Modeling competitive demand dynamics in multi-airport regions using time-series econometrics

Understanding the interaction among co-located airports is critical for accurate demand forecasting and air service planning. In regions with overlapping catchment areas, airline decisions at one airport can affect demand at the other. Nonetheless, few studies have quantified these interactions or their temporal dynamics. This study presents an applied empirical analysis using the Seasonal Autoregressive Integrated Moving Average with Exogenous Regressors (SARIMAX) framework to quantify cross-airport demand dynamics in multi-airport regions. First, we develop SARIMAX-based demand models for 16 U.S. metropolitan areas to evaluate how changes in seat capacity and average airfare at one airport are associated with demand shifts at both the focal and its co-airport. Second, we conduct a detailed market-level case study of the Dallas–Fort Worth metropolitan area, examining how demand in shared origin markets from Atlanta (ATL), Seattle (SEA), and Phoenix (PHX) is distributed between Dallas Fort Worth International Airport (DFW) and Dallas Love Field Airport (DAL). Enhanced service levels at one airport tend to coincide with rising demand at that airport and reduced demand at its co-airport. These insights provide an evidence base for transport policymakers, airline managers, and airport authorities to coordinate slot allocation, manage congestion, and plan infrastructure investments across multi-airport systems. • Develops a SARIMAX-based forecasting framework to quantify dynamic demand interactions between co-located airports. • Analyzes 16 U.S. multi-airport regions and reveals consistent cross-airport substitution patterns using revealed-preference data. • Demonstrates dual-scale modeling, combining aggregate analysis with a detailed OD-level case study of the DFW–DAL airport system. • Provides actionable insights for airport planning, airline scheduling, and congestion management in multi-airport environments.