At Aviation Festival Asia 2026 in Singapore, Fetcherr's VP of LMM, Hadar Sharvit, presented a keynote to explain how airlines can move from reactive, rule-based pricing to proactive, AI-driven revenue management. Sharvit has nearly a decade of experience building machine learning systems and a background in demand estimation, time-series analysis, and physics, and he oversees R&D for Fetcherr’s Market Model.
Sharvit opened up by explaining how standard traditional systems miss revenue opportunities. He mentioned that these often change prices in correlated and bulk-like moves. If airlines ran the same pricing decisions through an AI decision platform like Fetcherr’s Market Model, prices can adjust continuously and independently, instead of moving in bulk when legacy rules are triggered.
The Market Model is a deep neural network that consumes hyperdimensional market data, including historical prices, seasonality, competitor behaviour, and external signals such as fuel prices and major events, and predicts market dynamics like demand, elasticity, and willingness to pay. A simulation engine then tests millions of price points to discover the optimal policy, using traceable and readable actions, so we understand why the AI made a certain decision.
Watch Hadar Sharvit’s full keynote presentation below and see how airlines can move from reactive to predictive pricing.
