Senior Lecturer of Industrial and Systems Engineering Tao Ma. Image/Dani Orlando
Tao Ma is dedicated to leveraging advanced data analytics and machine learning techniques to revolutionize transportation systems and tackle urban mobility challenges. Previously a lecturer at Texas State University, Ma completed his Ph.D. in transportation systems engineering from the University of Toronto and held postdoctoral positions at the Technical University of Munich and The University of Texas at Austin. His research focuses on network-level traffic flow prediction, transportation systems modeling, and the application of statistical methods to urban mobility challenges. He joins USC Viterbi this year, where he will be teaching Predictive Analytics, Data Mining and Temporal & Spatial Data Analytics in the Daniel J. Epstein Department of Industrial and Systems Engineering.
USC Viterbi spoke to Ma about his background, research interests and what he hopes to bring to USC’s academic community.
How do your previous teaching experiences at Texas State University and the Technical University of Munich inform/influence your approach to teaching at USC?
Reflecting on my previous teaching experiences is crucial for shaping effective teaching practices at USC. It helps to refine teaching methods and improve overall effectiveness. Different student populations and campus environments usually exhibit different characteristics, which requires an adaptable instructional approach. Previous teaching experiences enable me to quickly understand and get acclimated to USC students’ values, academic needs, expectations, and campus culture. Leveraging diverse teaching experiences helps me develop new teaching methods and create a more engaging and effective learning environment to better serve USC students.
Can you share an example of how you incorporate real-world applications into your engineering courses?
Integrating real-world applications into engineering courses can significantly enhance student engagement and understanding. An example is the traffic prediction project as a component of Intelligent Transportation Systems (ITS) where the students were asked to develop a traffic prediction model. This project aims to address real-world issues such as traffic congestion and route optimization, as well as serve a purpose for urban traffic management and operations. Students work in teams to create a machine learning model that predicts traffic state based on historical and real-time sensor data. This project helps students apply theoretical knowledge to practical problems, enhances their data analysis and machine learning skills and provides insights into the complexities of real-world traffic management.
What do you find most rewarding about teaching engineering subjects like Operations Research and Statistics?
Teaching engineering subjects like Operations Research and Statistics is incredibly rewarding. These subjects equip students with powerful analytical tools to tackle complex problems and help them develop the ability to break down intricate issues into manageable parts and find optimal solutions. Operations Research and Statistics have vast real-world applications where students apply theoretical concepts to real-world scenarios and make a tangible impact. These subjects encourage critical thinking and data-driven decision-making. It’s rewarding to witness students evolve into critical thinkers who can approach problems methodically and make informed decisions based on data analysis. Operations Research and Statistics intersect with various fields such as economics, computer science and management. This interdisciplinary nature allows for broadening students’ perspectives and their way of thinking. On the other hand, teaching these subjects also means continuous learning for me. These fields are always evolving with new methodologies and technologies, keeping me engaged and intellectually stimulated.
Can you explain a recent project you’ve worked on in network-level traffic flow prediction?
I proposed and developed new predictive methods, i.e., functional time series and functional Neural Network approaches, that can accurately forecast traffic flow across an entire network of highways and urban roads, helping to manage congestion and optimize traffic management systems.
A functional time-series approach treats traffic data as continuous functions over time, allowing for daily traffic curve-based predictions.
A functional Neural Network approach leverages neural networks to capture complex, continuously evolving patterns and spatial relationships in the traffic data, providing robust predictions across the network.
This project highlights the potential of advanced statistical and machine learning techniques in addressing real-world traffic management challenges.
How do you see your research in transportation systems modeling and simulation benefiting urban planning and traffic management?
My research in transportation systems modeling and simulation can significantly benefit urban planning and traffic management. By accurately predicting traffic patterns and congestion points, urban planners can design more efficient road networks and traffic control systems. This helps in reducing bottlenecks and improving overall traffic flow. Simulation models can evaluate the impact of new infrastructure projects, such as roads, bridges and public transit systems, before capital investment. This allows decision-makers to make informed decisions, optimizing resource allocation and minimizing disruptions. Simulation tools can be integrated with real-time traffic data to provide dynamic traffic management solutions, such as adaptive traffic signal control, real-time route guidance and incident management. Overall, my research aims to provide comprehensive insights and practical solutions that enhance the efficiency, safety and sustainability of urban transportation systems.
Are there any new research directions or collaborations you’re excited to explore at USC?
There are several exciting research directions and potential collaborations I’m looking forward to exploring at USC, such as big data analytics, sustainable transportation, intelligent transportation systems, autonomous vehicles, smart urban mobility and public policy.
Published on September 30th, 2024
Last updated on September 30th, 2024
