Analytics tools are increasingly used for faster and better decision making in various business and management fields. For example, data mining techniques and predictive models are commonly used for effective marketing strategies and investment planning. In this business analytics project, we used 5-year hierarchical sales data from Walmart, the world’s largest company by revenue, to forecast daily sales for one month ahead. The project considered ten Walmart stores in the three US states (California, Texas, and Wisconsin). The dataset collected from each store included 3049 SKUs, 7 departments, 3 product categories, and 10 store details. In addition, the study used explanatory variables such as price, promotions, day of the week, and special events. Together, this robust dataset was used to build accurate and powerful forecasting models that can be used for pricing, inventory control, and logistics planning.
The result of this study will advance the theory and practice of business forecasting and operations. The methods developed can be applied in various business areas, such as customer segmentation, facility and capacity planning, and supply chain management. As business forecasting is essential and ubiquitous, this project will initiate our future work for more university-industry collaborations in business analytics field.
Faculty: Dr. Honggang Wang Students: Lei Wang, Allen Feng, Xiaoyi Lu, Jennifer Ware, and You Wang