Transition From Manual Predictive Process To Automated Predictive Analytics For A Retailer

After going through customers current manual process, our advance analytics team proposed an architecture with automated data ingestion, data preparation, self-learning predictive models and delivery of predictive outputs via Power BI share and collaborate features. Following were the important activities performed during the project: - Followed Data Science Process Lifecycle (TDSP) for the end to end implementation - Prepared data sets for Sales, Promotion, Store Remodeling, Workforce(Full Time/Part Time) and Products. - As part of the first data collection process all the raw data were stored in Azure Blog Storage. Data were later feed to Azure Machine Learning flow where they were cleaned and fed to actual model for training and prediction. - Output published through Web Service which later utilized in Power BI and other application - The implementation of the retail dashboard Power BI solution