Learn how the ZEOS replenishment optimization system achieves up to 22.1% GMV uplift by unifying probabilistic demand forecasting with risk-aware discrete event simulation.

Zalando Tech Blog is a platform where Zalando's engineering team shares insights, experiences, and best practices on technology and innovation. Covering topics such as software architecture, data science, and machine learning, Zalando's Tech Blog offers resources for developers and tech enthusiasts. Developers can learn about Zalando's engineering culture, projects, and technology stack through their blog posts and articles.

Zalando

Zalando's ZEOS inventory optimization system combines probabilistic demand forecasting with discrete event simulation to solve the inventory paradox in e-commerce. The system extends the classical (R,s,Q) replenishment policy with lifecycle-aware parameters and uses Monte Carlo simulation to optimize under uncertainty. By modeling full probability distributions instead of point forecasts and optimizing for the 75th percentile cost, the system achieved 22.1% GMV uplift, 33.6% availability improvement, and 23.6% better demand fill rates compared to human decisions across 2 million articles over 12 months. The approach proves that explicitly embracing uncertainty through probabilistic forecasting and risk-aware optimization delivers substantial business value.

Paper Announcement: A Practical Approach to Replenishment Optimization with Extended (R, s, Q) Policy and Probabilistic Models

The Design: A Unified Optimization Architecture

The Simulation Method: Discrete Event Modeling

Comparative Analysis & Ablation: Why it Works