Title: Intelligent Departure Metering Advisory Tool (I-MATE) for Managing Airport Airside Congestion
Speaker: Dr. Hasnain Ali, NTU Singapore
Date and time: 6 May 2025 (Tuesday), 9:30-10:30 a.m.
Venue: IEOR Seminar Room
Abstract:
Airport taxi delays pose significant challenges to airlines, passengers, and environmental sustainability. Departure Metering (DM) is an effective method to contain taxi delays by controlling departure pushback timings. In this work, we demonstrate the potential of Deep Reinforcement Learning (DRL) based DM method to reduce taxi delays by effectively transferring delays from taxiways to gates. We formulate the DM problem as a Markov Decision Process (MDP) and train a DM policy using simulations generated from historical airport surface movement data. Building on this, we develop an Intelligent Departure Metering Advisory Tool (I-MATE) that employs the trained DM policy to recommend pushback advisories to Air Traffic Controllers (ATCOs). Furthermore, we conducted validation experiments to assess the efficacy and acceptability of I-MATE in assisting ATCOs to manage airside traffic. The study revealed a spectrum of compliance with I-MATE recommendations among ATCOs, highlighting the complex interplay between human behavior and AI-driven decision support systems. Results demonstrated a 25.6% reduction in taxi delays with increased compliance with I-MATE advisories, though a slight 2.7% decrease in runway throughput was observed. While ATCOs rated I-MATE highly for usefulness and reliability, concerns regarding transparency and explainability emerged. ATCO feedback also emphasized the importance of considering individual differences and human factors in the design of such systems. Overall, this research underscores the promise of AI-based decision support systems in managing complex, dynamic, and high-stakes environments like airport airside.
Bio:
Hasnain Ali received the Master’s degree from the Indian Institute of Technology Delhi (IIT Delhi), in 2018 and the Ph.D. degree from the Nanyang Technological University (NTU) Singapore, in 2022. His research lies at the intersection of machine learning, optimization, and decision-making under uncertainty, with a focus on Human-AI collaboration in complex operational environments. He currently serves as a Research Fellow and Team Lead at the Air Traffic Management Research Institute (ATMRI), NTU Singapore, where he leads applied research projects on autonomous airport operations. He is a recipient of several prestigious awards, including the NTU Research Scholarship, the Mistletoe Research Fellowship, and the Singapore Data Science Consortium Fellowship, in recognition of his impactful, real-world research in intelligent transportation systems.