Best Paper Award at 14th APTE 2025

Lee Kong Chian Faculty of Engineering and Science (LKC FES) academic Ir Prof Dr Khoo Hooi Ling was honoured with the Best Paper Award at the 14th Asia-Pacific Conference on Transportation and the Environment (APTE 2025), held from 9 to 11 August 2025 in Hangzhou, China. The international conference was organised by Tsinghua University, China.

Prof Khoo’s award-winning paper, titled Multi-agent Reinforcement Learning for Traffic Signal Control: Comparison with Centralized, Decentralized and Hybrid Approaches, was one of only four papers selected for this prestigious recognition. The research team comprised LKC FES academics Prof Khoo, Dr Ong Chuan Fang, Prof Ts Dr Yau Kok Lim and PhD student Ong Lip Wei. Their study was supported by the Ministry of Higher Education (MOHE) under the Fundamental Research Grant Scheme (FRGS).

The paper explores the use application of multi-agent reinforcement learning (MARL) to improve urban traffic signal control, a crucial aspect of intelligent transportation systems. By comparing centralised, decentralised, and hybrid approaches under a unified experimental framework, the study demonstrated that the Value-Decomposition Network (VDN) hybrid model achieved superior stability, coordination, and reduced vehicle waiting times compared to traditional systems. The findings contribute significantly to the advancement of data-driven, adaptive traffic management strategies suitable for smart city applications.

Expressing her delight over the achievement, Prof Khoo said, “Winning the Best Paper Award is a deeply gratifying experience. It represents the culmination of persistent effort, teamwork, and dedication towards advancing intelligent traffic control research. This recognition reaffirms that our research direction in integrating artificial intelligence with real-world traffic management has both academic and societal value.”

She added, “This recognition adds strong validation and visibility to my ongoing research in intelligent and sustainable mobility systems. It underscores the relevance of applying advanced computational intelligence, such as multi-agent reinforcement learning, to address real-world transportation challenges. The award not only strengthens our research group’s reputation but also encourages us to continue pursuing impactful, policy-aligned studies that contribute to smarter and more adaptive traffic solutions for Malaysia and beyond.” 

The Best Paper Award celebrates the team’s outstanding contribution to intelligent transportation research, highlighting UTAR’s continuous efforts to advancing innovation and sustainable solutions aligned with global smart city initiatives.