A Binary Logistic Regression Model to Identify Key Aspects that Enhance Global Hub Airports Status

Abstract

This paper provides a tool to identify key aspects for an airport to achieve global hub status for a given airline and determines whether these factors are related to the facility's infrastructure, its region, or both. Despite the frequent use of the term ‘hub’, there is little academic consensus on its exact definition in air transport. Many define a hub based on passenger numbers rather than the concentration of flights and passengers from the main carrier. This study addresses this gap by analyzing the factors influencing the definition of a hub and the commonalities among global hubs. Data from 300 major airports, including internal variables (runways, terminals, gates and area) and external variables (economy, population, attractiveness), were collected. A Binary Logistic Regression (BLR) model identified key aspects influencing hub status, with the assistance of an Exploratory Factor Analysis (EFA) that grouped the variables into factors. The binary ‘hub’ variable was defined by the main carrier's activity and the Global Airport Connectivity Index (GACI). The factor with the highest coefficient primarily involved internal variables and, to a lesser extent, global attractiveness and population. The factor with the lowest coefficient related to the region economy. The BLR correctly identified hub status in 93.3% of cases, with 68.3% accuracy for hub airports. Airports not correctly identified by the model mostly present a lack or underutilization of existing infrastructure.