Trend AnalysisInterdisciplinary

Digital Twins for Urban Planning and Smart Cities

Urban digital twins create virtual replicas of entire citiesโ€”integrating real-time sensor data, 3D models, and AI simulation to let planners test interventions before deploying them in the physical world. Recent work addresses the gap between proof-of-concept demonstrations and operational urban governance.

By Sean K.S. Shin
This blog summarizes research trends based on published paper abstracts. Specific numbers or findings may contain inaccuracies. For scholarly rigor, always consult the original papers cited in each post.

Why It Matters

Cities house over half of humanity and generate roughly 75% of global carbon emissions. Urban planning decisionsโ€”where to build transit, how to zone development, where to plant treesโ€”have consequences that unfold over decades and affect millions. Yet historically, these decisions have relied on static models, political negotiation, and educated guesswork.

Digital twins change this calculus fundamentally. An urban digital twin is a dynamic, data-fed virtual replica of a city that continuously mirrors its physical counterpart. Traffic sensors feed real-time congestion data. Weather stations update microclimate models. Energy meters track building consumption. Within this digital mirror, planners can simulate interventionsโ€”"What happens to downtown temperatures if we add 10,000 trees?" or "How does a new metro line change commuting patterns?"โ€”without touching the physical city.

The technology has matured from isolated prototypes to operational systems. Singapore's Virtual Singapore, Helsinki's Kalasatama Digital Twin, and Shanghai's CityBrain represent different approaches to the same ambition: making urban governance evidence-based, predictive, and participatory. The 2024-2025 literature focuses on a critical transitionโ€”from technically impressive demonstrations to practically useful governance tools.

The Science

Systematic Landscape of Urban Digital Twins

Sacoto-Cabrera et al. (2025), with 8 citations, conduct a systematic review mapping the intersection of IoT, AI, and digital twins in smart cities. Their thematic analysis identifies five dominant application areas: energy management, traffic optimization, environmental monitoring, emergency response, and citizen engagement.

A key finding is the integration gap: most urban digital twin implementations excel in one domain (e.g., traffic) but struggle to connect across domains. A traffic digital twin and an energy digital twin for the same city may use different data formats, time scales, and spatial granularitiesโ€”making integrated analysis (e.g., "how does traffic rerouting affect building energy consumption?") technically difficult. The review identifies interoperability standards as the critical bottleneck for next-generation urban digital twins.

AI-Driven Predictive Planning

Kalfas et al. (2025), with 3 citations, evaluate AI-driven digital twin technology for predictive urban planning in Greek cities. Their case study measures four dimensions: simulation accuracy, real-time data processing capability, stakeholder engagement, and planning decision quality.

Results show that AI-enhanced digital twins significantly improve simulation accuracy compared to traditional planning models, particularly for complex multi-variable scenarios like mixed-use neighborhood development. However, the study identifies a persistent challenge: the accuracy of predictions depends heavily on the quality and completeness of input data. Greek cities, like many European municipalities, have incomplete digital infrastructureโ€”sensor coverage is uneven, historical data is fragmented, and building information models are outdated or absent.

The practical implication: digital twin deployment requires parallel investment in data infrastructure, not just software sophistication.

Open-Source Thermal Comfort Modeling

Lopez-Cabeza et al. (2025) demonstrate an open-source urban digital twin specifically designed to enhance outdoor thermal comfort in Huelva, Spainโ€”a city facing increasing heat stress from climate change. Their approach is deliberately constrained: rather than attempting a comprehensive city-wide digital twin, they focus on a single, well-defined problem with clear actionable outputs.

The system models microclimatic conditionsโ€”temperature, humidity, wind, solar radiationโ€”at street level, then simulates the thermal comfort impact of interventions: adding shade structures, changing surface materials, increasing vegetation, or modifying building heights. The open-source approach is strategic: it enables other mid-sized cities with limited budgets to adapt the framework to their own contexts.

This focused approach represents an important counterpoint to the comprehensive "whole city" digital twin ambitionโ€”sometimes a targeted tool that solves one problem well is more useful than a platform that attempts everything.

Open-Source AI Decision Support

Shulajkovska et al. (2024), with 15 citations, develop an open-source AI framework for urban decision support, emerging from the EU Urbanite H2020 project. Their system enables urban policymakers to conduct cost-benefit analyses of traffic management changes using AI models that learn from historical city data.

The framework's architecture separates domain logic (traffic models, emission models, economic models) from the AI reasoning layerโ€”allowing the same platform to be adapted across different cities with different data sources. This modular design addresses a key barrier to digital twin adoption: most cities cannot afford custom-built systems, and vendor lock-in to proprietary platforms creates long-term cost and governance risks.

Urban Digital Twin Maturity Levels

<
LevelCapabilityCurrent Adoption
1 - Static3D city model, no real-time dataWidespread (Google Earth, etc.)
2 - ConnectedReal-time sensor data integrationMajor cities (Singapore, Helsinki)
3 - AnalyticalAI-driven simulation and predictionPilot projects, limited domains
4 - IntegratedCross-domain analysis (traffic + energy + climate)Emerging, interoperability challenges
5 - AutonomousSelf-optimizing city systems with human oversightTheoretical, not yet operational

What To Watch

The democratization of urban digital twins through open-source frameworks is the trend to watch most closely. If digital twins remain expensive proprietary systems accessible only to wealthy global cities, they will widen the gap between data-rich and data-poor urban governance. Open-source approaches like those demonstrated by Lopez-Cabeza et al. and Shulajkovska et al. offer a pathway for mid-sized and developing-world cities to participate. Expect increasing integration of citizen-generated data (smartphone sensors, social media reports) into digital twins, and growing regulatory demand for digital twin-based impact assessments before major urban development projects are approved.

Explore related work through ORAA ResearchBrain.

References (5)

[1] Sacoto-Cabrera, E.J., Perez-Torres, A., & Tello-Oquendo, L. (2025). IoT, AI, and Digital Twins in Smart Cities: A Systematic Review. Smart Cities, 8(5), 175.
[2] Kalfas, D., Kalogiannidis, S., & Spinthiropoulos, K. (2025). Enhancing Predictive Urban Planning in European Smart Cities Through AI-Driven Digital Twin Technology. Urban Science, 9(7), 267.
[3] Lopez-Cabeza, V., Videras-Rodriguez, M., & Gomez-Melgar, S. (2025). An Open-Source Urban Digital Twin for Enhancing Outdoor Thermal Comfort in Huelva. Smart Cities, 8(5), 160.
[4] Shulajkovska, M., Smerkol, M., & Noveski, G. (2024). Enhancing Urban Sustainability: Open-Source AI Framework for Smart Cities. Smart Cities, 7(5), 104.
Shulajkovska, M., Smerkol, M., Noveski, G., & Gams, M. (2024). Enhancing Urban Sustainability: Developing an Open-Source AI Framework for Smart Cities. Smart Cities, 7(5), 2670-2701.

Explore this topic deeper

Search 290M+ papers, detect research gaps, and find what hasn't been studied yet.

Click to remove unwanted keywords

Search 7 keywords โ†’