[Archimedes Talk] Incident Management Modelling via AI - Prof. Simona Mihaita

 

Prof. Simona Mihaita (University of Technology Sydney, Australia)

This talk will delve into the transformative potential of AI modeling in Traffic Incident Management. It will begin by outlining the critical problem of efficiently predicting and managing traffic incidents. The discussion will then address the significant challenges faced, such as accident reporting errors, high dimensionality and lack of automated process.

The core of the talk will focus on innovative methodologies designed to tackle these issues. First, it will introduce the "Universal Bi-Level Framework for Traffic Incident Duration Prediction," a novel approach that enhances prediction accuracy. Next, the talk will explore "Data Fusion Techniques for Enhanced Incident Modelling," which integrates diverse data sources to improve the modeling of incidents including text. Lastly, it will cover "Methodologies for Traffic Disruption Segmentation," which aims to build an automatic procedure for detecting accidents in advance of their reporting time.

The session will conclude with a summary of how these innovations collectively advance traffic incident management, offering a more responsive and efficient approach to handling disruptions on the road.

 

Dr Mihaita is an Associate Professor at the University of Technology in Sydney, leading the Future Mobility Lab. Her research focus is to improve people's movement via artificial intelligence in a smart city context. Dr Mihaita holds several leadership roles securing funding over $5M in initiatives such as the Australian-Singapore Strategic Collaboration via a joint NRF-ARC Linkage Project, the "Premiere's Innovation Initiative" for easing congestion leading to ICMP (Intelligent Congestion Management Program), the "On-Demand Mobility" trials in Northern Beaches with Keolis Downer, the impact of Electric Vehicles with AEMO, the Automatic Road Star Rating with TomTom, the positioning accuracy of connected trucks with TfNSW, etc.  

 

She is a winner of the 2023 iAWARDS – for Digital Twin modelling, 2022 Women in AI Awards – Infrastructure category, a triple finalist in the 2021 IoT Awards – Smart Cities, 2021 ITS Australia for Transport Data Excellence and for R&D Excellence, the winner of the 2018 ITS Australia National Awards for Research Excellence, nominated TfNSW Expert panel member, finalist in Smart City Awards 2019, awarded Monash Business School Scholarship for Women in Leadership 2018, nominated for the 2019 Prime Minister's Science Awards, and interviewed by Channel 7 News and The Guardian.  

 

So far, Simona has 56 peer reviewed publications, out of which 14 in IEEE venues (including IEEE ITSC, and the top journal in transportation science – IEEE Transactions on ITS). Her publication rate since joining UTS in 2019 is 10.6 publications per year. Her Google Scholar citation score is 1130, h-index =16. She is cited by academics from University of College London, University of Calgary Canada, from USA, Australia, India, and Malaysia.