A state-of-the-art review on machine learning based municipal waste to energy system

Abstract

Municipal waste refers to a pool of different byproducts generated from domestic activities both in rural and urban areas. It is critical to consider strategies to effectively manage and treat municipal waste by establishing a waste-to-energy (WTE) system. However, waste-to-energy industries are facing several obstacles, including disruptive technologies, stringent government regulations, and some underdeveloped technological aspects. That is why, the researchers conducted a state-of-the-art review that aims to explore how machine learning models in WTE contribute to the achievement of sustainable development goals; second to highlight the strengths and weaknesses of machine learning techniques, and lastly to point out and evaluate the capabilities and flaws in the entire process and operation of WTE system through the use of machine learning, which would serve as a benchmark for a sound decision and policy-making as well as the basis to look into the areas for improvement. Results showed that within WTE systems, machine learning has greatly aided in the achievement of sustainable development goals (SDGs) by streamlining operations, increasing productivity, lessening environmental impact, and improving decision-making. Moreover, machine learning highlighted to foucus on solutions related to corrosion and deterioration occurring in the waste incinerator, chemical pollution in mechanical pre-treatment, and maintaining only an optimal emission in the WTE facility based on the prediction accuracies of 80% and 94% respectively.