<![CDATA[The escalating volume of waste, driven by insufficient public awareness and the lack of effective waste management systems, has become a significant environmental challenge. Improper waste disposal, particularly of non-biodegradable materials like plastics and metals, contributes notably to environmental pollution. In Riau Province, Indonesia, where the annual waste generation reaches 862,013 tons, only 60.71% is effectively managed. Despite the presence of 267 waste bank units, much of the plastic bottle and aluminum can waste remains improperly discarded. The high cost of commercially available Reverse Vending Machines (RVMs) further limits their widespread adoption, especially in regions like Riau. This study addresses these issues by proposing a cost-effective RVM that integrates Internet of Things (IoT) technology and a deep learning-based image classification model. The system enables users to exchange waste bottles for rewards through the Ecocycle mobile application, thus promoting waste sorting and recycling. The proposed model, tested on plastic and aluminum bottles, achieved 100% classification accuracy. Notably, this research bridges a critical gap by combining automated classification with IoT communication and incentive distribution in a low-cost, scalable system. The potential for this system to be expanded globally is evident, as it provides a feasible solution for large-scale waste management, particularly in regions lacking advanced waste infrastructure. Through both technological and behavioral approaches, this study contributes uniquely to the field of waste management by advancing accessible, effective solutions to foster environmental sustainability.]]>
Copyrights © 2025
