AI Powered Business Forecasting System Using ML and Dashboard

Abstract

This research work presents an advanced forecasting framework that aims to address inefficiencies in the supply chain and prevent any inventory imbalance using machine learning algorithms. Built upon an optimally designed XGBoost algorithm, this framework analyzes multi-year transactional information along with high-frequency time-series information for capturing volatile market conditions. The backend algorithmic framework is effectively integrated using a web-based Streamlit application which consolidates all past data along with future sales trends into live metrics. Through experimental analysis, it is demonstrated that the proposed framework empowers retail store managers, independent of programming skills, to set their optimal inventory levels.