Journal of Decision Making and Healthcare

Electronic ISSN: 3008-1572

DOI: 10.69829/jdmh

Predicting plastic waste mismanagement with machine learning models

Journal of Decision Making and Healthcare, Volume 3, Issue 1, April 2026, Pages: 1–20

FRANCIS OGOCHUKWU OKPALA

Department of Mathematics and Computer Science, Brandon University, Manitoba, Canada

GAUTAM SRIVASTAVA

Department of Mathematics and Computer Science, Brandon University, Manitoba, Canada


Abstract

Predicting plastic waste mismanagement is critical in evaluating recycling processes and reducing the havoc of plastic waste crises. Mismanagement is a term synonymous to inefficient utilization and improper accountability of plastic wastes along the generation and consumption chains. Efficient tracking of these chains can aid efficient deployment of resources and safe rectification for efficient plastic recycling. Machine learning models are predominantly known to thrive under enormous high-quality data. Popular regions of plastic waste mismanagement like Africa are usually prone to lack of insufficient data hence a robust method is needed. This work utlizes Africa as a casestudy for tracking plastic waste mismanagement due to the unavailability of quality data to reasonably make predictions. Consequently, this work leverages the environmental and societal correlation in regions of Africa to create a correlative feature engineering technique that is leveraged by machine learning models to predict plastic waste mismanagement in Africa. To show the superiority of this work, we leveraged less efficient machine learning models with our technique to guarantee better predictions for regions of plastic waste mismanagement. Analytically, our approach shows an accuracy of \(97.8\%\) with SVM, \(94.2\%\), with cat boost while other algorithms like KNN and linear regression showed \(73.6\%\) and \(65.3\%\), respectively. This framework explains how analyzing plastic waste in Africa can be approached and how it results can be used by industry practitioners and related software vendors in establishing improved practices and tools for plastic management.


Cite this Article as

Francis Ogochukwu Okpala and Gautam Srivastava, Predicting plastic waste mismanagement with machine learning models, Journal of Decision Making and Healthcare, 3(1), 1–20, 2026