【作者】 王锋；

【导师】 王竹荣；

【作者基本信息】 西安理工大学 ， 计算机应用技术， 2019， 硕士

【摘要】 随着国内民航业迅猛发展以及民航相关政策的颁布,近年来航空出行受越来越多旅客的青睐。但是由于航空出行受很多因素干扰,经常会使航班发生不可预期的延误,中断等,给航空公司以及旅客带来巨大的经济损失和不便。如何处理不正常航班,建立与实际情况相符的航班恢复数学模型,设计出更高效的求解算法,使航空公司因不正常航班的影响损失最小,受扰旅客总延误时间最少,已经成为各个航空公司以及学术界关注的热点问题。本文研究从国内外航空公司对不正常航班处理的标准出发,对受扰航班恢复问题和航班--旅客整体恢复问题进行深入的分析和研究。为了处理不正常航班恢复问题,即机场因为天气原因临时关闭而引起的航班中断,从不同的航班恢复场景出发,同时考虑决策变量和场景因素,构建了能处理恢复时间确定和不确定两种情况的航班恢复数学模型。首先建立具有确定恢复时间的优化数学模型M1,其目标函数是最小化航班交换,航班取消和航班延误的花费。其次,对于恢复时间不确定的情况,建立航班恢复数学模型M2,用于对恢复时间确定情况下获得的结果进行随机优化,包括航班重排,航班延误及宵禁处理,使得因恢复时间不确定而引起的总花费最小。再次,通过模型M1和M2的加权组合建立航班恢复问题的随机模型M,来解决航班恢复时间确定以及恢复时间不确定两种情况。最后,将实际数据用于验证所提模型,借助Lingo11.0来对所提问题模型进行优化求解,来证明所建模型的可行性和正确性。为了处理航班--旅客整体恢复问题,所提模型以最小化航班和旅客的恢复成本为第一优化目标,借助经济学投资组合的概念和可信性理论,以旅客和航班的最小组合风险为第二优化目标。模型考虑航班取消,航班延误,航班置换,旅客迁转,乘客座位降级,以及宵禁等因素,在航班发生延误情况下,重点分析了航班和旅客的优先级恢复策略以及恢复过程中航班和旅客的组合风险。为合理并高效求解所建立航班--旅客恢复模型,根据有关知识信息设计出一种符合所建立模型的启发式优化算法。使用某机场历史航班数据作为算例,对所提数学模型的正确性和有效性进行验证。实验结果表明优化模型相对于传统经验恢复方法使总延误时间和因航班取消而受扰旅客数量都有所下降。更多还原

【Abstract】 With the rapid development of the domestic civil aviation industry and the promulgation of relevant civil aviation policies,air travel has been favored by more and more passengers in recent years.However,due to many factors,air travel often causes unpredictable delays,interruptions which bring huge economic losses and inconvenience to airlines and passengers.How to deal with abnormal flights,establish a mathematical model of flight recovery that is consistent with the actual situation,and design a more efficient algorithm to solve the problem,so as to minimize the loss caused by abnormal flights and minimize the total delay time of the affected passengers,has become a hot issue of concern for various airlines and academia.This paper studies the domestic and international airlines’ standards for the handling of abnormal flights,and conducts in-depth analysis and research on the problem of disturbed flight recovery and flight-to-passenger recovery.For different scenarios of flight recovery problems,a mathematical model for flight recovery is constructed while considering the factors of decision variables and scenarios.The model deals with the problem of optimized objective functions and the constraints based on actual requirements.The mathematical optimization model Ml with determining recovery time is established to minimize the objective function of aircraft swapping,flight cancellation and flight delay cost.In the case of uncertain recovery time,the mathematical model M2 of flight recovery was established to randomly optimize the results obtained under the condition of uncertain recovery time,including flight rescheduling,flight delay and curfew treatment,so as to minimize the total cost caused by uncertain recovery time.For actual flight recovery time,the stochastic model M of the flight recovery problem is established by the combination of the constructed model Ml and M2.Finally,the actual data is used to verify the proposed model,and lingo11.0 is used to optimize and solve the proposed problem model,so as to prove the feasibility and correctness of the model.In order to deal with the flight-passenger overall recovery problem,the model established in this paper takes minimizing the recovery cost of flights and passengers as the first optimization objective,and takes the minimum portfolio risk of passengers and flights as the second optimization objective with the help of the concept of economic portfolio and the credibility theory.The model considers factors such as flight cancellation,flight delay,flight replacement,passenger transfer,passenger seat degradation,and curfew,etc.In the case of flight delay,the priority recovery strategy of flight and passenger and the combination risk of flight and passenger in the recovery process are analyzed emphatically.In order to solve the flight-passenger recovery model reasonably and efficiently,a heuristic optimization algorithm was designed according to the relevant knowledge information.The historical flight data of an airport is used as an example to verify the correctness and validity of the proposed mathematical model.The experimental results show that the optimization model has a decrease in the total delay time and the number of passengers affected by flight cancellation compared to the traditional experience recovery method.更多还原