A reinforcement learning framework using pointer networks to address tampering of pollutant control systems in freight transportation

Bidgoli, Masoumeh Messi and Demir, Emrah ORCID: https://orcid.org/0000-0002-4726-2556 2025. A reinforcement learning framework using pointer networks to address tampering of pollutant control systems in freight transportation. Expert Systems with Applications , 129178. 10.1016/j.eswa.2025.129178 Item availability restricted.

PDF - Accepted Post-Print Version Restricted to Repository staff only until 27 July 2026 due to copyright restrictions. Download (1MB)

Abstract

Road freight transportation, heavily reliant on diesel fuels, is a major source of emissions (i.e., air pollutants and greenhouse gases). To address this, governments enforce regulations like type approval, regular roadworthiness tests, and technical roadside inspections. While type approval ensures new vehicles meet emission standards, emissions worsen as vehicles age or when emissions control systems are tampered with or removed. Similarly, roadside screening is a timely, cost-effective way to detect high-emitting vehicles. Authorities can also identify tampered vehicles that fraudulently pass inspections but emit heavily in real-world conditions. However, logistics service providers (LSPs) often prioritize cost savings over environmental concerns and resort to vehicle tampering instead of fleet upgrades. Using the Stackelberg game to analyze decision-makers ’ behavior, this paper explores the conflict between LSPs and government agencies. A deep reinforcement learning (DRL) algorithm with pointer networks (PN) is proposed to solve this complex problem. Computational results from randomly generated instances highlight that the algorithm significantly outperforms existing methods and achieves convergence reasonably.

Item Type:	Article
Date Type:	Published Online
Status:	In Press
Schools:	Schools > Business (Including Economics)
Publisher:	Elsevier
ISSN:	0957-4174
Date of First Compliant Deposit:	26 July 2025
Date of Acceptance:	25 July 2025
Last Modified:	28 Jul 2025 11:45
URI:	https://orca.cardiff.ac.uk/id/eprint/180051

Actions (repository staff only)

Edit Item