This project involves the creation of code to generate synthetic data. In situations where access to real data is limited, generating synthetic data becomes the easiest means to test models. The project focuses on the problem of demand forecasting, which includes loading the dataset, performing data exploration and analysis, data cleansing, creating visualizations to draw insights about the data, performing scaling, encoding, and applying models such as Linear Regression and Random Forest along with cross-validation. Finally evaluation is done.