航空客運網路之形成與發展

Exploring Passenger Air Network Formation and Development

王郁珍(Yu-Chen Wang)

指導教授 ： 汪進財

國立交通大學/管理學院/運輸與物流管理學系/博士(2020年)

摘要

航空網路的形成及發展，其背後的驅動因素及如何回應市場需求與經營環境的變化，一直是機場營運者及航空公司所關注的課題。從微觀的角度視之，航空網路係由許多的機場及航線所串連組成的，機場連結性是最為外界熟知且已被廣泛應用於評估機場或航空公司航網便捷性、競爭力及機場發展的重要指標。機場直接連結不僅是航線形成的首要型態，也是航網發展的重要基石，因此，本研究首先提出以機場網路單元(Airport Network Unit, ANU)的概念釐清機場連結性的定義、型態，建立研究機場連結性的概念架構，並應用於亞洲機場連結力的探討，及作為研究航網形成及發展的基礎論述。
從供給的角度觀之，航網的發展包括新航線的開闢及既有航線的航班頻次增加，除探討那些因素影響機場直接連接、形成新航線以外，航線的航班頻次對於航網的拓展也具有相當的重要性。因此，為能對航網發展具有更詳細而全面的了解，及尋找潛在的航線市場，本研究以二部模型(Two-part model)構建影響因素與航線連結及航線航班頻次的關係，並採用Innovata 2016年的航班資料及相關影響因素的變數資料蒐集，建立主要亞州機場航網形成及發展模式，結果顯示市場需求(機場運量、綜合競爭力)對於航線的形成及航班頻次均具有顯著的正向影響效果，而低成本航空公司的營運則對航線航班頻次顯示有強化的效果；至航線距離、樞紐效應、多機場都會區等因素則不利於新航線的闢駛與航班頻次量的增加。另外，從航線距離與代表市場需求的機場運量和競爭力指數的影響結果顯示，愈大的需求市場愈能吸引更多的航線連結，即所謂的大者恆大；而當航線距離越長時，提供高頻率直達航線的機會就越少。結合這兩個變量的結果，反映出航線大多在一定距離內與大型機場直接連接，顯示機場航網具有小世界網路的特性。
為深入探討各影響因素在不同的市場特性是否對航線航班頻次具有相同的影響效果，本研究利用分量迴歸模式檢視影響因素的異質性效果。結果顯示機場運量、航線距離、是否有低成本航空營運、共用班號的航班比例等因素對航線航班數具有異質性效果。例如對於那些需求量偏低、尚未完全開發的航線市場，引進低成本航空公司營運或提高共用班號的航班比例，對航線航班頻次所產生的增加效果，明顯大於那些需求較大且已發展成熟的航線市場。而航線距離在發展相對成熟的市場則顯示具有較大的彈性。
囿於資料限制，本研究所引用之資料雖然無法完整全面的展現航網發展的樣貌，然而所提的研究架構、模式建立及結果分析，對於機場營運、航空公司及政府政策制訂，具有相當參考及指引，並可供作後續此領域的進一步探討。

關鍵字

航網發展 ； 機場連結性 ； 航線航班頻次 ； 二部模型 ； 影響因素 ； 異質性效果

並列摘要

Aviation network has changed significantly since open sky policy. How it formed and developed is always the major concern of airport operators and airlines. Airport connectivity is the most well-known and has been applied in various studies to understand how network transformation is related to market growth and how it influences airport development. Airport direct connections are not only the basic and core type of route formation but also embrace network development. Nevertheless, airport connectivity lacks a common definition and measurement. Aiming to explore passenger air network formation and development, first of all, this study represented airport connectivity by introducing the concept of airport network unit (ANU), which helps to clarify the content and types of airport connectivity. A conceptual research framework of airport connectivity and air network is developed, and applying to measure the major Asian airport connectivity. These have set the stage for exploring air network formation.
From the supply perspective, air network development consists of forming new routes and increasing flight frequency on existing routes. Thus, in addition to the focus on the direct connections driven by growing demand in the market, flight frequency on a route plays an equally critical role in exploring the characteristics of the route network. Simultaneous analysis of potential connections and route flight frequency in the air network contributes to a detailed and comprehensive understanding of network development. Therefore, two-part models for the Asia airport route network are developed. Flight frequency is used as the dependent variable and factors affecting route connection and flight frequency are used as the explanatory variables. Innovata 2016 flight data and the collected factors data are used for model estimation. The results indicated that the zero-inflated negative binomial regression model outperformed the others. All estimated results, including the sign and magnitude, are significant and reasonable. Based on the odds ratio and effects of the explanatory variables, the results reflect that routes are mostly connected directly to big airports within a specific range. The findings correspond to the characteristics of a small-world network structure. In addition, among other factors, low-cost carriers and visa exemption show a positive effect on route flight frequency.
In order to analyze the affecting factor’s effect in various route markets, this study presents an explorative analysis of the factors driving route traffic and examines the heterogeneity of factors that affect route flight frequency by using a quantile regression methodology. The results indicate that factors related to airline operations indeed exhibit a heterogeneous effect on various markets. Particularly, the factor effects of having low-cost carriers (LCCs) in operation and the rate of codeshare flights showed a descending effect as the quantiles increase. This finding implies that routes in the low quantiles may represent an under-cultivated market, the potential of which is not yet fully exploited. On the other hand, routes in the upper quantiles represent markets that are fully exploited and tend to be mature. Thus, the under-cultivated markets are a supply-driven niche. As a consequence, LCCs and codeshare flights can boost and enhance the potential of the market. Compared to enhancing factors, route distance showed a larger elasticity of impedance in fully exploited markets.

並列關鍵字

Air network development ； Airport connectivity ； Route flight frequency ； Affecting factors ； Two-part model ； Heterogeneous effect

參考文獻

Airports Council International. (2017). Annual World Airport Traffic Report 2017 edition.

Azzam, M., Klingauf, U., Zock, A. (2013). The accelerated growth of the worldwide air transportation network. The European Physical Journal Special Topics 215: 35. https://doi.org/10.1140/epjst/e2013-01713-7

Bagler, G. (2008). Analysis of the airport network of India as a complex weighted network. Physica A: Statistical Mechanics and Its Applications, 387(12), 2972–2980. https://doi.org/ 10.1016/j.physa.2008.01.077

Barabási, A., Albert, R. (1999). Emergence of Scaling in Random Networks. Science, 286(5439)(October), 509–512. https://doi.org/10.1126/science.286.5439.509

Bhadra, D., Kee, J. (2008). Structure and dynamics of the core US air travel markets: A basic empirical analysis of domestic passenger demand. Journal of Air Transportation, 14, 27–39. https://doi.org/10.1016/j.jairtraman.2007.11.001