EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Optimizing Public Transport Infrastructure Through AI-Driven Reliability Prediction: A Data-Driven Approach

Ioannis Marios Andreadis, Georgios Georgiadis, Ioannis Politis

Smart Cities · 11.06.2026 · 9 (6) · 99

DOI: 10.3390/smartcities9060099 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Smart Cities, online erschienen am 11.06.2026. Verfasst von Ioannis Marios Andreadis und 2 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Public transport reliability largely determines the performance of smart urban mobility systems, as it directly affects passenger satisfaction and network efficiency. However, the strategic planning of public transport infrastructure is often carried out without dynamic, data-driven insights into operational performance, instead relying solely on static historical records of network operations. This study develops a data-driven framework based on the XGBoost machine learning algorithm to support the prioritization of infrastructure interventions by predicting delay severity and identifying reliability hotspots along an urban bus route. Delay severity is categorized into three classes (minor, moderate, and severe), using a model that incorporates spatial, temporal, operational, and meteorological variables. The XGBoost framework achieves a high predictive performance, with classification accuracies of 91.5% and 89.7% for the outbound and inbound bus route directions, respectively. Feature importance analysis indicates that seasonal and meteorological variables are critical factors influencing delay severity, highlighting the role of broader external environmental conditions on corridor performance. Furthermore, spatial analysis identifies specific bus stops with high delay probabilities, indicating hotspots where infrastructure upgrades should be prioritized at the stop and corridor levels. This study proposes a decision-support tool that enables targeted infrastructure investments at locations where they are most needed, contributing to more efficient and resilient public transport systems in smart cities.

Zitieren

APA

Andreadis, I. M., Georgiadis, G., & Politis, I. (2026). Optimizing Public Transport Infrastructure Through AI-Driven Reliability Prediction: A Data-Driven Approach. Smart Cities, 9(6), 99. https://doi.org/10.3390/smartcities9060099

Vancouver

Andreadis IM, Georgiadis G, Politis I. Optimizing Public Transport Infrastructure Through AI-Driven Reliability Prediction: A Data-Driven Approach. Smart Cities. 2026;9(6):99. doi:10.3390/smartcities9060099

BibTeX
@article{Andreadis2026,
  author = {Andreadis, Ioannis Marios and Georgiadis, Georgios and Politis, Ioannis},
  title = {Optimizing Public Transport Infrastructure Through AI-Driven Reliability Prediction: A Data-Driven Approach},
  journal = {Smart Cities},
  year = {2026},
  volume = {9},
  number = {6},
  pages = {99},
  doi = {10.3390/smartcities9060099}
}
Kostenlos starten →