Data Science for Transportation

Abstract The growing deployment of vehicles equipped with driver assistance and automated driving systems presents new challenges for crash record classification, as existing police-reported databases vary considerably in the completeness and consistency of automation-level metadata. This study benchmarks five tabular machine learning and deep learning models: random forest, XGBoost, MambaAttenti…

Abstract In this paper, the existence of a statistically significant relation between microscopic traffic variables that reflect the individual driving behavior with macroscopic traffic dynamics at network level is investigated. Empirical evidence and machine learning techniques are used, applied on a publicly available dataset of vehicle trajectories recorded using Unmanned Aerial Units in the c…

We introduce a topological data analysis (TDA) framework to characterize departure‐delay co-occurrence in Japan’s domestic airline networks. By constructing a delay‐based filtration on daily delayed flight networks for All Nippon Airways (ANA) and Japan Airlines (JAL), we track how airports form delay co-occurrence loops through persistent homology using Vietoris–Rips complex filtration technique…

Abstract This study paved the way for developing digital twins of smart and emerging urban mobility systems, using shared mobility services such as ridesharing as a key case study. As cities contend with challenges, such as traffic congestion, environmental sustainability, and transportation equity, shared mobility platforms (e.g., UberPOOL and Lyft Shared) have emerged as promising solutions. Le…

Abstract In this paper, we present a novel approach for real-time detection of non-recurrent traffic patterns in urban roadway networks leveraging advanced machine learning techniques explained by traffic flow theory. The methodology comprises two key components. First, an LSTM-based Autoencoder is employed to extract typical expected traffic patterns from raw traffic data. Second, a clustering t…

Abstract Data science in transportation networks (DSTNs) refers to using diverse types of spatio-temporal data for various transportation tasks, including pattern analysis, traffic prediction, and traffic controls. Graph neural networks (GNNs) are essential in many DSTN problems due to their capability to represent spatial correlations between entities. Between 2016 and 2024, the notable applicat…

Accurate network macroscopic fundamental diagram (NMFD) estimation requires accurate spatial mean speed estimates that are hard to get from loop detector devices (LDD) as they capture local speeds only. This study introduces a correction method able to reconstruct the mean speed from loop data leveraging floating car devices (FCD) during the training phase. This significantly improves LDD-based N…