E-waste management in Nepal: A case study overcoming challenges and opportunities

Abstract

The main objective of this paper is to design and implement a novel automatic E-Waste Management System using machine learning algorithms. A Waste Management System (WMS) is entirely manual, and it was extended by an IoT sensor-based alert system that can inform the overflowing of waste to the corresponding people for immediate action. It was not efficient in terms of time and cost. Also, WMS does not care about disposing of or recycling electronic wastes (E-waste). With the development of information technology, the use of electronic devices has considerably increased. Electronic devices like laptops, monitors, mobiles, headphones, tablets, and others depreciate in their value and use, adding to the amount of e-waste generated. Although this adds to the quantity of e-waste that threatens the environment, the brighter side is the possibility of extracting rich, recycled minerals. Various e-waste management systems have been invented for the optimal utilization and handling of e-waste. Surveys continue to study the possibility of utilizing e-waste using image processing techniques. The image processing is done with various deep-learning algorithms that provide better classification and prediction accuracy. This paper presents a deep learning model to improve the accuracy of e-waste prediction and management systems. The proposed deep learning model consists of a CNN algorithm for extracting features, and the extracted features are processed with an RBM model for better accuracy. An open-source dataset is considered for validating the proposed system, and the results obtained are compared with various existing approaches. The comparison shows that the proposed algorithm provides better prediction accuracy (96%) of the e-waste.