Network centrality is an intermediary between airport resource utilization and air traffic generation. A central position in the network with frequent and regular flights with hub nodes can boost air traffic by providing better accessibility, resulting in more efficient use of airport resources. However, this relationship has been largely ignored in the literature. Using data from the Turkish airport industry, this paper proposed a weight-restricted Network Data Envelopment Analysis model, which considers network centrality measures as the cornerstone intermediates that establish the link between airport resources and the traffic volume handled. In the first stage, called networkability, assets such as runways, terminals, aprons, and special purpose vehicles, and exogenous factors including population, socio-economic development, and tourist arrivals are used to accomplish the network integration with other airports, as measured by degree centrality, betweenness centrality, and eigenvector centrality. In the second stage, called traffic generation, this network integration allows for aircraft movements and workload unit to be handled. Criteria weights of model variables were calculated using Criteria Importance Through Intercriteria Correlation. The main findings indicate that 1) the weight-restriction procedure improved the robustness of Network DEA, 2) the proposed two-stage structure reveals whether performance losses are due to networkability or traffic generation capabilities and helps to identify the right policies for performance improvement, 3) the Turkish airports generally suffer from the inability to establish connections in the domestic network, 4) the pandemic has significantly improved the domestic networkability of airports due to mandatory direct flights while devastating the traffic generation capability, 5) low betweenness centrality is the main reason for weak networkability, and 6) good networkability may not ensure air traffic generation.

中文翻译：

---

网络中心性驱动的机场效率：权重受限的网络 DEA

网络中心性是机场资源利用和空中交通产生之间的中介。网络中的中心位置，具有枢纽节点的频繁和定期航班，可以通过提供更好的可达性来促进空中交通，从而更有效地利用机场资源。然而，这种关系在文献中很大程度上被忽视了。本文利用土耳其机场行业的数据，提出了一种权重限制的网络数据包络分析模型，该模型将网络中心性度量视为建立机场资源和处理的交通量之间联系的基石中间体。在第一阶段，称为网络化，利用跑道、航站楼、停机坪和专用车辆等资产以及人口、社会经济发展和游客人数等外生因素来完成与其他机场的网络整合，衡量标准为度中心性、介数中心性和特征向量中心性。在称为流量生成的第二阶段，这种网络集成允许处理飞机起降和工作负载单位。使用标准重要性通过标准间相关性计算模型变量的标准权重。主要发现表明 1) 权重限制程序提高了网络 DEA 的稳健性，2) 所提出的两阶段结构揭示了性能损失是由于网络能力还是流量生成能力造成的，并有助于确定性能改进的正确策略， 3）土耳其机场普遍无法在国内网络中建立连接，4）疫情由于强制直飞，显着改善了机场的国内网络能力，同时破坏了客流量生成能力，5）介数中心性低是主要因素网络能力弱的原因，以及 6) 良好的网络能力可能无法保证空中交通的产生。