Accurate demand forecasting is an important part of supply chain management, allowing businesses to optimize inventory, be less wasteful, and improve customer satisfaction. However, it is very challenging to predict demand accurately because of the complexity in seasonal variations, holiday periods, promotional activities, and sales trends that change with time. Our study addresses this challenge by analyzing the comprehensive sales dataset of Walmart, which brings in unique insights due to its extensive network of stores across a huge variety of locations. The real strength of this dataset lies in its rich historical data that captures how, in real-world retail, dynamics, seasons, and consumers’ purchasing behaviors change across different product categories and price points. Modern machine learning approaches are contrasted with traditional time series methods in our research. The four models presented in this paper include Random Forest with feature engineering, Random Forest with parameter tuning, Gradient Boosting, and SARIMAX (Seasonal AutoRegressive Integrated Moving Average with eXogenous regressors) with exogenous variables. To improve model performance, we incorporated external factors into the models: holiday periods and promotional events through feature engineering. We also tried optimization of parameters in individual models and ensemble modeling to improve the accuracy in predicting. Of special value is the SARIMAX model for its ability to include external variables in its time series framework. For a comprehensive evaluation, we used several performance measures such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R-squared values. The conclusion presents a necessity for the application of domain knowledge through feature engineering as well as the potential to mix traditional time series analysis with modern machine learning techniques. This research gives useful information regarding ways of improving forecasting in retail supply chains. It allows for data-based solutions regarding better inventory management and planning toward customer demand.