Technology Category
Planning, Routing, and Scheduling: Planning with Language ModelsSearch and Optimization: Learning to SearchNatural Language Processing: Learning & Optimization for NLP
Data-driven jet fuel demand forecasting: A case study of Copenhagen Airport
🤖 AI Summary
Inaccurate aviation fuel demand forecasting hampers supply chain optimization, as existing approaches rely heavily on expert judgment or deterministic models and lack data-driven, long-horizon predictive capabilities. To address this, we propose a hybrid data-driven framework for Copenhagen Airport that integrates multiple exogenous variables and—novelty—synergistically combines an LSTM-based sequence-to-sequence architecture with Facebook Prophet for 30-day rolling forecasts. Evaluated on real-world market data, the hybrid model achieves a 18.7% reduction in MAPE over standalone models and conventional methods, demonstrating superior robustness during periods of high demand volatility. This work empirically validates the efficacy of integrating deep learning with classical time-series modeling for long-term aviation fuel forecasting. Moreover, it delivers a production-ready decision-support tool for fuel distributors, advancing the intelligent,精细化 (fine-grained) transformation of aviation energy supply chains.
Technology Category
Application Category
📝 Abstract
Accurate forecasting of jet fuel demand is crucial for optimizing supply chain operations in the aviation market. Fuel distributors specifically require precise estimates to avoid inventory shortages or excesses. However, there is a lack of studies that analyze the jet fuel demand forecasting problem using machine learning models. Instead, many industry practitioners rely on deterministic or expertise-based models. In this research, we evaluate the performance of data-driven approaches using a substantial amount of data obtained from a major aviation fuel distributor in the Danish market. Our analysis compares the predictive capabilities of traditional time series models, Prophet, LSTM sequence-to-sequence neural networks, and hybrid models. A key challenge in developing these models is the required forecasting horizon, as fuel demand needs to be predicted for the next 30 days to optimize sourcing strategies. To ensure the reliability of the data-driven approaches and provide valuable insights to practitioners, we analyze three different datasets. The primary objective of this study is to present a comprehensive case study on jet fuel demand forecasting, demonstrating the advantages of employing data-driven models and highlighting the impact of incorporating additional variables in the predictive models.
Problem
Research questions and friction points this paper is trying to address.
Forecasting jet fuel demand to optimize aviation supply chains
Addressing the lack of machine learning studies for fuel prediction
Evaluating data-driven models for 30-day fuel demand horizon
Innovation
Methods, ideas, or system contributions that make the work stand out.
Uses machine learning models for jet fuel demand
Compares traditional time series with LSTM networks
Analyzes three datasets for reliable forecasting insights
🔎 Similar Papers
No similar papers found.
