Shahram Shahinpour  |  Sabre

Traditional airline fleet assignment models (FAMs) often rely on leg-based demand and disregard complex network interactions, which limits their ability to maximize the total profitability of a schedule. To address market needs and remain competitive, airlines are increasingly interested in optimizing capacity allocation at a granular level, driven by passenger demand and fares for individual cabin classes. In this talk, we present a cabin-level, origin-and-destination fleet assignment model that overcomes the shortcomings of traditional FAMs and empowers airlines to optimize their capacity using detailed itinerary information.

Ahmed Abdelghany  |  Embry-Riddle Aeronautical University

This study introduces an airline network optimization framework that treats passenger demand as endogenous to seat capacity allocation decisions. Empirical analysis indicates that Origin–Destination (OD) demand responds nonlinearly to capacity additions, following market-specific saturation curves represented by asymptotic exponential or stepwise functions. In this setting, demand influences capacity allocation decisions, while capacity simultaneously stimulates demand, creating a bidirectional endogenous relationship. The proposed model determines optimal weekly aircraft and seat allocations across OD markets while enforcing aircraft flow balance constraints throughout the network. Assuming fixed fares and known operating costs, the objective is to maximize network profit through coordinated fleet deployment, flight frequency assignment, and demand-responsive capacity planning.

Mahekha Dahanayaka  |   University of Twente, Netherlands / KLM Royal Dutch Airlines

Robust flight scheduling designs airline timetables that remain operationally reliable under real-world uncertainty. At its core lies the delay propagation problem: independent delays propagate forward through aircraft rotation chains whenever scheduled ground time is insufficient to absorb them, tightening connection slack until misconnections cascade across the network. Despite significant advances, existing approaches share a structural limitation. They measure propagation cost in operational terms, delay minutes or passengers disrupted, ignoring the revenue consequences of where propagation strikes. Yet revenue at risk when a high-value connection fails can surpass that of a routine connection by orders of magnitude, leaving revenue-sensitive connections systematically under-protected. This study addresses this asymmetry by assigning each connection a commercial weight reflecting the passenger revenue at risk if delay propagation disrupts it. Scheduling decisions are steered toward connections where commercial exposure is greatest, re-timing departures and swapping aircraft duties, within feasible bounds, to maximise expected commercial value under stochastic delay propagation.

Jeffery Oboy | PA Consulting

Engine maintenance scheduling requirements have often been treated as a one-way street within broader airline planning context. Network planners and schedulers of course recognize the opportunity cost from TechOps restrictions over where and when specific tails must be for maintenance, but little has progressed in terms of integrated planning. This presentation will share transformative advances in computer processing and mathematical modelling that are enabling a new frontier in engine and broader airline planning along with findings to some of the most common objections.

Andreas Hottenrott  | Kearney

Flight schedule design is evolving from purely commercial planning to an integrated decision process that anticipates operational impact already in the strategic planning phase. We leverage historical data and Monte-Carlo methods to stress-test future schedules before operations begin. In our approach, we simulate aircraft rotations, delay propagation, day-of-ops disruptions, and smart mitigations such as aircraft swaps, translating complex operations into planning-relevant metrics: punctuality, passenger disruption, and costs.

Our approach enables airlines to compare schedule alternatives, quantify the value of robustness measures, and make decisions beyond gut feeling. Case studies at Lufthansa Group show high predictive accuracy and benefits enabled. The simulation core is now being embedded into an integrated optimizer to build schedules that are both commercially attractive and operationally robust.

Kevin Bryan and Maura Twillman  |  U.S. Department of Transportation Office of Aviation Analysis

On January 31, 2023, the U.S. Department of Transportation (DOT) issued a final rule updating the Origin-Destination Survey of Airline Passengers (O&D). Since the July 1, 2025, implementation deadline, all U.S. certificated and commuter air carriers have transitioned to this updated methodology. This presentation provides a comprehensive review of the updated O&D (DB1C) framework, focusing on:

• Methodological Shifts: Insights into updated carrier ticket submission logic and refined DOT-derived variable processing.
• Data Enhancements: An in-depth examination of new variables and the modernized DB1C output formats designed for high-resolution analysis.
• Accessibility & Distribution: Guidance on the digital channels available for external users to acquire and utilize these expanded datasets.

Attendees will gain a clear understanding of how this high-fidelity dataset enhances the industry's ability to analyze carrier competition and pricing trends in our increasingly data-driven aviation landscape.

Narges Sereshti  | Air Canada

Airline schedule design requires balancing operational efficiency with robustness under uncertainty across large-scale, highly constrained networks. We present an integrated framework combining predictive modeling, discrete-event simulation, and large-scale optimization to improve schedule resilience. In the first phase, supervised learning models and a simulation engine are used to estimate delay propagation risk and identify structurally vulnerable connections within the flight network. These outputs are translated into quantitative penalties and robustness indicators that inform downstream optimization. The second phase consists of a time-space network optimization model for aircraft routing, formulated to minimize a composite objective including propagated delays, passenger misconnections, and operational inefficiencies (e.g., towing). The model explicitly enforces maintenance requirements, fleet compatibility, and turnaround constraints. Due to the scale of the problem—millions of decision variables and constraints—we employ graph-based preprocessing, and decomposition techniques to ensure computational tractability. This work demonstrates how simulation-informed parameters can be embedded within a deterministic optimization framework to better capture and resolve operational risk. The resulting approach enables more robust schedule solutions while maintaining feasibility within real-world airline constraints.

Antonio Montaruli  |  HEC Montreal / KLM Royal Dutch Airlines / University of Twente

The airline planning process comprises multiple sequential stages, from strategic network planning to tactical flight scheduling, each traditionally addressed in isolation. Network planning jointly optimizes route selection, service frequencies, and fleet composition, while flight scheduling assigns departure times within bank structures to maximize hub connectivity. Despite strong interdependencies, where network outputs define scheduling inputs and schedule quality directly affects connectivity and passenger demand, existing approaches optimize each stage independently, often leading to suboptimal decisions. This work proposes an integrated optimization framework bridging network planning and flight scheduling for hub-and-spoke networks. The framework captures demand–supply interactions through an empirical demand model in the network planning component, and optimizes departure times within banks to maximize connectivity in the scheduling component. By coupling these stages, network decisions account for downstream scheduling implications, while scheduling decisions are informed by the strategic network structure. The approach is assessed on realistic hub-and-spoke instances.

Josephine Dietrich |  Data Powered Aviation Intelligence (DPAI)  and John Pepper  |  DPAI, Bermudair

The airline planning process comprises multiple sequential stages, from strategic network planning to tactical flight scheduling, each traditionally addressed in isolation. Network planning jointly optimizes route selection, service frequencies, and fleet composition, while flight scheduling assigns departure times within bank structures to maximize hub connectivity. Despite strong interdependencies, where network outputs define scheduling inputs and schedule quality directly affects connectivity and passenger demand, existing approaches optimize each stage independently, often leading to suboptimal decisions. This work proposes an integrated optimization framework bridging network planning and flight scheduling for hub-and-spoke networks. The framework captures demand–supply interactions through an empirical demand model in the network planning component, and optimizes departure times within banks to maximize connectivity in the scheduling component. By coupling these stages, network decisions account for downstream scheduling implications, while scheduling decisions are informed by the strategic network structure. The approach is assessed on realistic hub-and-spoke instances.

Fred Gardi  |  Hexaly

Mixed‑Integer Linear Programming (MILP) has been the dominant optimization framework in Operations Research for several decades. While it has proven extremely powerful, it is also well known that MILP formulations can become unwieldy when confronted with large‑scale, highly combinatorial, non‑convex, or structurally rich problems, particularly in application domains such as routing, scheduling, and packing.

Hexaly is an industrial optimization solver built around a hybrid, post‑MILP approach. Rather than relying primarily on linearization techniques and classical branch‑and‑bound‑centered workflows, it combines heuristic and exact methods and draws inspiration from multiple paradigms, including Mixed‑Integer Programming, Constraint Programming, Nonlinear Programming, and Black‑Box Optimization. A central design objective is modeling expressiveness and openness: enabling users to formulate problems closer to their natural combinatorial structure, while allowing diverse algorithmic components to interact in a complementary manner.

In this talk, I will present the guiding principles behind this approach, with a particular focus on discrete optimization problems where Hexaly currently demonstrates its strongest performance, such as large‑scale routing, scheduling, and packing. I will discuss how hybridization manifests not only at the algorithmic level, but also—crucially—within the modeling layer. Finally, I will provide a transparent overview of the solver’s current algorithmic status, supported by selected performance benchmarks.