Leveraging Advanced Technologies for (Smart) Transportation Planning: A Systematic Review

The Urban Mobility Challenge: Moving Beyond Disconnected Tech Pilots

Cities across the globe are grappling with the compounded pressures of vehicular congestion, transport-related carbon emissions, and roadway safety risks. While the vision of “smart mobility” has generated a surge in emerging technologies—from Internet of Things (IoT) traffic sensors and connected vehicles to machine learning models and full-scale urban digital twins—the actual academic and operational landscape has remained highly fragmented.

All too often, municipal authorities and tech providers launch isolated pilot projects that fail to address the systemic interdependencies of urban transport. How exactly do specific emerging technologies align with specific bottlenecks on the ground? And how can researchers systematically map these relationships beyond purely qualitative surveys?

Our open-access systematic review in Sustainability (MDPI), titled “Leveraging Advanced Technologies for (Smart) Transportation Planning”, addresses this question by combining the established PRISMA systematic literature protocol with an innovative Sentence-BERT (SBERT) natural language processing approach to quantitatively evaluate how advanced technologies intersect with real-world mobility challenges.

An International Collaboration: Pusan National University & University of Liverpool

This publication represents an international research collaboration between researchers at the Department of Urban Planning and Engineering at Pusan National University (Busan, South Korea) and the Geographic Data Science Laboratory at the University of Liverpool (UK).

The Korea-UK Digital Twin Partnership

This study builds on the ongoing Korea-UK Digital Transport Project supported by the Liverpool City Region Combined Authority (LCRCA), Podaris, and South Korean partners. The project aims to cross-pollinate computational methodologies between the polycentric, heritage-constrained urban network of the Liverpool City Region and the hyper-dense, transit-oriented coastal metropolis of Busan.

Read University of Liverpool announcement on project expansion

Methodological Innovation: Quantitative SBERT-Based Concept Mapping

Traditional systematic literature reviews rely predominantly on manual thematic coding, which can introduce subjective interpretation when mapping interdisciplinary literature. In this study, we coupled PRISMA screening guidelines with a transformer-based NLP workflow:

1. PRISMA Screening

Screened global peer-reviewed literature across Web of Science and Scopus (2013–2024), isolating 26 in-depth core studies on smart transport technology implementations.

2. SBERT Embeddings

Utilized Sentence-BERT to generate high-dimensional vector representations of challenges, technologies, and policy strategies extracted from the literature corpus.

3. Semantic Alignment

Computed cosine similarity matrices to objectively measure the quantitative alignment between specific transport problems and their enabling technologies.

Key Findings: Where Advanced Technologies Deliver the Greatest Value

The SBERT analysis uncovered distinct technological clusters and clear linkages showing where specific digital tools deliver measurable improvements:

Technology Cluster	Primary Transport Focus	Core Mechanism	Key Impact
Digital Twins	Network Simulation & Congestion	Dynamic spatial digital replicas mirroring real-time traffic	Enables scenario stress-testing before implementing physical road reconfigurations.
IoT & Sensor Networks	Real-Time Data Collection	Inductive loops, roadside cameras, GPS probe feeds, air quality sensors	Provides high-frequency telemetry required for proactive traffic management.
AI & Machine Learning	Predictive Analytics & Safety	Short-term flow forecasting, incident detection, automated computer vision	Shifts traffic operations from reactive clearing to proactive bottleneck prevention.
Mathematical Optimization	Public Transit & Multimodal Scheduling	Network graph algorithms, headway optimization, fleet dispatching	Maximizes bus/rail efficiency and synchronizes intermodal transfers.

One of the most compelling insights from the alignment scores is the inseparable relationship between road congestion and public transit optimization. The analysis confirms that roadway automation alone (such as adaptive traffic light signaling) yields diminishing returns unless coupled directly with high-frequency, reliable public transport and active mobility alternatives.

Strategic Lessons for Urban Planners & Policymakers

From an urban governance and planning perspective, the study outlines three pivotal recommendations for cities embarking on smart mobility strategies:

Invest in Municipal Data Infrastructure First: Advanced AI and digital twins are only as viable as the underlying data pipeline. Cities must prioritize open, standardized, and interoperable spatial data backbones rather than purchasing proprietary, closed vendor solutions.
Embed Digital Twins into Statutory Planning: Digital twin simulations should not remain isolated university experiments; they must be integrated into official long-term transport masterplans, allowing planners to simulate emissions and equity impacts before committing public funds.
Cross-Border Technology Transfer: International comparative research between cities with different urban morphologies (such as Liverpool in the UK and Busan in South Korea) provides vital stress-testing for mobility models under diverse governance and density conditions.

Access the Open-Access Paper

The full article is published open access in Sustainability under the Creative Commons Attribution (CC BY) licence:

Read Article on MDPI (Open Access) University News Release

Citation (BibTeX)

@article{son2025leveraging,
  author  = {Son, Heejoo and Jang, Jinhyeok and Park, Jihan and Balog, {\'{A}}kos and Ballantyne, Patrick and Kwon, Heeseo Rain and Singleton, Alex and Hwang, Jinuk},
  title   = {Leveraging Advanced Technologies for (Smart) Transportation Planning: A Systematic Review},
  journal = {Sustainability},
  volume  = {17},
  number  = {5},
  pages   = {2245},
  year    = {2025},
  doi     = {10.3390/su17052245},
  url     = {https://doi.org/10.3390/su17052245}
}

Paper Overview & Key Details