Artificial Neural Networks (ANNs) have obtained significant results across different machine learning applications, but their computational complexity and energy consumption remain significantly higher than those of biological neural circuits. Spiking Neural Networks (SNNs), which utilize sparse and event-driven spike-based computation, offer a biologically inspired alternative with potential advantages in terms of energy efficiency and hardware compatibility. In the context of Amadeus Revenue Management, where forecasting models have to balance accuracy, scalability, and interpretability, SNNs represent a promising research direction. This thesis investigates the applicability of SNNs to simplified Revenue Management forecasting tasks. Regression experiments are conducted on linear and sinusoidal functions, followed by parameter estimation tasks that mimic the fitting of analytical forecasting models to historical demand data.