Zalando Deploys ZEOS Inventory Optimization Tool
Zalando recently published in Nature Scientific Reports a practical replenishment optimization paper describing the ZEOS tool, which combines LightGBM probabilistic forecasts, discrete-event simulation, and an extended (R, s, Q) policy optimized via Monte Carlo. In a computational backtest (Oct 2023–Sep 2024) across ~2 million articles and ~800 merchants, the engine achieved +22.11% GMV, +21.95% gross margin, 91.14% fill rate, and 86.40% availability.
Key Points
- 1Implements simulation-driven probabilistic replenishment using LightGBM, DES, and extended (R, s, Q) policy.
- 2Optimizes risk-aware decisions by minimizing the 75th percentile cost across thousands of Monte Carlo futures.
- 3Delivers business uplifts in backtest: +22.11% GMV, +21.95% margin, 91.14% fill rate.
Scoring Rationale
Strong peer-reviewed evidence and large-scale backtest drive score; limited by focus on fashion e-commerce rather than cross-industry generality.
Sources
Public references used for this report.
Practice with real Retail & eCommerce data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Retail & eCommerce problems
