Aircraft spare parts inventory management is a critical challenge in aeronautical maintenance; it requires a balance between operational readiness and sustainability. Airlines seek to avoid costly stockouts that disrupt service while minimizing both the financial and ecological burdens of overstocking. This study presents an integrated multi-objective optimization framework for sustainable aviation logistics that combines machine-learning forecasting, circular-economy principles, and green logistics optimization :a feature engineering pipeline augmented with sustainability metrics, a hybrid XGBoost-LSTM forecasting model incorporating sustainability-based weights, and a circular economy recommendation engine. Results indicate that embedding sustainability criteria directly into predictive logistics frameworks improves both operational efficiency and resource recovery — without treating the two as competing objectives.

Managing spare parts inventory for aircraft presents a significant challenge in aviation maintenance, requiring careful coordination between keeping operations running smoothly and maintaining sustainable practices. Airlines must prevent expensive parts shortages that could ground flights while also reducing both the financial costs and environmental impact of holding too much inventory. This research introduces a comprehensive multi-objective optimization approach for sustainable aviation supply chains that integrates machine learning predictions, circular economy concepts, and environmentally conscious logistics optimization. The framework includes a feature engineering process enhanced with sustainability measurements, a combined XGBoost-LSTM prediction model that incorporates sustainability-focused weightings, and a circular economy recommendation system. The findings show that integrating sustainability factors directly into predictive logistics systems enhances both operational performance and resource recovery, demonstrating that these goals can work together rather than against each other.