Utilizing Random Forest with iForest-Based Outlier Detection and SMOTE to Detect Movement and Direction of RFID Tags

In recent years, radio frequency identification (RFID) technology has been utilized to monitor product movements within a supply chain in real time. By utilizing RFID technology, the products can be tracked automatically real-time. However, cannot detect movement and direction of tag. This study investigates performance machine learning (ML) algorithms passive tags. The dataset this was created by considering variety conceivable tag motions directions that may occur actual warehouse settings, such as going inside out gate, moving close turning around, static statistical features are derived from received signal strength (RSS) timestamp Our proposed model combined Isolation Forest (iForest) outlier detection, Synthetic Minority Over Sampling Technique (SMOTE) Random (RF) shown highest accuracy up 94.251% compared other ML models detecting addition, we demonstrated classification could applied web-based monitoring system, so tagged move or through gate correctly identified. is expected improve on status (being delivered) automatically.