Recent Advances in IoT-Based Smart Pharmacies for Optimizing Stock Management with Siamese Heterogeneous Convolutional Neural Networks: A Systematic Review

Abstract

The integration of Internet of Things (IoT) and artificial intelligence (AI) has significantly improved pharmaceutical inventory management by addressing the limitations of traditional systems, such as inaccurate tracking, drug shortages, and medication wastage. IoT-enabled smart pharmacy systems use technologies like sensors, RFID tags, and smart shelves to continuously monitor inventory in real time. These systems generate large volumes of data, which can be analyzed using advanced deep learning techniques to enhance decision-making. In particular, Siamese Heterogeneous Convolutional Neural Networks (SHCNNs) have shown strong potential in optimizing stock management by learning similarity patterns between historical consumption data and real-time inventory records. This enables accurate demand forecasting, anomaly detection, and intelligent classification of pharmaceutical stocks, ultimately reducing shortages and minimizing losses due to expired medications.This review examines recent developments in IoT-based smart pharmacy systems with a focus on AI-driven inventory optimization techniques. The findings highlight that integrating IoT infrastructure with deep learning models significantly enhances supply chain efficiency and inventory accuracy. Despite these advantages, challenges such as data security, system interoperability, and high implementation costs remain barriers to widespread adoption. Future research should focus on developing hybrid AI models and leveraging edge-based IoT analytics to improve scalability, reduce latency, and ensure reliable performance in real-world pharmacy environments.