🤖 AI Summary
To address coarse-grained customer journey analysis, fragmentation between inventory and sales data, and the difficulty in quantifying RFID value in retail, this paper proposes an end-to-end analytical framework integrating RFID trajectory data with point-of-sale (POS) transaction records. It introduces a lightweight ETL architecture and a behavior–product co-occurrence–oriented logical data model to construct a multidimensional spatiotemporal data warehouse. Furthermore, a Balanced Scorecard (BSC) is adopted to systematically evaluate RFID deployment impact across four dimensions: financial performance, customer experience, internal processes, and learning & growth. Experiments demonstrate that the framework improves inventory count accuracy to 99.2%, reduces shrinkage in high-risk zones by 37%, and identifies three distinct customer path–conversion rate association patterns. The approach delivers a reusable methodology and empirical validation for data-driven, fine-grained retail operations.
📝 Abstract
The article focuses on researching a system for processing and analyzing tracking data based on RFID technology to study the customer journey in retail. It examines the evolution of RFID technology, its key operating principles, and modern applications in retail that extend beyond logistics to include precise inventory management, loss prevention, and customer experience improvement. Particular attention is paid to the architecture for data collection, processing, and integration, specifically the ETL (extract, transform, load) methodology for transforming raw RFID and POS data into a structured analytical data warehouse. A detailed logical database model is proposed, designed for comprehensive analysis that combines financial sales metrics with behavioral patterns of customer movement. The article also analyzes the expected business benefits of RFID implementation through the lens of the Balanced Scorecard (BSC), which evaluates financial performance, customer satisfaction, and internal process optimization. It is concluded that the integration of tracking and transactional data creates a foundation for transforming retail into a precise, data-driven science, providing unprecedented visibility into physical product flows and consumer behavior.