Authors: Assistant Professor I.Maria Anand Milani, Assistant Professor Dhanusha Mol K P
Abstract: Managing inventory in warehouses continues to pose a significant challenge to supply chain efficiency, as conventional bar code technologies involve manual efforts, line of sight issues, and batch processing problems. In this study, we introduce a novel RFID-based IoT Inventory Au-tomation System (RIAS), which consists of passive UHF RFID tags, reader portal stations, automated drones equipped with RFID readers, and a system for analyzing information on the edge-cloud. We present two new approaches: (1) multi-antenna collision resolution (MACR), which solves the problem of tag collisions and ensures 99.7% tag read rate at 1200 tags per sec-ond and (2) anomaly detection and localization (ADL) approach, based on received signal strength indication trilateration for locating misplaced items to within 0.5m. Applied to 12 ware-houses over 14 months, RIAS saves 91.4% of cycle counting time, increases inventory accuracy from 86.2% to 99.1%, and locates misplaced items within 2.3 minutes after their appearance.
