Graduating doctoral students Yulin Liu and Stephen Wong presented their dissertation work May 1, 2020 at the ITS Berkeley Transportation Seminar.
Yulin Liu: Machine Learning for Trajectory Modeling in the National Airspace Systems: Data Mining, Inference, and Deep Generative Models
Abstract: In this thesis, we present data-driven methods and algorithms to address the emerging challenges for performance assessment and predictive analysis in the national airspace system (NAS). Among these challenges, routing efficiency has received more and more attentions in the research community due to its core position in aviation economics and environmental studies, which motivates us to seek answers to fundamental questions such as "what are the mechanisms and causes to inefficient en route operations?" and more importantly, "what can we do to improve it, presumably in a predictive way?" In this research, we have explored three aspects regarding these open questions:
a) Examine a macroscopic comparison for flight en route inefficiency among different departing airports, arrival airports, seasons, and flight lengths through large-scale fixed effect models;
b) Propose mechanisms and statistical learning methods to explore the causal reasons behind inefficient en route operations;
c) Develop an end-to-end deep generative recurrent neural network to predict real-time 4D aircraft trajectories based on high-fidelity methodological datasets.
Bio: Yulin Liu is a UC Berkeley Civil and Environmental Engineering PhD candidate. His research focuses on developing machine learning and deep learning algorithms in air traffic flow management, and has been working under the National Center of Excellence for Aviation Operators (NEXTOR II) consortium since 2016. Yulin received his B.S. in Civil Engineering and a second degree in Economics from Tsinghua University in 2015.
Stephen Wong: Compliance, Congestion, and Social Equity: Tackling Critical Evacuation Challenges through the Sharing Economy, Joint Choice Modeling, and Regret Minimization
Abstract: Evacuations are a primary transportation strategy to protect populations from natural and human-made disasters. Recent evacuations, particularly from hurricanes and wildfires, have exposed three critical evacuation challenges: 1) persistent evacuation non-compliance to mandatory evacuation orders; 2) poor transportation response, leading to heavy congestion, slow evacuation clearance times, and high evacuee risk; and 3) minimal attention in ensuring all populations, especially those most vulnerable, have transportation and shelter. As natural disasters and human-made events more severely impact populations due to factors including climate change, land development, and population shifts, local and regional governments across geography types (rural to urban) need to develop effective evacuation strategies that move all people to safety. To tackle these challenges, this research conducts an examination of three innovative opportunities:
1) the feasibility of the sharing economy and emerging mobility in evacuations;
2) the implications of evacuation choice-making, particularly joint choices; and
3) the suitability of alternative decision rules, specifically regret, to describe evacuee choice-making. Employing data from individuals impacted by disasters in the U.S., this research aims to build more resilient communities to handle acute shocks (i.e., disasters and non-natural hazards), as they relate to transportation and begins to develop empirically driven evacuation strategies for governmental agencies to prepare for, respond to, and recover from disasters.
Bio: Stephen Wong is a UC Berkeley Civil and Environmental Engineering Doctoral Candidate studying transportation engineering. He received a B.S. in Civil Engineering and second major in Sociology from Johns Hopkins University and an M.S. in Civil and Environmental Engineering from UC Berkeley. He will be finishing his Ph.D. in May 2020. His research focuses on understanding evacuation behavior from natural hazards and how to leverage shared mobility and sheltering resources to support disaster response and relief. He is currently a research fellow for the National Science Foundation Graduate Research Fellowship Program and the Dwight D. Eisenhower Graduate Research Fellowship Program. He is also a graduate student researcher for the California Resilient and Innovative Mobility Initiative (CA RIMI), an initiative led by the University of California Institute of Transportation Studies (UC ITS). Stephen is a member of the Transportation Research Board (TRB) committee on Disaster Response, Emergency Evacuations, and Business Continuity. He is a member of the Emergency Management and Evacuation Working Subgroup for the International Association for Fire Safety Science. Stephen has also received the Eno Center for Transportation Future Leader Fellowship and the National Science Foundation Graduate Research Opportunities Worldwide (GROW) Fellowship.