Dr. Haoning Xi

Lecturer (Assistant Professor)

The University of Newcastle

My research lies at the intersection of Engineering and Business, utilizing my expertise in Data Analytics to transform mobility services. Focusing on Shared Mobility, Mobility-as-a-Service (MaaS), and Public Transport, my research aims to enhance these systems through innovative data-driven approaches. My core expertise includes Data Analytics, Business Intelligence, Operations Research, Statistics, Algorithm Design, Game Theory, Data Mining, Machine Learning, and Deep Learning.

Shared Mobility and Mobility-as-a-Service
My research aims to revolutionize traditional mobility services by integrating data analytics and advanced algorithms. By examining millions of smart card transactions from buses, ferries, trams, and trains, my research uncovers valuable insights into user travel behaviors and preferences. This understanding allows for the development of personalized weekly and monthly subscription bundles, combining various transport services with additional non-mobility options such as restaurant coupons, fitness memberships, and more. This approach seeks to enhance accessibility, customization, and overall user satisfaction.

Predictive Analysis and Intelligent Decision Making
By leveraging sophisticated AI and Machine Learning algorithms, my research focuses on predictive analysis to forecast peak travel times, potential bottlenecks, and service disruptions. This capability enables transport authorities to implement proactive measures, ensuring uninterrupted services and optimized operational efficiency. The insights derived from predictive analysis inform intelligent decision-making for transport scheduling, route optimization, and infrastructure investments. Real-time data and advanced algorithms allow mobility systems to dynamically adjust service schedules and resource allocations, achieving a balance between service availability and operational efficiency.

Data-Driven Optimization
Data-driven optimization, such as “smart predict-then-optimize” and “end-to end predict-then-optimize” models are central to my research. By continuously analyzing and applying data insights, transport services can be dynamically optimized to improve operational performance and user experiences.