Market Intelligence and Analytics

Airport Intelligence Series Probabilistic Demand Forecasting for Airport Master Planning: Hyderabad Case Study April 2026 7 min read Master plans, capital programmes, and terminal sizing decisions worth hundreds of millions of dollars are routinely built on single-line demand projections. The question is not whether the forecast will be inaccurate — it will be — but how much, and in which direction, and what that means for infrastructure that takes a decade to plan, fund, and build. Probabilistic forecasting offers a fundamentally different approach: instead of a single number, it delivers a distribution of plausible outcomes, each with an associated likelihood. This opinion piece applies that approach to paint a range of scenarios for both domestic and international passenger traffic at Hyderabad International Airport (HYD), one of the fastest-growing southern aviation hubs in India. Hyderabad International Airport Case Study Domestic traffic has quadrupled in a decade; international connectivity has expanded to 26 scheduled destinations with total passenger traffic volumes exceeding 29 million in FY 2025. There is a Master Plan update underway that charts the next phase of growth at Hyderabad airport and will make key decisions on the delivery timeline of big-ticket items such as a North Code F Runway, new Passenger Terminal and associated landside facilities. A total capital expenditure outlay of over INR 14,000 crore ($1.5 Billion) is anticipated in the next 5-7 years. IndiGo is the dominant carrier with a market share of over 71% on the domestic side. This carries both advantages and risks. Their wide-ranging domestic exposure adds risk for the growth of local traffic – a key component of the overall aviation traffic growth anticipated at the airport. A Herfindahl-Hirschman Index of above 5000 represents a highly concentrated market, tying the success of the airport closely to the success of Indigo’s operation at HYD. The disruptions in December 2025 due to the failure to adapt to the new FTDL norms are an example of the demand risk (albeit temporary in this case) for established monopolies. Domestic passenger traffic clocked 12.8% lower in December 2025 at HYD. On the international front, the ongoing West Asia conflict continues to impact air travel to the Gulf corridor, a segment that accounts for approximately 65% of HYD’s international seat supply. The Historical Record: From 7 MAP to 29 MAP in a Decade HYD’s growth story is overwhelmingly a domestic aviation story. In FY2010, the airport handled 4.8 million domestic passengers. By FY2025, traffic volumes had scaled 24.4 million, a CAGR of approximately 10.7% over 15 years. The airport grew on steroids pre-COVID period: between FY2015 and FY2019, domestic traffic grew at a CAGR of roughly 23%, propelled by IndiGo’s aggressive capacity deployment. The transfer traffic at HYD also increased significantly with IndiGo’s Domestic to Domestic hubbing strategy. International traffic has followed a parallel but slower trajectory. From 1.9 million in FY2011, international passengers grew to 4.0 million by FY2019 (CAGR of approximately 9.2%), Recovery after COVID has been robust: Traffic volumes in FY2024 scaled 4.2 million international passengers (+23.2% YoY), and FY2025 reached a new peak of 4.7 million. The international share of total traffic remained stable at 16-18% in recent times. Source: AAI / Avinia Labs Methodology: SARIMAX and Monte Carlo Simulation An analytical framework based on a Seasonal ARIMA model with exogenous variables (SARIMAX), overlaid with 1,000-iteration Monte Carlo simulation was used to generate probabilistic confidence bands. The model was run independently for domestic and international traffic. The COVID-19 period (March 2020–December 2022) was treated as a temporary exogenous shock using a binary variable. Model parameters were selected using the Akaike Information Criterion (AIC). Monte Carlo residuals were sampled from the post-COVID normalisation period (January 2023 onwards), ensuring that the simulated uncertainty bands reflect current operating conditions rather than pandemic-era noise. Domestic: The IndiGo Adjustment Since the December 2025 disruption represented a temporary exogenous shock, the data point was not utilized for the SARIMAX model to avoid any near-term downward recency bias. The temporary phenomena is evidenced in the flatlining of January 2026 volumes as compared to January 2025. International: The Middle East Disruption Scenario For international traffic, distinct macro-scenarios with geopolitical disruption were analyzed. The disruption scenario quantifies the impact of the West Asia conflict on HYD’s Gulf-dependent international traffic. The modeling assumed that the disruption would continue till May 2026 and drop in volumes to the Gulf region would be confined to the ongoing month of March through May (representing 65% of total international supply from HYD). Some recovery was assumed to happen in April. Projections to FY2030: Three Scenarios Across Two Segments From the 1,000 Monte Carlo simulated demand paths, three scenarios drawn from the Monte Carlo distribution are most informative: P25 (conservative, exceeded 75% of the time), P50 (base case, the forecast mean), and P85 (optimistic, exceeded only 15% of the time). The P25-P85 range defines the corridor within which traffic is most likely to fall under the modelled conditions. Domestic Traffic Projections Under the P50 base case, domestic traffic grows from 24.4 million in FY2025 to approximately 32.2 million by FY2030 representing aCAGR of roughly 5.7%. The P25 conservative path (resulting in 27.1 million passengers in FY2030) reflects a world where IndiGo’s fleet recovery is delayed, economic headwinds dampen travel demand, and competing airports absorb a larger share of southern India’s growth. The P85 optimistic scenario reaches approximately 40.0 million (CAGR of 10.4%), driven by full IndiGo fleet restoration, aggressive deployment by competitors, and GDP growth above 7%. Source: Avinia Labs SARIMAX-Monte Carlo model The steep upside CAGR (10.4%) vs muted downside (2.1%) suggests:The market is supply-constrained in the base case.When supply is released → demand responds stronglyThe spread between P25 and P85 widens dramatically over the horizon: from 5.6 million in FY2027 to 12.9 million by FY2030. A single-line forecast would entirely obscure this aspect. International Traffic: Geopolitical Disruption The geopolitical disruption scenario tells a different near-term story. By removing Gulf seat supply for March-May 2026, FY2027 international traffic volumes at the P50 level are lower by 380,000 passengers

Airport Intelligence Series The Single Line Forecast Will Be Wrong. Here’s How to Plan for It – Washington Dulles Airport Case Study March 2026 7 min read Airport traffic forecasting has traditionally been an exercise in false precision. Master plans, capital programmes, and terminal sizing decisions worth hundreds of millions of dollars are routinely built on single-line demand projections that imply a certainty no forecaster can honestly claim. The question is not whether the forecast will be wrong — it will be — but how wrong, in which direction, and what that means for infrastructure that takes a decade to plan, fund, and build. Probabilistic forecasting offers a fundamentally different approach: instead of a single number, it delivers a distribution of plausible outcomes, each with an associated likelihood. This opinion piece applies that approach to the international traffic at Washington Dulles International Airport (IAD), one of the fastest-growing international gateways in the United States. Washington Dulles Case Study IAD’s new use and lease agreement includes a $9.0B capital program for a 15-year period. That’s not chump change for an anticipated 13 million annual passenger growth from now. What if the growth doesn’t materialize? IAD has been in the news lately – US Department of Transportation (DOT) had issued a Request for Information (RFI) seeking design, financing and construction concepts ideas for rebuilding IAD. The responses that came in ranged from some practical ideas to pure marketing sells. During the RFI stage, an earlier opinion piece by Avinia Labs on the Airport P3 Investment Decision Framework in the context of IAD was published, which presented a single line demand total passenger forecast and highlighted the high market share concentration. United Airlines dominance at IAD carries both advantages and risks. Their wide-ranging international exposure adds risk for the growth of international traffic – a key component of the overall aviation traffic growth anticipated at the airport. A Herfindahl-Hirschman Index of 3,900 represents a highly concentrated market, tying the success of the airport closely to the success of UA’s operation at IAD. The Historical Record: A Story of Disruption and Recovery IAD’s international passenger trajectory over the past decade illustrates precisely why rigid point forecasts fail and why the airport’s current position demands closer attention than a simple recovery story would suggest. Before COVID-19, international traffic at IAD was relatively stable, hovering between 8.0–8.5 million passengers from 2016 to 2019 with modest annual growth below 1.5%. Growth was incremental, driven primarily by gradual expansion in long-haul connectivity to Europe and the Middle East rather than structural network transformation. The pandemic caused a severe contraction in 2020, with international passengers falling by over 76% to roughly 2 million. Recovery began in 2021 and accelerated sharply in 2022 as travel restrictions lifted. By 2023, IAD had surpassed its pre-pandemic peak, reaching 9.35 million passengers. Growth continued in 2024 (10.38 million) and 2025 (10.53 million), marking three consecutive record years. Monthly peaks now exceed 1.1 million passengers — 22% higher than the 2019 high. Unlike the pre-COVID plateau, the recent expansion reflects structural network broadening rather than incremental frequency additions. Total annual international flights rose 12% between 2023 and 2025. Growth has been driven by new airline entrants and new destinations across Europe, Latin America, Africa, and Asia, including significant capacity additions and aircraft upgauging (e.g., A380 deployment). While post-2020 compound growth (~39%) is inflated by the recovery base effect, even measured from 2019 levels, international traffic is growing at approximately 3.8% annually — outperforming the long-term U.S. international average of 2–3%. United Airlines, as the dominant home carrier operating 68% of flights at IAD, underpins the entire network, up 15% over two years. Its regional partners Republic Airline and GoJet nearly tripled their international feed operations from 494 to over 2,200 flights, reflecting United’s strategic push to broaden international connectivity through its IAD hub. The airport is not simply recovering; it is consolidating its position as a connecting hub for international traffic originating from secondary US markets – a structural shift with meaningful implications for how the forecast range should be interpreted. Monthly Seasonality: Understanding the Peaks International passenger traffic at IAD follows a pronounced seasonal pattern that is critical for capacity planning, staffing, and terminal design. The peak travel season runs from June through August, driven by summer leisure demand, diaspora travel, and student movements. These peak months carry roughly 70-80% more traffic than the winter trough months of January and February, when volumes typically drop to between 600,000 and 700,000 passengers. The peak-to-trough ratio at IAD approximately 1.7:1 to 1.8:1 for international traffic is moderately high by US gateway standards, though less extreme than leisure-dominated airports in Florida or the Caribbean. Comparing peak months across years reveals a structural upward shift: June 2024 traffic exceeded June 2023 by approximately 10%, and June 2023 itself was roughly 15% above June 2019. This is not merely recovery it is structural growth layered on top of the seasonal pattern. The shoulder months of April-May and September–October have also strengthened, partly driven by airline schedule changes and the expansion of year-round long-haul services. This broadening of the demand profile is a positive development for asset utilisation, though the summer peak remains the dominant design constraint. Methodology: SARIMAX and Monte Carlo Simulation For this analysis, Avinia employed a Seasonal ARIMA model with external inputs (SARIMAX) a statistical framework purpose-built for time-series data exhibiting both trend and cyclical behaviour. The SARIMAX model captures three key drivers of airport traffic: long-term structural growth in demand, predictable monthly seasonality operating on a 12-month cycle, and short-term fluctuations around the trend. An explicit trend component was included to reflect the sustained post-recovery growth trajectory observed at Dulles. Critically, the COVID-19 period from March 2020 to December 2022 was treated as a temporary exogenous shock rather than a permanent structural change. A binary external variable was introduced to isolate the pandemic’s effect on passenger volumes, allowing the model to learn from the underlying demand dynamics without being distorted by the crisis period. This methodological

Airport Intelligence Series Top 10 U.S. Domestic Airport Pairs September 2025 4 min read The U.S. domestic market continues to be shaped by hub dynamics and route specific supply decisions. Unsurprisingly, the airports that dominate the list are serving some of the most populated catchment areas. Drawing on supply data [1] of seats for three consecutive summers (June to August of 2023, 2024 and 2025, referred to as SYY from here on), the analysis identifies the dominant carriers by airport pair and highlights the trajectory of their dominance. In S25, the largest legacy carriers; United Airlines, Delta Airlines, American Airlines – carved out a market share of 66% across the top 10 domestic airport pairs with United Airlines far ahead of the pack with 28% market share of the top 10 US domestic airport pairs. The pecking order for the top 10 domestic pairs tell an interesting story. Largely, the pairs have held their ranking in S25 as compared to S24 with at most a drop or gain in ranking by one position for some airport pairs. A three-year comparison of summer schedule across top airport pairs reveals how carriers have fortified their dominance, how low-cost and ultra low-cost carriers (LCCs and ULCCs) are tactically adjusting, and route-specific supply decisions are tightly aligned with broader network planning objectives. The JFK>LAX corridor is the busiest U.S. domestic airport pair in S25 with a 10.3% YoY growth from S24, and represents a strategic battleground for transcontinental traffic. The entry of Frontier with a 4.9% share in S25 signals possible ULCC experimentation, though the pair remains highly concentrated among legacy and hybrid carriers – Delta ~ 44% and JetBlue 33% and American 19% market share. LGA>ORD has dropped 1 position to #2 ranking with a 4.4% YoY growth in S25 as compared to S24. The move up in the ranking from 7th position in S23 reinforces its status as a high-frequency business trunk pair. United Airlines has increased its market share (by 6 percentage points) to 48% as the carrier continues leveraging its stronghold at Chicago O’Hare. The casualty of the battle has been American and Delta – losing market share by 5.0 and 3.0 percentage points respectively. Spirit Airlines coming out of bankruptcy has filed for second bankruptcy within a year due to its inability to fix its cost structure. Operating expenses have continued to exceed revenue, forcing the airline to cut several loss-leading routes. Its market share has dropped from 9.8% in S24 to 3.0% in S25, and the company plans to further reduce its presence in the near future as part of additional cost-saving measures. On the West Coast, LAX>SFO route posted a 3.8% increase in seat supply in S25 as compared to S24, and retained its #3 rank in seat supply volumes. United and Delta are dominant with 33% and 25% market share respectively. Notably, American Airlines has ramped up service and increased market share from 2.6% to 6.6%, while Frontier and Horizon have emerged with ~5% shares respectively, reflecting diversification of supply. In contrast, JetBlue and Alaska have fully exited as part of the larger network restructuring to cut unprofitable routes. ORD>LAX, one of the fastest growing pair, increased seat supply by 12% and moved up one spot to #4 position in S25 – United (47.8%) and American (40.0%) dominating the pair. Frontier is the new kid on the block with a 3.4% market share. At #5 position, the seat supply for the East Coast corridor LGA>ATL fell sharply (-13.9% drop) with the exit of JetBlue and the reduced service from Spirit, illustrating the challenges of contesting in Delta’s fortress hubs. Delta increased its dominance to nearly 70% market share. Again, Frontier was the only other airline to grow seat volumes by 1.5%. Other competitors, including JetBlue and Spirit, exited. In the #6 position DEN>PHX corridor, not surprisingly dominated by Southwest, who remains the market leader (48.0%) but has ceded ground to United (23.6% market share, 2% seat supply growth in S25). While Frontier’s growth has plateaued, American has rebounded modestly. The market contracted by 7.0% in S25 compared to S24. At the #7 position EWR>LAX market contracted by 1.2% in S25. United was the dominant carrier (69.2%), with declining seat volumes for JetBlue, Alaska, and Spirit, reflecting incumbents’ advantage in long-haul markets. This dominance is expected to increase with United’s moving their transcontinental premium “PS” service from JFK to EWR from October. At the #8 position, JFK>SFO route is dominated by Delta and JetBlue controlling ~69% combined share, Alaska Airlines continued to retreat, cutting seat volumes by 5.5% in S25 and 8 percentage points compared to S23. This route grew 5.4% in seat supply in S25. Leisure markets – #9 EWR>MCO is stagnant with S25 volumes similar to S24 while #10 PHL>MCO volumes in S25 are down compared to the S24 by 4.0%. Overall, S25 trends illustrate a re-concentration of supply around high-demand transcontinental and business-focused corridors, led by legacy carriers strengthening their share amid selective ULCC entry and exit. The overall seat supply in S25 for the top 10 domestic markets increased by 1.2%. United, Delta, and American have garnered nearly two-thirds of the market share of the top 10 domestic markets, cementing their network advantages, while ULCC entries remain limited. [1] Source: Cirium SRS Analyzer Share Share Share

Airport Intelligence Series Predictive Analytics for Airport Capacity Planning September 2025 3 min read Why Passenger Demand Forecasting at a granular level matters Traditional planning cycles, seasonal schedules, and monthly roll-ups can’t keep pace with real-world variability: shifting booking patterns, weather, special events, and disruptions. Passenger demand is the thread that ties it all together. If airport operators can predict inflows and outflows through curb, check-in, security, immigration, boarding, and arrivals, they can right-size resources and make targeted, timely interventions. That translates to shorter queues, more consistent dwell times, and smoother peaks and troughs that ripple less across the day. Variability of the demand from one day to another day is what keeps some of the capacity and operations planning folks up all night. Traditional ways of estimating passenger demand through historical load factor (% of seats filled by flight) and applying a show up profile (also known as the distribution of time that passengers arrive at an airport) by route/market are very static ways of estimating demand. The precision is indeed lacking to be able to have a high degree of confidence in the numbers. Airports are interested in predicting hourly and daily demand with a higher degree of confidence interval, and not just quarterly or annual demand to streamline resourcing allocation decisions. Use Case – Profile Estimation for a Typical Day at a Major Indian Airport Avinia Labs, using several years of tower data for a major Indian airport, has demonstrated the utility of machine learning (ML) models to forecast typical weekday and weekend patterns, as well as the busiest days of the year such as Diwali or Christmas Eve. In this use case SARIMA model was used, which is among the more popular time series forecasting techniques. 11 months (January to November) 2019 of tower data was used to train the model to produce daily and hourly profiles for the month of December. Specifically, the weekly seasonality was tested and separate daily profiles for a Friday and Sunday was tested against actuals. This also included testing the busiest travel day in the year, which fell on Christmas Eve. The comparison is shown in the graphic below. It shows a close match between the predicted values and actuals except for the busiest day of the year outlier. Busiest travel days require bespoke models instead of one-size-fits-all averages, improving accuracy when passenger experience is most sensitive. Modelling the busiest day would require training the model on a few years of the busiest day to enable it to predict with a higher precision going forward. However, there are some limitations as with any model. Larger datasets are required to improve accuracy. As models are exposed to larger, more diverse datasets, they learn from anomalies, become more robust across airport types, and improve reliability. Modeling extreme cases such as the busiest day of the year will require representative data to train the model for busiest days over several years. Forecasting is part science and part Art. Any application without a subjective context or interpretation can be misleading. Patterns tend to overestimate “causality”. Any constraints and other “temporal” factors need to be considered. Got a use case idea? Just hit reply and tell us what analysis you would find most useful— we can crunch the numbers at your beloved airport. Share Share Share

Airport Intelligence Series India Passenger Traffic Market Deep Dive July 2025 2 min read Passenger traffic levels in India have surpassed its pre-COVID benchmark, with total passenger traffic clocking 412 million in 2025 [1], a 20 percent increase compared to 2019. India represents the third busiest domestic O-D market. Where do we go from here? 2.5x growth to 1+ billion passenger volumes by 2040. There are several scenarios that could play out on the back of anticipated economic growth, air travel penetration, emerging geopolitical and trade barriers. Each of these scenarios generate a very different set of numbers in a 15-year time frame from today. The moderate growth scenario (also referred to as the diversification scenario) or where Indian carriers will continue to tap new markets and propensity to fly will substantially increase in the next 10 years.   In the diversification scenario, Avinia Labs projects a robust growth in passenger demand at an India wide level on the back of record fleet orders and expansion of airport infrastructure. The passenger volumes will be a tad over 1 billion passengers in 2040, reflecting a CAGR of 6.6 percent from 2025 to 2040. Near-term growth is projected to be robust, with a CAGR exceeding 8 percent through 2030, pushing India wide annual passenger volumes to 600 million. The diversification scenario assumes structural transformation in the aviation landscape, with significant airport infrastructure investments in non-metro regions and increasing passenger willingness to fly. The supply side tells a slightly different story based on confirmed aircraft orders. The commercial fleet has grown at a 7.6 percent CAGR over the past decade, reaching over 860 aircraft as of December 2024[2]. As of January 2025, Indian airlines had collectively placed gross orders for 1,800 aircraft [3]. The net additions (excluding wet leases) by 2030 are expected to be around 600+ aircraft assuming a standard rate of retirements. This addition could increase the supply by about 1.2 million additional domestic air traffic movements (ATMs). As can be seen in the chart above, the supply trajectory tracks above the demand trajectory for the moderate growth scenario by about 100 million passengers annually. This surplus supply is a theoretical construct. The surplus provides the industry with operational flexibility, the ability to open new routes, and room to absorb future demand shocks. Over time, supply is expected to adjust in line with demand, driven by India’s highly elastic demand–supply pricing dynamic. However, if demand outpaces expectations and leads to higher load factors, the resulting shortfall may be addressed through interim solutions, similar to the contingency measures implemented last year in response to engine-related groundings. Analysing demand-side forecasts alongside supply-side investments at the national level provides stakeholders with a strategic lens to guide infrastructure development, fleet planning, and capital allocation over the next 15 years. The logical next step would be to drill down to the regional level to identify specific demand–supply gaps that state or regional authorities could address. We plan to explore this in future editions of the newsletter. We would love to get your feedback on the numbers. [1] https://www.aai.aero/ [2]https://www.iata.org/en/iata-repository/publications/economic-reports/aviation-in-india [3]https://www.boeing.com/commercial https://www.airbus.com/en/products-services/commercial-aircraft/orders-and-deliveries Share Share Share