Retailers often find it hard to carry out demand planning for different product categories. This project provides a good time-series approach to demand forecasting that can help with inventory and replenishment optimization resulting in reduced operational costs. The project designs a predictive model using lightGBM
This is a topic I learned from kaggle competition during my final semester as a Masters student in Computer Science at Santa Clara University. This competition is an exercise to explore different time series techniques used for forecasting. The project determines the appropriate demand forecasting technique for retail stores factoring in the seasonality aspect. This specific approach was referenced from this kaggle submisssion with some modifications.
Demand Forcasting for a Retail Store.ipynb - a notebook that implements the modeling process
https://towardsdatascience.com/machine-learning-for-store-demand-forecasting-and-inventory-optimization-part-1-xgboost-vs-9952d8303b48 https://www.kaggle.com/ekrembayar/store-item-demand-forecasting-with-lgbm