的处理具有一定的参考研究价值.     

离港航班延误的动态预测研究

现有航班延误预测方法大多是对航班延误进行非动态预测,不能随着航班数据的不断增加而有效更新,进而使得预测结果的实时性较差.因此,提出一种基于时间序列的航班延误动态预测方法.首先,利用小波分解技术对延误时间序列进行平稳化处理;然后,采用自回归滑动平均(ARMA)模型对离港航班延误进行预测;最后,利用支持向量机模型对预测结果进行修正,得到最终的离港航班延误预测值.结果表明,该动态预测方法能较好地提高预测精度.     

一种面向航班准点率保障的航班协同调度评价模型

为了保障航班的准点率,对航班调度过程中空管、航空公司和机场之间的协同状况进行评价,构建一种面向航班准点率保障的航班协同调度评价模型。基于某一时间段内机场航班进港时间平均偏离值、出港时间平均偏离值、保障完成时间平均偏离值、目标撤轮挡时间准确率和计算起飞时间准确率,建立航班协同调度的指标体系;采用关联规则挖掘的方法确定各指标对航班协同调度及航班延误影响的阈值;结合模糊层次分析法和综合评价法,对各单位在航班调度过程中的协同状况给出量级评价。实验表明:该模型能够科学有效地评价各单位在航班调度过程中的协同状况,对于优化航班保障、提高航班放行正常率具有重要意义。     

航班延误分布式检测与仿真

关于航班延误问题,已成为是困扰民航部门和广大旅客的热点,涉及多个环节和多种影响因素的航班延误状态检测是难点.针对某天某时刻航班实际运行发生的情况,将生物免疫系统机制与机场航班运行机制联系起来,为提高检测准确率,提出了一种分布式人工免疫模型的机场航班延误实时检测方法,采用分布式检测模型的特点,用各周次训练出来的检测器联合实时对下一时刻的机场离港航班延误状态进行检测,仿真结果表明方法不仅能够实时准确地检测机场离港航班状态,而且能够准确预测下一时间段累计延误航班数量,为航班延误预警提供决策支持.     

航班延误传播分析方法研究

为避免以往主要针对局部机场网络开展航班延误传播分布和现实数据观察等表象研究,发掘大面积航班延误的要素影响规律,提出了基于航班频率偏好的航空演化网络模型,并将经典传染病SIS传播分析方法与航班延误传播过程相结合,构建基于航空演化网络的延误传播模型,以及延误传播概率及其与影响要素之间的分析方法.基于东航实际运营数据,得到延误传播概率与航线距离、航班频率、实际缓冲时间、进离港总延误时间4个影响因素之间的回归量化关系,并由此给出了控制航班频率、避免频繁起落、降低航线繁忙程度、增大缓冲时间的结论.     

基于贝叶斯网络的航班延误波及研究

在民航业内,航班延误波及问题一直是影响航班延误的一个主要因素。基于贝叶斯网络（BN）,讨论了在繁忙的枢纽机场,其航班延误的波及问题。在实验中使用的数据,皆来自国内某大型航空公司的真实记录。通过建立延误波及模型和贝叶斯网络模型,探讨了相关航班中,进港延误和航班取消对离港延误的影响。学习的结果显示了进港延误（Arrival-delay）对离港延误（Departure-Delay）的波及现象的存在;以及波及现象在不同情况下的程度不同;相应的可采取的应对方式亦不相同。其中航班取消是一种釜底抽薪的应对方法,可以在一定程度上削弱上述条件下的延误波及,其削弱程度与航班取消的架次有关。基于该研究可以在机场发生大规模延误时,提供一个基本的参考。     

航班延误树的构造与波及分析

由于一架飞机在一天中要执行多个航班,各航班之间存在前后衔接关系,因此,一个航班的延误会波及到下游许多其它航班。重点研究飞机和机组资源对于航班延误与波及的影响,给出延误树的生成过程,通过初始航班延误的触发,动态建立以该航班为根结点的航班延误树,并根据统计结果给出相关量值。实例分析了初始航班延误发生的时刻、持续时间与波及的程度,以期辅助优化飞机与机组排班,减少航班延误。     

多机场地面等待策略的建模与仿真

针对我国空中交通流量管理中出现的大规模航班延误情况，根据多个机场实际起飞和降落容量的约束，建立了一种考虑航班取消的多机场地面等待策略数学模型，并提出求解该模型的有效算法。在所提模型与算法的基础上，选择我国三大机场实际数据进行仿真。对比了取消航班与不取消航班两种情况结果，仿真结果验证了模型与方法的有效性与实用性，给出了最优排序的航班时刻表，从而支持了关键航班的运行，减少了航班总延误。     

Applying digital twins for the management of information in turnaround event operations in commercial airports

The aerospace sector is one of the many sectors in which large amounts of data are generated. Thanks to the evolution of technology, these data can be exploited in several ways to improve the operation and management of industrial processes. However, to achieve this goal, it is necessary to define architectures and data models that allow to manage and homogenise the heterogeneous data collected. In this paper, we present an Airport Digital Twin Reference Conceptualisation’s and data model based on FIWARE Generic Enablers and the Next Generation Service Interfaces-Linked Data standard. Concretely, we particularise the Airport Digital Twin to improve the efficiency of flight turnaround events. The architecture proposed is validated in the Aberdeen International Airport with the aim of reducing delays in commercial flights. The implementation includes an application that shows the real state of the airport, combining two-dimensional and three-dimensional virtual reality representations of the stands, and a mobile application that helps ground operators to schedule departure and arrival flights.     

Reducing airport gate blockage in passenger aviation: Models and analysis

Commercial flights are typically assigned to an arrival gate at their destination station (airport) prior to their departure from their origin station. Although the gate is scheduled to be available when the flight arrives, this is not always the case in practice. Due to variability in departure and flight times, the arriving flight might arrive early, the previous flight departing from the gate might depart late, or both. When a flight arrives at its scheduled gate but has to wait because the preceding aircraft is still occupying that gate, we refer to this as gate blockage. Gate blockage can have many negative impacts, including passenger delays, missed connections, and increased fuel burn. Our research is focused on incorporating the inherent stochasticity of the system into the planning process to reduce the prevalence and impact of gate blockage. Specifically, we formulate an optimization problem to assign flights to gates so as to minimize the expected impact of gate blockage. We use historical data to predict delay distributions and conduct experiments to assess both the computational tractability of our approach and its potential for improvement in solution quality over existing approaches.     

考虑任务均衡的加油车动态调度问题

为了提高机场资源利用率以及航班的正点率,合理调度机场加油车为航班进行燃油加注服务势在必行。考虑到航班实际时刻的不可预知性,需建立动态规划时间窗,为窗口依据航班预计时刻捕捉到的航班构建机场车辆调度模型,利用自适应分支定价算法求解模型,从而配置车辆与人员,规划车辆的行驶路线,衔接任务,实现车辆行驶时间最短,加油员工作量均衡的目的。在华北某机场真实数据上进行仿真实验,结果表明,自适应分支定价算法优化下的车辆行驶时间与加油员工作量的标准差分较节约算法分别减小1.38%和7.47%,验证了算法的优势,且算法在机场其他地面服务规划问题中同样适用。     

基于松弛算法的停机位分配优化方法

针对机场近机位资源紧缺及实际航班到离港时间偏离计划时间对停机位分配所造成的扰动，提出了在同机位相邻航班间加入缓冲时间的停机位分配调度方法。首先，建立了以机位空闲时间、远机位占用时间最小为目标的鲁棒性停机位分配模型；然后，设计了一种基于双目标的拉格朗日松弛优化算法，并使用次梯度算法求解拉格朗日松弛算法中的对偶问题。基于国内某枢纽机场运行数据的仿真结果表明，所提方法的优化方案和原始机位分配方案相比，机位使用量和机位空闲时间分别降低了15.79%、7.56%，机位占用率提高了18.72%，并且冲突率降低到3.57%，达到了有效提高停机位利用率与鲁棒性的目的。     

连续航班延误与波及的贝叶斯网络分析

针对空运系统日益严重的航班延误,尝试将贝叶斯方法应用于航班数据分析,重点考虑同一飞机飞行连续航班的情况。借助Netica软件包,建立贝叶斯网络模型。通过贝叶斯网络推理进行连续航班延误波及分析,并用实际航班数据进行测试。结果表明,概率统计意义下,模型能够清晰反映连续航班延误原因分布、过站时间差分布和按时间段的延误波及情况。     

The effect of four-day round trip flights over 10 time zones on the circadian variation of salivary melatonin and Cortisol in airline flight attendants

The effects of four-day round flights (Helsinki-Los Angeles-Seattle-Helsinki) were studied on the circadian rhythms of salivary melatonin (MT) and Cortisol (COR) in 35 flight attendants. The mean age of the subjects was 33 ± 7 years (median 34, range 21-50). Five 24 h profiles of unstimulated saliva were collected at 2 h intervals (except at 04:00) before, during, and after the four day flight. Salivary MT and COR were determined by radioimmunoassay. Both MT and COR exhibited a clear circadian rhythm with acrophases before the flight at 03:03 (MT) and 09:08 (COR). Two days after the westward flight from Helsinki to Los Angeles, the MT rhythm (circadian acrophase) had delayed 4 h 51 min and the COR rhythm 3 h 55 min compared to the control day before the flight. Two days later, during the last day in the USA, the MT rhythm had delayed 5 h 59 min and the COR rhythm 5 h 29 min as compared to the situation before the flight. After four days of the eastward flight from Seattle to Helsinki, the circadian acrophase of MT was still 1 h 35 min delayed compared to the control day before the westward flight. The results indicate that the restitution time of five days at the home base is on the average proper for recovery, if a four day round flight over 10 time zones takes four days or less. The resynchronization rate of salivary hormones after westward, outgoing flights is faster than the resynchronization rate after the eastward return flights.     

Visual Analytics for Exploring Local Impact of Air Traffic

The environmental and noise impact of airports often causes extensive political discussion which in some cases even lead to transnational tensions. Analyzing local approach and departure patterns around an airport is difficult since it depends on a variety of complex variables like weather, local and general regulations and many more. Yet, understanding these movements and the expected amount of flights during arrival and departure is of great interest to both casual and expert users, as planes have a higher impact on the areas beneath during these phases. We present a Visual Analytics framework that enables users to develop an understanding of local flight behavior through visual exploration of historical data and interactive manipulation of prediction models with direct feedback, as well as a classification quality visualization using a random noise metaphor. We showcase our approach using real world data from the Zurich International Airport region, where aircraft noise has led to an ongoing conflict between Germany and Switzerland. The use cases, findings and expert feedback demonstrate how our approach helps in understanding the situation and to substantiate the otherwise often subjective discourse on the topic.