-
-
Dashboard
-
Initial ideas
This project presents an end-to-end intelligent optimization system for airline catering operations, developed for the HackMTY 2025 hackathon. The solution integrates advanced machine learning and operations research techniques to address three critical challenges in gategroup's Pick & Pack process: consumption prediction, expiration date management, and productivity estimation. At its core, the system employs SARIMA time series models to forecast product demand with passenger-specific granularity, combines Monte Carlo simulation for risk-adjusted inventory planning, and utilizes linear programming (PuLP) to optimize cart configurations under multiple constraints including weight limits, expiration dates (FIFO), and time windows. The optimization engine maximizes expected revenue while minimizing waste by intelligently balancing sale probabilities, product costs, and inventory availability. Additionally, a Random Forest regression model predicts drawer assembly times based on complexity metrics, enabling better workforce planning. The entire workflow is orchestrated through an automated pipeline that processes flight schedules, dynamically allocates inventory batches, simulates consumption patterns, and updates stock levels in real-time. This data-driven approach transforms traditional manual processes into a scalable, intelligent system that reduces food waste, optimizes resource allocation, and improves operational efficiency across global catering operations.
Log in or sign up for Devpost to join the conversation.