Case Study: Lenovo achieves 90%+ accurate, rapid sell-out forecasts with DataRobot

Lenovo Computes Supply Chain and Retail Success with DataRobot

Lenovo Brazil needed to balance supply and demand for thousands of laptops shipped to retailers each week but struggled to predict sell-out volume quickly and accurately. With only two people maintaining slow, hand-coded R models and limited data-science resources, Lenovo turned to DataRobot and its automated machine learning platform to scale predictive modeling and improve forecast accuracy.

DataRobot automated model building, ran many algorithms to produce a ranked Leaderboard, and provided Feature Impact/Feature Effects for clear interpretability, enabling Lenovo to make proactive inventory and production decisions. The results were dramatic: model creation fell from 4 weeks to 3 days, productionalization from 2 days to 5 minutes, accuracy rose from under 80% to 87.5% (and later over 90%), DataRobot users grew from 2 to 10 with a web simulator for 20+ stakeholders, and Lenovo surged to lead B2C notebook volume share in Brazil.

Lenovo

Rodrigo Bertin

Senior Business Development Manager