交通 | 思虑供需交互下的航空网络优化问题
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://p3-sign.toutiaoimg.com/tos-cn-i-qvj2lq49k0/bf69f8c4435d4a3a8777bc87a23b1013~noop.image?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1725614950&x-signature=i4ZwxfNgHOaEbPwj5cO0MCmCAhY%3D" style="width: 50%; margin-bottom: 20px;"></div>
<h1 style="color: black; text-align: left; margin-bottom: 10px;">编者按:</h1>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">本文提出了一种<span style="color: black;">包括</span>供需交互<span style="color: black;">功效</span>的航空网络规划模型 (ANPSD),该模型<span style="color: black;">同期</span><span style="color: black;">思虑</span>了航线<span style="color: black;">选取</span>、航班频率和机队<span style="color: black;">构成</span>等问题,还<span style="color: black;">捉捕</span>了航空<span style="color: black;">机构</span>的供应和乘客<span style="color: black;">需要</span>之间的相互依赖关系。作者将<span style="color: black;">需要</span>实证函数与 ANPSD 模型相结合,<span style="color: black;">研发</span>了一种名为 2 ECP 的精确割平面算法来<span style="color: black;">处理</span>此混合整数非凸优化模型,并针对一家欧洲航空<span style="color: black;">机构</span>的网络进行案例分析。</span></p>
<h1 style="color: black; text-align: left; margin-bottom: 10px;">1. 引言</h1>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">在过去的几十年里,航空业在<span style="color: black;">全世界</span>范围内高速发展,但<span style="color: black;">同期</span><span style="color: black;">亦</span>面临着<span style="color: black;">各样</span><span style="color: black;">外边</span>冲击所带来的<span style="color: black;">需要</span>波动方面的挑战。本文在战略层面提出了一种<span style="color: black;">思虑</span>供需交互<span style="color: black;">功效</span>的航空网络优化模型 ANPSD (Airline Network Planning with Supply and Demand interactions)。它<span style="color: black;">首要</span>从人口、地理、经济等<span style="color: black;">方向</span>出发,利用两<span style="color: black;">周期</span>最小二乘法得到了一个实证模型来估计乘客<span style="color: black;">需要</span>。其次本文在<span style="color: black;">思虑</span>路线<span style="color: black;">选取</span>,航班频率以及机组<span style="color: black;">形成</span>等问题的<span style="color: black;">基本</span>上,融入乘客<span style="color: black;">需要</span>函数,从而<span style="color: black;">研发</span>出一种新型的航空网络优化模型。<span style="color: black;">文案</span>证明了此模型的非线性与非凸性后,运用一种基于<span style="color: black;">外边</span>近似的精确割平面算法来对此进行求解,结果<span style="color: black;">显示</span>此算法能在较短的时间内返回比基准更<span style="color: black;">优秀</span>的<span style="color: black;">方法</span>。最后,作者将该模型应用于意大利航空<span style="color: black;">机构</span>的大陆网络,并从管理的<span style="color: black;">方向</span><span style="color: black;">供给</span>了关于<span style="color: black;">怎样</span>平衡相互竞争的网络规划<span style="color: black;">目的</span>的战略见解。</span></p>
<h1 style="color: black; text-align: left; margin-bottom: 10px;">2. 模型描述</h1>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://p3-sign.toutiaoimg.com/tos-cn-i-qvj2lq49k0/c4192b5fec10416da488ad7243e632b0~noop.image?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1725614950&x-signature=oSZUbymPLpaD0Oc9u%2BhUzUsg17s%3D" style="width: 50%; margin-bottom: 20px;"></div>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">如Fig. 1所示,该建模框架<span style="color: black;">包含</span>两个<span style="color: black;">重点</span>元素:<span style="color: black;">需要</span>模型和优化模型。</span></p>
<h1 style="color: black; text-align: left; margin-bottom: 10px;">2.1 <span style="color: black;">需要</span>模型</h1>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"><span style="color: black;">需要</span>模型将航班时刻表、机票价格、客流和社会经济变量的历史数据<span style="color: black;">做为</span>输入。 它<span style="color: black;">运用</span>两<span style="color: black;">周期</span>最小二乘法和工具变量来估计<span style="color: black;">每一个</span>航段的每月乘客<span style="color: black;">需要</span>,其中<span style="color: black;">包含</span>直达和转机乘客,以<span style="color: black;">处理</span>内生性问题。(1)式是作者得出的<span style="color: black;">需要</span>函数,其<span style="color: black;">捉捕</span>了乘客<span style="color: black;">需要</span>与服务频率和网络结构<span style="color: black;">体积</span>(由辐条数量<span style="color: black;">表率</span>)的依赖关系。</span></p>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://p3-sign.toutiaoimg.com/tos-cn-i-qvj2lq49k0/9e2306ce1b1b440f8425d7d63b5372b3~noop.image?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1725614950&x-signature=6QjsveCeIOapuekRIVowy%2BiO40A%3D" style="width: 50%; margin-bottom: 20px;"></div>
<h1 style="color: black; text-align: left; margin-bottom: 10px;">2.2 优化模型</h1>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">该航空网络优化决策模型 ANPSD <span style="color: black;">思虑</span>了<span style="color: black;">包含</span>航线<span style="color: black;">选取</span>、航班频率和机队<span style="color: black;">形成</span>等<span style="color: black;">原因</span>,站在<span style="color: black;">拥有</span>轴辐式 (Hub-and-spoke) 结构的单个航空<span style="color: black;">机构</span>的视角,使其利润最大化。</span></p>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://p3-sign.toutiaoimg.com/tos-cn-i-qvj2lq49k0/04f713ad53984133aa5cb59d2954b836~noop.image?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1725614950&x-signature=QZ9C%2BWBjPykro8sf90at44P%2FhA4%3D" style="width: 50%; margin-bottom: 20px;"></div>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://p3-sign.toutiaoimg.com/tos-cn-i-qvj2lq49k0/e315985e868f4e7881a7c605c624d398~noop.image?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1725614950&x-signature=D9VoPk5JthkMy7o%2FMFI5lFMNitU%3D" style="width: 50%; margin-bottom: 20px;"></div>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://p3-sign.toutiaoimg.com/tos-cn-i-qvj2lq49k0/64b0885ca05e45f8a5d6afe03e894c51~noop.image?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1725614950&x-signature=J1pRRzRGU%2BLZ%2FgQv2rc927mdHpQ%3D" style="width: 50%; margin-bottom: 20px;"></div>
<h1 style="color: black; text-align: left; margin-bottom: 10px;">3. 求解算法</h1>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">在约束 (11) 的<span style="color: black;">功效</span>下,经过计算与证明,此模型为非线性非凸问题。在先前的<span style="color: black;">科研</span>中,有学者运用 因子<span style="color: black;">纠正</span>了凸函数的基于梯度的线性近似以<span style="color: black;">处理</span>非凸性问题,此<span style="color: black;">办法</span>被<span style="color: black;">叫作</span>为 ECP (extended cutting-plane)。在本文中,作者<span style="color: black;">经过</span>一组<span style="color: black;">外边</span>近似半超平面替换约束 (11) 来<span style="color: black;">处理</span>主问题,每次迭代后都会生成两个新的半超平面,并更新上限和下限,<span style="color: black;">最后</span>收敛到 ANPSD 的最优解。作者<span style="color: black;">按照</span>两个离散化基准<span style="color: black;">评定</span> 2 ECP 的计算性能:Convex combination (CC) 与 Logarithmic branching convex combination (LOG) 基准。CC<span style="color: black;">办法</span>直接用一组分段线性超平面来近似非线性函数。LOG<span style="color: black;">办法</span>将非线性函数逼近为离散点的凸组合,此<span style="color: black;">办法</span>关键点在于其用许多二元变量和约束<span style="color: black;">要求</span>重新表述了凸组合,这些变量和约束<span style="color: black;">要求</span>随或多面体的数量呈对数增长。数值实验<span style="color: black;">显示</span>,2 ECP 算法与基于离散化和线性化的最新基准相比,实现了更强、更稳定和更一致的性能。 <span style="color: black;">另外</span>,它<span style="color: black;">亦</span>是一种精确<span style="color: black;">办法</span>,它在每次迭代时<span style="color: black;">供给</span>最优性的<span style="color: black;">处理</span><span style="color: black;">方法</span>,而不是仅生成可行的<span style="color: black;">处理</span><span style="color: black;">方法</span>。</span></p>
<h1 style="color: black; text-align: left; margin-bottom: 10px;">4. 案例分析</h1>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://p3-sign.toutiaoimg.com/tos-cn-i-qvj2lq49k0/870eca258d684834868b8692ae5388f1~noop.image?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1725614950&x-signature=MNZPous1jINTZm8RtyeblS08LTI%3D" style="width: 50%; margin-bottom: 20px;"></div>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://p3-sign.toutiaoimg.com/tos-cn-i-qvj2lq49k0/9e8a5c39c72d47e69c90c19bb0b1d60f~noop.image?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1725614950&x-signature=wqJtiWqCF5FxjDPj0lvvWPXhU%2Fk%3D" style="width: 50%; margin-bottom: 20px;"></div>
<h1 style="color: black; text-align: left; margin-bottom: 10px;">5. 总结与展望</h1>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">本文<span style="color: black;">研发</span>了一种新型的建模和算法框架来优化航空网络,该模型与忽略供需相互<span style="color: black;">功效</span>的<span style="color: black;">办法</span>相比,更能够<span style="color: black;">捉捕</span>服务频率、网络结构和乘客<span style="color: black;">需要</span>之间的相互<span style="color: black;">功效</span>。<span style="color: black;">由于</span>该模型强调了<span style="color: black;">区别</span>网络规划优化决策、航线<span style="color: black;">选取</span>、飞行频率和机队<span style="color: black;">构成</span>的重要性,<span style="color: black;">因此</span>其<span style="color: black;">能够</span><span style="color: black;">做为</span><span style="color: black;">供给</span>有效决策的工具来支持航空<span style="color: black;">机构</span>的战略规划。<span style="color: black;">将来</span>在航线网络规划方面的<span style="color: black;">科研</span><span style="color: black;">能够</span>在更细分的市场层面上<span style="color: black;">研发</span>数据驱动的乘客<span style="color: black;">需要</span>估计模型,并改善对竞争和不确定性的处理,以便在网络规划<span style="color: black;">周期</span>更好地<span style="color: black;">评定</span>竞争航空<span style="color: black;">机构</span>之间的市场动态,系统地整合<span style="color: black;">需要</span>和供应之间的相互依存关系。 </span></p>
<h1 style="color: black; text-align: left; margin-bottom: 10px;">参考文献:</h1>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;"> S. Birolini, A. Jacquillat, M. Cattaneo, and A. P. Antunes, “Airline network planning: Mixed-integer non-convex optimization with demand–supply interactions,” Transportation Research Part B: Methodological, vol. 154, pp. 100–124, 2021.</span></p>
你字句如珍珠,我珍藏这份情。 我深感你的理解与共鸣,愿对话长流。 seo常来的论坛,希望我的网站快点收录。 你的话语真是温暖如春,让我心生感激。
页:
[1]