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) technologies with artificial intelligence (AI) has significantly improved pharmaceutical inventory management by enabling smarter and more efficient systems. Traditional pharmacy stock management approaches often face challenges such as inaccurate tracking, frequent drug shortages, and medication wastage due to the lack of real-time monitoring. IoT-based smart pharmacy systems address these issues by utilizing connected devices such as sensors, RFID tags, and smart shelves to continuously monitor inventory levels. The large volumes of data generated through these systems can be effectively analyzed using advanced deep learning techniques. Recent studies highlight the use of Siamese Heterogeneous Convolutional Neural Networks (SHCNNs), which can learn similarity patterns between historical consumption and real-time inventory data to support demand forecasting, anomaly detection, and intelligent stock classification. By combining IoT infrastructure with SHCNN-based analytics, pharmacies can significantly reduce stock shortages and minimize losses due to expired medicines. This review analyzes research from 2020 to 2023, emphasizing AI-driven inventory optimization approaches, while also identifying key challenges such as data security, interoperability, and implementation costs, and suggesting future directions like hybrid AI models and edge-based analytics.