Title: Optimising e-waste management and warranty fraud detection through ensemble machine learning in remanufacturing

Authors: Elif Kongar; Gazi Murat Duman; Surendra M. Gupta

Addresses: Department of Economics and Business Analytics, University of New Haven, Orange Campus, 584 Derby-Milford Rd., Orange, CT 06477, USA ' Department of Economics and Business Analytics, University of New Haven, Orange Campus, 584 Derby-Milford Rd., Orange, CT 06477, USA ' Mechanical and Industrial Engineering, Northeastern University, Boston, MA 02115, USA

Abstract: Fraud in warranty claims can lead to substantial financial losses and reputational damage for companies. Previous research identified various sources of fraud, including customers, service agents, and third-party providers. This study explores the detection of warranty fraud in the remanufacturing industry by leveraging an ensemble machine learning approach. The proposed methodology integrates random forest, gradient boosting machine (GBM), and XGBoost algorithms to enhance fraud detection accuracy. The synthetic dataset simulates real-world warranty return conditions incorporating variables such as claim amounts, service costs, and customer histories. The ensemble model combines multiple algorithms to improve fraud detection performance, achieving high predictive accuracy. Our findings highlight the effectiveness of advanced analytics in fraud detection while supporting sustainable remanufacturing practices and responsible electronic waste management.

Keywords: clean innovation; e-waste; fraud; predictive modelling; remanufacturing; risk management.

DOI: 10.1504/IER.2025.150084

Interdisciplinary Environmental Review, 2025 Vol.24 No.4, pp.369 - 384

Received: 12 Nov 2024
Accepted: 01 Mar 2025
Published online: 28 Nov 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article