On April 20th, 2020, Inon Peled successfully defended his Ph.D. thesis entitled “Machine Learning Methods for Transportation under Uncertainty.“
He was supervised by prof. Francisco Pereira and Prof. Justin Dauwels (NTU).
Examiners were Prof. Pierre Pinson (DTU, chairman), Prof. Hans van Lint (TU Delft), and Prof. Constantinos Antoniou (TUM).
Find below the abstract of the thesis! For more information, feel free to contact Inon at firstname.lastname@example.org.
Transportation is rife with uncertainty, e.g., due to sudden disruptions and incomplete knowledge. Properly modelling this uncertainty is thus crucial for effective Transport practitioning. Fortunately, Transportation is also a rich source of data, from which Machine Learning models can extract useful patterns. The works in this Ph.D. thesis deal with Machine Learning methods for handling uncertainty in Transportation. In these works, we find that recent technological advances can alleviate the degradation of data-driven prediction models under road incidents, for which we offer a dedicated framework. We also advise to explicitly model an inherent limitation in Transportation demand observations, for which we offer two non-parametric alternatives. For dynamic operation of shared mobility services, we demonstrate the benefits of preserving a full uncertainty structure of demand, and we also quantify the relationship between predictive quality and subsequent service optimization.