Accurate sales forecasting is essential for effective decision-making in the retail sector, particularly for large-scale operations like Walmart. This study aims to build a robust machine learning model to predict weekly sales for Walmart stores using historical data from 2024 to 2025. The model incorporates key features such as temperature, fuel price, Consumer Price Index (CPI), unemployment rate, and holiday indicators to enhance forecasting accuracy. Advanced machine learning algorithms, including Linear Regression, Decision Trees, Random Forest, XGBoost, and Neural Networks, are employed to train and validate the model. Performance is assessed using evaluation metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and R² Score. The primary goals of this research are to support data-driven decision-making, optimize inventory management by minimizing stockouts and overstock scenarios, and improve business efficiency through better planning and resource allocation. The study demonstrates how predictive analytics can be leveraged to enhance strategic sales operations and profitability in the retail industry.