| 392 | 0 | 4 |
| 下载次数 | 被引频次 | 阅读次数 |
【背景】机场时刻资源供不应求的问题主要基于世界机场航班时刻指南采用行政手段将机场航班时刻分配给航空公司，机场容量不能满足航空运输需求将会造成机场严重拥堵和航班延误。现有航班时刻配置研究大多针对一个机场的航班时刻进行单独配置，配置结果可能导致航班公司在两个机场获得的航班时刻不匹配进而无法安排航班。【目标】机场网络航班一体化配置中同时考虑航班两端时刻配置，并开发求解算法解决模型规模大而求解困难的问题，具有重要研究意义。【方法】建立了一个机场网络航班时刻配置模型，用于在网络层面管理机场航班时刻，并开发了基于深度强化学习（DQN）和邻域搜索（NS）相结合的NS-DQN算法用于模型求解，该算法利用DQN跳出邻域搜索过程的局部最优，从而提升求解速度和效果。【数据】采用中国大陆机场网络2023年7月24日至30日航班时刻配置进行模型的求解。【结论】与直接使用Gurobi求解器相比，NS-DQN算法可以在2.75 h内得到全局最优解，大幅减少了模型求解所需要的计算时间。
Abstract：[Background] Airport capacity that cannot meet the demand for air transportation can lead to severe congestion and flight delays. Currently, the issue of insufficient airport slot resources is mainly addressed through administrative means, following the Worldwide Airport Slot Guidelines, to allocate airport slots to airlines. Existing slot allocation studies often focus on individual airport slot allocation, which may result in mismatched slot allocations at two airports for an airline, making it impossible to schedule flights. [Objective] Airport network flight-slot integration considers the slot allocation at both ends of the flight, but faces challenges such as the large scale of the problem and difficulty in finding a solution. [Methods] This study establishes an airport network flight-slot allocation model for managing airport slots at the network level. To improve the model-solving efficiency,an NS-DQN algorithm that combines deep reinforcement learning(DQN) with neighborhood search(NS) was developed. The algorithm leverages DQN to escape the local optima in the neighborhood search process, thereby enhancing the algorithm's speed and effectiveness. [Data] The algorithm is applied to solve the airport network flight slot allocation model for China's mainland airports from July 24th to 30th, 2023. [Conclusions] The results show that compared to directly using the Gurobi solver, the NS-DQN algorithm can obtain the global optimal solution within 2.75 h, significantly reducing the computational time required for model solution.
[1]中国民用航空局.中国民航2022年8月份主要生产指标统计[EB/OL].(2022-09-13)[2024-08-05].https://www.caac. gov. cn/XXGK/XXGK/TJSJ/202209/P020220920312815593554.pdf.
[2]中国民用航空局.民航航班时刻管理办法[S].北京：中国民用航空局, 2018.
[3]王艳军,水笑雨,王梦尹.机场航班时刻资源管理研究进展[J].北京航空航天大学学报, 2024, 50(4):1065-1076.WANG Yanjun,SHUI Xiaoyu,WANG Mengyin. Research progress on airport slot allocation[J]. Journal of Beijing University of Aeronautics and Astronautics, 2024,50(4):1065-1076.
[4] DANIEL D, Wang Y J. Slot allocation in a multi-airport system under flying time uncertainty[C]//International Workshop on ATM/CNS 2022. Tokyo:IWAC, 2022:hal-03852039.
[5] LIU C, LIAO C, HANG X, et al. Slot allocation in a multi-airport system under flying time uncertainty[J].Transactions of the Japan Society for Aeronautical and Space Sciences, 2024, 67(3):127-135.
[6] XU M, WANG M, WANG Y, et al. Robust estimation of airport declared capacity[C]//2022 IEEE 25th International Conference on Intelligent Transportation Systems(ITSC). Macau:IEEE, 2022:3807-3812.
[7] ZHAO X, WANG Y, LI L, et al. A queuing network model of a multi-airport system based on point-wise stationary approximation[J]. Aerospace, 2022, 9(7):390.
[8] WANG M, WANG Y J, LIAO C H, etal. Slot allocation for a multiple airport system:equity and efficiency[C]//Fifteenth UAS-Europe Air Traffic Magement Research and Delevopmen Seminar. Savannah:ATM, 2023:hal-04119202.
[9] FENG H, HU R, ZHANG J, et al. An integrated slot allocation model for time-space-dimensional noise reduction[J]. Transportation Research Part D:Transport and Environment, 2023, 121:103845.
[10]水笑雨,王艳军,王子明.考虑机场公平性的机场群航班时刻分配[J].航空学报, 2023, 44(08):165-181.SHUI Xiaoyu, WANG Yaijun, WANG Ziming, et al. Airport cluster flight schedule allocation considering fairness among airports[J]. Acta Aeronautica et Astronautica Sinica, 2023, 44(08):165-181.
[11] ZOGRAFOS K G, SALOURAS Y, MADAS M A. Dealing with the efficient allocation of scarce resources at congested airports[J]. Transportation Research Part C:Emerging Technologies, 2012, 21(1):244-256.
[12] BOYAC?B, ZOGRAFOS K G, GEROLIMINIS N. An integrated optimization-simulation framework for vehicle and personnel relocations of electric carsharing systems with reservations[J]. Transportation Research Part B:Methodological, 2017, 95:214-237.
[13] RIBEIRO N A, JACQUILLAT A, ANTUNES A P, et al.An optimization approach for airport slot allocation under IATA guidelines[J]. Transportation Research Part B:Methodological, 2018, 112:132-156.
[14] JACQUILLAT A, ODONI A R, WEBSTER M D. Dynamic control of runway configurations and of arrival and departure service rates at JFK airport under stochastic queue conditions[J]. Transportation Science, 2017, 51(1):155-176.
[15] FAIRBROTHER J, ZOGRAFOS K G, GLAZEBROOK K D. A slot-scheduling mechanism at congested airports that incorporates efficiency, fairness, and airline preferences[J]. Transportation Science, 2020, 54(1):115-138.
[16] BIROLINI S, JACQUILLAT A, SCHMEDEMAN P,et al. Passenger-centric slot allocation at schedule-coordinated airports[J]. Transportation Science, 2023, 57(1):4-26.
[17] WANG Y J, SHUI X Y, WANG M Y, et al. Slot allocation for a multi-airport system and its impact on the equity among the airports and airlines[EB/OL].(2023-01-19)[2024-08-01]. http://dx.doi.org/10.2139/ssrn.4330306.
[18] WANG Y, WANG M, XU W, et al. Secondary trading of airport slots:issues and challenges[J]. Chinese Journal of Aeronautics, 2023, 36(12):1-12.
[19] WANG Y, LIU C, WANG H, et al. Slot allocation for a multiple-airport system considering airspace capacity and flying time uncertainty[J]. Transportation Research Part C:Emerging Technologies, 2023, 153:104185.
[20] BERTSIMAS D, LULLI G, ODONI A. An integer optimization approach to large-scale air traffic flow management[J]. Operations Research, 2011, 59(1):211-227.
[21] COROLLI L, LULLI G, NTAIMO L. The time slot allocation problem under uncertain capacity[J]. Transportation Research Part C:Emerging Technologies, 2014, 46:16-29.
[22] PELLEGRINI P, BOLI?T, CASTELLI L, et al. SOSTA:an effective model for the Simultaneous Optimisation of airport SloT Allocation[J]. Transportation Research Part E:Logistics and Transportation Review,2017, 99:34-53.
[23] BENLIC U. Heuristic search for allocation of slots at network level[J]. Transportation Research Part C:Emerging Technologies, 2018, 86:488-509.
[24]吴慎之,王子明,杭旭.机场群航班时刻表优化方案研究[J].交通运输工程与信息学报, 2022, 20(4):136-146.WU Shenzi, WANG ziming, HANG Xu, etal. Optimization schemes for schedules intervention in a multi-airport system[J]. Journal of Transportation Engineering and Information, 2022, 20(4):136-146.
[25]吴慎之.区域内多机场航班时刻协同优化[D].南京：南京航空航天大学, 2022.WU Shenzi. Collaborative optimization of flight time at multiple airports in the region[D]. Nanjing:Nanjing University of Aeronautics and Astronautics, 2022.
[26]聂建雄，刘畅，王艳军.机场群容量资源战略一体化配置方法[J].交通运输工程与信息学报，2023, 21(4):115-128.NIE Jianxiong, LIU Chang, WANG Yanjun. Integrated method for strategically allocating capacity resources in a multiple airport system[J]. Journal of Transportation Engineering and Information, 2023, 21(4):115-128.
[27] RIBEIRO N A, JACQUILLAT A, ANTUNES A P. A large-scale neighborhood search approach to airport slot allocation[J]. Transportation Science, 2019, 53(6):1772-1797.
[28] LEE I, SIKORA R, SHAW M J. A genetic algorithmbased approach to flexible flow-line scheduling with variable lot sizes[J]. IEEE Transactions on Systems,Man, and Cybernetics, Part B(Cybernetics), 1997, 27(1):36-54.
[29] ARAGON-GóMEZ R, CLEMPNER J B. Traffic-signal control reinforcement learning approach for continuoustime Markov games[J]. Engineering Applications of Artificial Intelligence, 2020, 89:103415.
[30] BANGUI H, BUHNOVA B. Recent advances in machine-learning driven intrusion detection in transportation:survey[J]. Procedia Computer Science, 2021, 184:877-886.
[31] CHEN X, GUO X, MENG J, et al. Research on ATO control method for urban rail based on deep reinforcement learning[J]. IEEE Access, 2023, 11:5919-5928.
[32] SHU S, ZHAO J, HAN Y. Signal timing optimization for transit priority at near-saturated intersections[J]. Journal of Advanced Transportation, 2018:8502804.
[33] DING Y, WANDELT S, WU G, et al. Towards efficient airline disruption recovery with reinforcement learning[J]. Transportation Research Part E:Logistics and Transportation Review, 2023, 179:103295.
[34] ALCARAZ J J, LOSILLA F, CABALLERO-ARNALDOS L. Online model-based reinforcement learning for decision-making in long distance routes[J]. Transportation Research Part E:Logistics and Transportation Review, 2022, 164:102790.
[35] AKHTER S, NURUL M, JAFOR S. Modeling ant colony optimization for multi-agent based intelligent transportation system[J]. International Journal of Advanced Computer Science and Applications, 2019, 10(10):277-284.
[36] BORG J P, ZITOMER D H. Dual-team model for international service learning in engineering:remote solar water pumping in Guatemala[J]. Journal of Professional Issues in Engineering Education and Practice, 2008, 134(2):178-185.
[37] XU M, AN K, VU L H, et al. Optimizing multi-agent based urban traffic signal control system[J]. Journal of Intelligent Transportation Systems, 2019, 23(4):357-369.
[38] XIONG C, ZHU Z, CHEN X, et al. Optimal travel information provision strategies:an agent-based approach under uncertainty[J]. Transportmetrica B:Transport Dynamics, 2018, 6(2):129-150.
[39] WANG J, WU J, CHE Y. Agent and system dynamicsbased hybrid modeling and simulation for multilateral bidding in electricity market[J]. Energy, 2019, 180:444-456.
基本信息:
DOI：10.19961/j.cnki.1672-4747.2024.09.008
中图分类号:V355;TP18;F562
引用信息:
[1]胡浩然,王艳军,范晰桐.基于强化学习和邻域搜索的机场网络航班时刻配置算法[J].交通运输工程与信息学报,2025,23(02):161-170.DOI:10.19961/j.cnki.1672-4747.2024.09.008.
基金信息:
国家自然科学基金民航联合基金重点项目：机场群航班时刻资源优化配置技术与方法研究项目（U2033203）