- 408-924-7560
- mineta-institute@sjsu.edu
- Donate
SMART-TWIN Phase II: Implementation Roadmap - Intelligent Digital Twin Ecosystem for Adaptive Transportation Systems
The SMART-TWIN Phase II project advances the development of an AI-driven Digital Twin (DT) platform for real-time transportation management in Fresno and the Central Valley. Digital Twins are virtual models that continuously mirror real-world traffic systems using live sensor data, infrastructure information, and predictive analytics. These models enable transportation agencies to monitor conditions, forecast congestion, simulate incidents, and evaluate signal or routing strategies before implementing them in the field.
As traffic demand, safety concerns, and climate-related disruptions increase, regional agencies require smarter decision-support tools. SMART-TWIN focuses on integrating multiple pilot corridors into a scalable digital ecosystem that provides real-time insights for congestion management, safety monitoring, and sustainable mobility planning. Rather than full automation, the project emphasizes practical decision support to help agencies respond more quickly and effectively.
The project includes system integration, AI-based prediction models, and real-world pilot testing. Outcomes will demonstrate improved traffic coordination, enhanced safety, and more efficient infrastructure planning. In addition to research impacts, SMART-TWIN strengthens workforce development by engaging students in hands-on digital infrastructure and applied AI training, positioning Fresno State and its partners as leaders in intelligent transportation innovation.
University:
Mineta Consortium for Transportation Mobility
Principal Investigator:
Hovannes Kulhandjian
PI Contact Information:
California State University Fresno
Dates:
January 2026 to March 2027
Implementation of Research Outcomes:
The SMART-TWIN Phase II project will produce a scalable, AI-enabled Digital Twin prototype for real-time transportation management in Fresno and the Central Valley. Key outputs include a modular cloud-edge digital twin architecture with secure data integration pipelines; predictive AI models (e.g., LSTM and Graph Neural Networks) for congestion and incident forecasting; and simulation-based evaluation tools using platforms such as SUMO and AnyLogic for “what-if” scenario analysis. The project will also deliver an interactive 3D visualization dashboard for live monitoring and decision support, along with performance benchmarking reports and replication guidelines for regional deployment. Additional outcomes may include peer-reviewed publications, open datasets (where permitted), invention disclosures, and potential patent filings related to AI-driven traffic prediction and digital twin system integration. Together, these outputs establish a practical, deployable framework for data-driven, AI-assisted transportation operations.
Impacts/Benefits of Implementation:
Implementation of SMART-TWIN Phase II will enhance transportation system safety, efficiency, and reliability by providing agencies with real-time visibility and predictive insights into traffic operations. The AI-enabled Digital Twin will allow planners and operators to anticipate congestion, detect incidents earlier, and evaluate response strategies before deploying them in the field. This proactive decision-support capability can reduce crash risk, minimize travel-time variability, lower vehicle idling and emissions, and improve overall corridor performance.
By integrating simulation and forecasting tools into daily operations, the project supports data-driven signal timing adjustments, coordinated corridor management, and more resilient responses to disruptions such as crashes, construction, or severe weather. These improvements can lead to measurable reductions in congestion-related costs, fuel consumption, and operational inefficiencies.
The research will also inform regional transportation policy and planning by providing validated performance metrics and scenario analyses that guide infrastructure investments and smart mobility strategies. Deliverables such as replication guidelines, technical documentation, and potential patentable AI-based forecasting or system-integration methods may influence best practices for digital infrastructure deployment across mid-sized urban regions. Collectively, SMART-TWIN strengthens the foundation for safer, more sustainable, and cost-effective transportation management in California’s Central Valley and beyond.
Project Number:
2617
-
Contact Us
San José State University One Washington Square, San Jose, CA 95192 Phone: 408-924-7560 Email: mineta-institute@sjsu.edu
