Study of Automated E-Waste Classification Techniques

Electronic waste (e-waste) is India’s fastest-growing waste, projected to surge by about 31% annually according to the Central Pollution Control Board. However, the Ministry of Environment reports a mere 33% recycling rate, leaving 67% of E-waste untreated. E-waste encompasses valuable and reusable materials, presenting an opportunity for economic gains through effective recycling. The efficiency of the entire recycling process depends on the first pre-processing step, the separation step. However, manual sorting of e-waste poses dangers which include exposure to contaminants, potential worker health problems, time intensive processes and prone to human error. Manual sorting also increases the risk of contact with toxic materials, which has a negative impact on worker wellbeing, also affecting the sorting efficiency. Thus, in this rapidly evolving landscape of e-waste management, innovative automated e-waste classification systems are implemented which utilizes machines and various algorithms for faster and accurate sorting improving the efficiency, accuracy and enhanced consistency in sorting reducing human efforts to a larger extent. This study undertakes a methodical examination of literature pertaining to diverse e-waste classification systems. It comprehensively explores the features of current systems, outlines their constraints, and contemplates potential avenues for future development. This paper provides an elaborate synopsis to assist the research community in crafting effective automated e-waste classification systems, leveraging deep learning and machine learning methodologies.