VN2 Inventory Planning Challenge
Share:
Test your inventory planning skills: cut holding costs, prevent shortages, and master real-world inventory planning.
An ordering system that doesn’t just forecast — it understands relationships and anticipates demand.
Most inventory systems still think in rows and tables.
But real demand lives in relationships: stores ↔ products, orders ↔ shipments, peers ↔ trends, and time flowing through it all.
Inspired by the latest Relational Graph Transformer (RelGT) research and our own work on RELIA, we built a temporal relational graph that learns directly from these connections — and turns probabilities into calibrated, cost-aware orders .
In this post, I share:
- how we move from transactional data to a time-aware graph,
- how attention over neighborhoods replaces handcrafted features,
- how calibrated probabilities drive smarter base-stock decisions,
and how RelGT’s ideas (multi-element tokens, local+global attention) can shape the next generation of retail AI.
💡 The goal: an ordering system that doesn’t just forecast — it understands relationships and anticipates demand.
Read the full story 👉 https://medium.com/@nasdag/building-a-practical-graph-native-ordering-system-584a5b4f6784
Keep up to date by participating in our global community of data scientists and AI enthusiasts. We discuss the latest developments in data science competitions, new techniques for solving complex challenges, AI and machine learning models, and much more!
Comments