In this lab, we integrated a Business Intelligence layer into the Pine Valley Furniture Company (PVFC) system by implementing a Recommendation System. The system utilizes historical order data to provide personalized product suggestions to customers and demand forecasting insights to management.
The implementation focuses on three primary functionalities:
Products.aspx and Orders.aspx pages.
When a user selects a product, the system queries the database to find items frequently purchased together with the selected item using Association Rule Mining principles.
DemandForecasting.aspx dashboard for managers.
It analyzes the correlation between "Primary" items (e.g., Dining Tables) and "Secondary" items (e.g., Dining Chairs) to forecast inventory needs when sales of primary items spike.
The system consists of an ASP.NET Web Forms frontend communicating with a backend SQL Server database.
Complex logic for association rules and forecasting is handled via optimized SQL queries and VB.NET code-behind logic.
We also explored the use of a dedicated Web API (PVFCRecommendationAPI) for decoupling the recommendation engine from the main website logic.
We maintained the "Premium" aesthetic of the PVFC website while adding non-intrusive recommendation banners. The demand forecasting charts provide a clear visual representation of expected vs. actual sales, helping in effective inventory planning.