Real-time RFID-based item tracking using IoT & efficient inventory management using Machine Learning

Internet of Things (IoT) and Machine Learning (ML) based connected intelligence framework has become a technology enabler in various segments of supply chain like inventory management. In this paper, we have proposed an IoT-cloud architecture for passive RFID tag based real-time Stock Keeping Units (SKUs) tracking and ML algorithms for predictive stock analysis. The SKUs are fitted with RFID tags, those are scanned at entry and exit of the warehouse and this real-time data is sent to cloud server over internet. The acquired data is then processed using ML based software engine using techniques like classification, training and testing. We have considered, ABC Inventory Classification for the SKUs in warehouse and three ML algorithms as: Support Vector Machine (SVM), K-nearest neighbors (KNN) and Bayes for predictive analysis of SKUs in inventory. The result indicates SVM is outperforming with 84.8% accuracy. The accuracy of KNN is 83.6% and Bayes at 74%.