A Machine Learning Approach for Pyrolysis of Plastic Waste: An Overview of Types, and Applications

Pyrolysis pathways have emerged as one of the most promising approaches to managing plastic waste(PW). However, the commercial use of pyrolysis is hampered by a number of unresolved blind spots, andthe industrial use of such technologies has been limited. Even though global kinetic techniques have beenused for many years to explain pyrolysis models, new developments in computational modeling, suchmachine learning (ML), provide fresh perspectives. An overview of the types and applications of MLtechniques for pyrolysis of PW is the main topic of this review. The outcome shown that ML is a potent toolfor pyrolysis process improvement, interpretation, and predictive modeling. In addition to enablingreasonably accurate forecasting of product yields and properties, ML algorithms such as artificial neuralnetworks, genetic algorithm-artificial neural networks, support vector machines, and decision treetechniques can also give researchers insights into complex pyrolysis kinetics so that process parameters canbe further tuned to achieve desired results. All things considered, pyrolyzing PW using ML approaches canincrease product production and quality, reduce environmental effect, and promote sustainable PWmanagement initiatives.