Lenovo, a multinational company headquartered in China and the United States, is one of the largest technology companies in the world, with $43 billion in revenue, hundreds of millions of customers, and four devices sold per second. Lenovo's "Intelligent Transformation" initiative is aimed at transforming its customers’ experiences with technology through constant innovation and a broad range of connected devices.
A key factor in Lenovo’s success is its global growth and the end-to-end control the company has over its supply chain. With a presence in more than 160 countries, Lenovo continues to expand into emerging markets to drive new retail business outcomes and opportunities. Brazil is a primary emerging market for Lenovo and the company sought to use machine learning (ML) tools to help predict sell-out volumes in the country. This was crucial for improving many aspects of the business, from supply chain logistics to marketing efforts, while the company worked to establish itself among customers and retailers in the South American market.
However, the Lenovo Brazil team didn’t have the right resources available to predict sell-out volume at scale.
The company set out to identify a tool that could automate all of the modelling and forecasting steps it needed to take to drive predictions. The team turned to DataRobot, an AWS Partner Network (APN) Advanced Technology Partner and AWS ML Competency Partner. DataRobot offers an automated machine learning platform for data scientists of all skill levels to quickly build and deploy accurate machine learning models. And by running DataRobot Cloud on AWS, users across the organization can build world-class prediction models in minutes.
Previously, the Lenovo Brazil team would select an algorithm, run the cleansed datasets through the model, work on parameter optimization, retrain the model, score it, and adjust it to improve its accuracy. Using DataRobot, the team was able to automate and accelerate its previously arduous model-building process.
With accurate sell-out volume predictions for up to four weeks ahead, the Lenovo Brazil team can work with business stakeholders within Lenovo, as well as local Brazilian retailers, to define actions in the appropriate time, avoiding a lack or excess of inventory. Achieving a better balance of supply and demand improves the bottom line for both Lenovo and their participating retailers in Brazil.
Since DataRobot implemented their solution, Lenovo's model creation time has decreased from four weeks to three days. Model production time has decreased from two days to five minutes. Additionally, prediction accuracy has increased from less than 80 percent to 87.5 percent.
Learn more about how Lenovo uses ML to drive better business outcomes every day.