Machine Learning-Assisted Solid Waste Life-Cycle Management

¹ KPR Institute of Engineering and Technology, Coimbatore, Tamil Nadu, India
² Blue Hora University, College of Engineering and Technology, Ethiopia
³ Howest Global, Coimbatore, Tamil Nadu, India

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

The increasing rate of urbanization in India has increased the levels of plastic wastes in the cities, making the traditional method of managing wastes, which is based on a fixed route, inadequate to manage the situation. The overflowing bins, improper collections, fuel wastage, and inefficient sorting of wastes indicate the urgent need to manage plastic wastes in a more efficient, anticipatory manner. This paper proposes a machine learning framework for the efficient management of the plastic waste life cycle, thereby supporting anticipatory urban planning. The proposed framework correlates historical plastic wastes in cities with their respective demographic data, such as population size and growth rates, to make predictions about the plastic wastes that will be generated in the near future. The proposed framework uses Extreme Gradient Boosting to make predictions about the plastic wastes that will be generated in the near future, while K-Means clustering is used to cluster the cities based on their plastic wastes, thereby enabling anticipatory planning for each city individually. The proposed framework predicts that the city of Coimbatore in India will generate about 4,024.79 tons of plastic wastes every day by the year 2026. The proposed framework is highly efficient, but any unforeseen changes in the demographics of the city may impact the performance of the proposed framework, making it imperative to use it for other types of wastes as well.