Multiobjective Optimization of Plastic Waste Sorting and Recycling Processes Considering Economic Profit and CO₂ Emissions Using Nondominated Sorting Genetic Algorithm II

Abstract

Plastic waste has become a severe threat to the environment as increasing amounts of plastic waste are generated every year. To solve this problem, it is crucial to increase the recycling rate of plastic waste with proper sorting and recycling processes. However, sorting and recycling costs vary depending on the specific process, and CO 2 is inevitably generated during recycling. Therefore, this study developed a novel multiobjective optimization model based on mixed-integer nonlinear programming to optimize plastic waste sorting and recycling processes according to target polymer types while maximizing net profit and minimizing CO 2 emissions. Recycling solutions were proposed using a nondominated sorting genetic algorithm, which enabled the selection of a portfolio of plastic recycling methods for each plastic type depending on the importance of the two objectives: maximum net profit and minimum total CO 2 emissions. As a result, Pareto-optimal solutions with a net profit distribution of 35− 1936 million USD/year and CO 2 emissions of 9.7−17.0 kt/year were obtained. Furthermore, the Pareto-optimal front was analyzed to provide representative optimal solutions, which can provide decision makers with a wide range of choices when determining process specifications.

Multiobjective Optimization of CO2 Emission and Net Profit for a Naphtha Cracking Furnace Using a Deep Neural Network with a Nondominated Sorting Genetic Algorithm

“…Thus, an optimization process using numerous case studies considering both the feed composition and product price requires a great deal of computational cost and might take several days. This problem can be solved by employing machine learning (ML) algorithms. − Hence, a deep neural network (DNN)-based model for the NCC cracking furnace is developed in this work to predict the product yield, and a nondominated sorting genetic algorithm II (NSGA-II) is used to solve the multiobjective problem Figure shows an overview of this study.…”