Sales Forecasting solution for Overstocking & Stock Outs!

Customer Profile

India’s Leading retail Chain with more than 270 stores

Problem Statement

In fast-moving consumer retail, overstocking and stock-outs lead to spiraling costs, reduced profits & customer churn. Primary reasons for these two problems are inaccurate planning due to unavailability of the data-driven solution and unaccounted external factors such as weather, holidays, economic situation, public events, etc.

BluePi’s Solution

BluePi has built a Sales Forecasting solution that reduces

• Uncertainty in the inventory procurement

• Overstocking & stock-outs Using this system, the stores don’t have to rely on in-store manager’s intuition and instead use Artificial Intelligence. There are two parts of the solution:

• Reporting: Centralized system builds on Redshift generates and shares to all stakeholders.
• Forecasting: Forecast sales for a week in advance. Amazon Sagemaker is a great tool that we have used to build, train and deploy forecasting models

Value Added

• Per article category forecasting with an accuracy of 80%
• Reducing stock-outs and overstocking by at least 20%.

AWS Service used

EC2, S3, DMS, Redshift, Sagemaker