ML-based demand forecasting solution to improve forecast accuracy while reducing costs by 5x
Built an ML-based demand forecasting solution enhance visibility in the supply chain based on their market share reducing the data run-time
A leading alcoholic beverages company operating across 180+ countries was finding it challenging to estimate their market share in the industry. The customer had limited business datasets for market share estimation and relied on other sources such as Nielsen, IWSR, and more. The right estimation would help them predict demand and optimize supply chain to grow their business. The estimation required them to model around 200,000 time series, which is a sequence of data points, considering the micro and macro exogenous factors in a quick and coherent manner.
Sigmoid developed an agile demand forecasting solution by pulling data from Azure Data Lake storage, preprocessing it to remove outliers and running feature engineering on it. We trained several ML and deep learning models that ran parallely across all countries on Azure Kubernetes Service (AKS) to generate a trained model representing the behavioral aspects of the data to determine specific patterns. The trained model was used to create forecasts that were closely monitored through Grafana. To modularize the code, Sigmoid built the entire framework on Kedro that offers best practices for software engineering and makes the code easy to scale. In addition, we operationalized the final solution by parameterizing the end-to-end pipeline countrywise with auto-enabling features on AKS that scale up and down based upon the number of countries in the run.
Sigmoid helped the customer predict nearly accurate demand forecasting to optimize their supply chain, improve inventory management, reduce overhead cost, enhance collaboration with suppliers, and improve risk mitigation.
faster data run time
reduction in forecasting cost
reduction in compliance and overhead cost