Smart Medication Disposal Systems: An IoT-Based Approachto Managing Unused Pharmaceuticals

Abstract

Disposal of unused pharmaceuticals improperly is a threat to human health and the environment. Current IoT -based disposal systems tend to be based on generic classifiers and non-adaptive routing approaches that are inadequate when it comes to sensor uncertainty, scalability, and sustainable optimization. To overcome these limitations, we propose an extended implementation of IoT -based framework that utilizes two new algorithms, a FuzzySense Neural Mapper (FSNM) and RepelPath Optimization Framework (RPOF). The framework starts with the multi-sensor data capture in internet-enabled smart disposal bin utilizing RFID, weight and chemical sensors. FuzzySense Neural Mapper applies fuzzy logic and neural feature extraction to handling uncertainty in indicator 1:::” FALSE submitted to heterogeneous sensors to robustly categorize pharmaceuticals into groups, expiration, and hazardous conditions. The structured outputs are then fed to the RepelPath Optimization Framework that uses repel-force-guided graph optimisation, multi-objective heuristics to calculate safe and eco-efficient disposal/recycling routes, whilst disallowing hazardous accumulation, improper dumping. Validation was performed on the Pharmaceutical & Biomedical Waste Dataset (Roboflow, open-access) that consists of annotated biomedical and pharmaceutical waste images. The presented results indicate that FSNM-RPOF pipeline has superior-quality performance in terms of classification accuracy, anomaly detection, and disposal-route efficiency, when compared to the baseline IoT approaches to pharmaceutical waste disposal, highlighting the potential of sequential fuzzy-neural mapping and repel-based optimization in sophisticated, sustainable pharmacological waste management.