Real-Time Demand Forecasting

The model which we have presented is a Linear Regression Model. In the results above we see that predictions can be done on the basis of the data available and is approximately accurate. An accurate forecast is very important for the demand planning team. The data used in this project and building the model is using the sales-in data for different stores. The important factor to be considered is the stability of the model and removing the game-playing. A community version of a platform is used to build the model. Linear Regression model is developed in pyspark. After the results are generated, dataframe of results is validated and generated and is sent backto the Azure SQL database to be used in Power BI.In the future work, different techniques will be considered and researched. Time-Series and Machine Learning to be built in one platform and check how the minimization of mse produces the forecast. The predictions can be hyper parameterized to give more accurately tuned results. Also, in the PowerBI report more measures and visualizations can be made on basis of individual's thought process.

Autorentext

Dr. Punit Gupta ist außerordentlicher Professor in der Abteilung für Computer und Kommunikation an der Manipal Universität Jaipur, Jaipur. Er erhielt 2010 einen B.Tech-Abschluss in Informatik und Ingenieurwesen von Rajiv Gandhi Prouduogiki Vishwavidyalaya, Madhya Pradesh.