Special Issue: “Applications of Big Data in Public Transportation Systems”

Featured Application

This Editorial aims to summarize the contents of the five scientific papers included in the Special Issue “Applications of Big Data in Public Transportation Systems”.

1. Introduction

The landscape of urban planning and city morphology has undergone an unprecedented transformation in recent decades, largely driven by the emergence of big data [1,2,3]. As urbanization accelerates and population densities continue to increase, the demand for efficient and sustainable mobility solutions has become increasingly urgent. In this context, robust public transportation systems play a crucial role in supporting sustainable urban development. They provide a viable alternative to private vehicle use, alleviate traffic congestion, and contribute to reducing greenhouse gas emissions [4,5].

While the feasibility of collecting large-scale transportation data now far exceeds that of traditional survey-based methods, an important challenge remains: how can these vast data resources be effectively utilized to support practical and responsible decision-making? The rapid growth of available data does not automatically translate into improved planning or operational outcomes unless appropriate analytical frameworks are developed to extract meaningful insights.

This Special Issue was conceived to address the above challenge by bringing together recent studies that move beyond simple data collection toward advanced analytical applications in public transportation systems. The contributions explore how big data and modern analytical approaches, including machine learning, simulation modeling, and statistical analysis, can enhance the management and resilience of contemporary transportation networks, as well as our understanding of this field.

2. Summary of Contents

The five papers included in this Special Issue cover a diverse range of topics, including passenger flow prediction in rail systems, commuting behavior analysis, connected vehicle safety, transportation simulation and data completion, and bus travel time reliability. Collectively, these contributions demonstrate the growing role of advanced data analytics and artificial intelligence in tackling complex transportation problems.

2.1. Data Completeness and Reliability in Transportation Analytics

A primary challenge in the use of large-scale transportation data is its inherent incompleteness and uncertainty. Missing trajectories, sparse observations, and inconsistent data quality may limit the effectiveness of data-driven analysis and operational applications. Two studies in this Special Issue address these technical challenges.

Wang et al. (2024) (Contribution 1) addressed incomplete location-based service data by proposing a dual-driven simulation framework. By integrating mesoscopic multimodal transportation simulations with empirical big data in Hangzhou, the authors demonstrated that simulation-based approaches can effectively reconstruct missing trajectories and improve data completeness. Their method achieved a travel mode identification accuracy of 95.3%, illustrating how simulation models can complement real-world data to enhance transportation analysis.

Yin et al. (2024) (Contribution 2) focus on improving the reliability of bus travel time predictions. Traditional deterministic prediction methods often fail to capture the inherent variability of urban traffic conditions. By combining bootstrap techniques with an analysis of driving style similarities across multiple bus routes, the authors developed a prediction interval approach that quantifies uncertainty in travel time estimates. This method provides passengers with more reliable information and enhances the credibility of public transit information systems.

2.2. System Resilience and Safety in Intelligent Transportation Systems

As transportation systems become increasingly digitalized and interconnected, maintaining system resilience and operational safety has become an important research priority. Disruptions in infrastructure or communication networks can significantly affect system performance, particularly in highly integrated intelligent transportation environments.

Fan et al. (2024) (Contribution 3) addressed the difficulty of predicting passenger origin–destination matrices during rail transit disruptions. Using a deep counterfactual inference framework, the authors estimated passenger flow patterns under both normal and incident conditions. Their analysis reveals the spatiotemporal propagation effects of operational incidents and indicates that power equipment failures generate the most significant delays within the network. The study demonstrates how advanced machine learning techniques can support more effective incident response and operational management in urban rail systems.

Nagy et al. (2024) (Contribution 4) examined the safety challenges associated with vehicle-to-everything communication systems under unreliable network conditions. By incorporating a safety indicator into vehicle control algorithms that accounts for both vehicle dynamics and communication network parameters, their proposed method enhances operational stability when network quality deteriorates. This work contributes to improving the robustness and safety of connected vehicle systems.

2.3. Travel Behavior Analysis Using Multimodal Transportation Data

In addition to operational applications, large-scale transportation datasets also provide new opportunities to revisit long-standing theories in travel behavior research. High-resolution and multimodal datasets allow researchers to analyze travel patterns with greater detail and accuracy than was previously possible.

Yang et al. (2024) (Contribution 5) used comprehensive multimodal transportation data from Portland, Oregon, to examine the long-debated commuting time paradox. Their analysis indicates that average commuting time has increased slightly despite improvements in travel speeds. This increase can be attributed to a shift toward slower, non-driving modes and an increase in commuting distances. The study highlights the importance of disaggregate analysis when evaluating changes in travel behavior and demonstrates how multimodal datasets can provide deeper insights into urban mobility dynamics.

3. Conclusions

The studies included in this Special Issue demonstrate that the value of big data in transportation lies not merely in its scale, but in its effective application through appropriate analytical methods. From reconstructing missing mobility trajectories and improving travel time prediction reliability to analyzing system resilience and revisiting established behavioral theories, they illustrate how advanced data-driven approaches can transform raw data into actionable insights for transportation planning and operations.

As urban transportation systems continue to evolve in response to technological innovation and growing mobility demands, the integration of big data analytics, intelligent modeling techniques, and multidisciplinary perspectives will become increasingly important. We hope that the research presented in this Special Issue will stimulate further studies aimed at enhancing the resilience, transparency, and efficiency of public transportation systems.