A Comprehensive Study of Solid Waste Classification using Deep Learning and Machine Learning Techniques

Solid Waste Management (SWM) involves the strategic planning and procedural management of solid wastes, which is a key factor in sustainable development of municipalities from physical, logical, and legal-political perspectives. Municipal Solid Waste (MSW) is typically managed using four methods: composting, recycling, landfilling, and incineration for reuse. The absence of proper on-site solid waste treatment increases the burden on urban municipalities, as the growing volume of waste demands methods such as recycling and source separation, along with classification for disposal. In mining activities, large volumes of Mining-Associated Solid wastes (MASW) are generated across various stages including exploration, separation, concentration, crushing, and sorting. Machine Learning (ML) and Deep Learning (DL) techniques have been employed in solid waste classification to categorize wastes as organic and recyclable. The conventional ML and DL techniques include Artificial Neural Network (ANN), Support Vector Regression (SVR), Convolutional Neural Network (CNN), and Graph Long Short-Term Memory (GLSTM) employed for solid waste classification. The performance metrics of accuracy, precision, recall, and F1-score are used to evaluate the models’ effectiveness of ML and DL techniques in solid waste classification.