Digital Twins as a Catalyst for Sustainable and Smart Cities

Principal Investigator(s): Junfeng Jiao PhD, Dev Niyogi PhD, Yiming Xu PhD

Project Partners: The University of Texas at Austin School of Architecture, The University of Texas at Austin’ Jackson School of GeoScience, UT Good System, City of Austin Fire Department

Research Project Funding: $241,478

Project Start and End Date: Oct 1st, 2023 – September 30st, 2024

Project Description: This project aims to develop an urban digital twin for the city of Austin that assists city planners and managers to build a sustainable and smart city. The proposed urban digital twin will incorporate a data management and visualization platform, a real-time city monitoring system, an integration of predicting models, and a dynamic urban simulation environment to achieve effective city management, better resource allocation, more efficient transportation operation, and more proactive responses to risks. The data management and visualization platform will store and publish static and real-time urban data. The platform will enable API access and data download to facilitate third-party use. The real-time city monitoring system will access and process multiple data sources, including public real-time dataset, camera, and road sensors, for traffic monitoring and accident detecting. The digital twin will incorporate a traffic predicting model and a risk predicting model using graph-based deep learning methods. We will also build a dynamic urban simulation environment for the city of Austin, including a 3D city model, a road network model, and a traffic simulator. These digital twin modules will operate cooperatively by interacting with each other to synchronize real-world and virtual information. The expected outputs of this project include online platforms, software, technical reports, and research papers.

US DOT Priorities: This project fits well within the US DOT strategic goals of Transformation and Climate and Sustainability, aiming to develop a comprehensive, dynamic, and interactive urban digital twin that will enable multi-source data management, real-time risk detecting, and reliable simulation, helping city planners and managers build and maintain a sustainable and smart city. This project also contributes to CCST’s Focus Area 5: Smart Cities & Innovative Adaptation and Mitigation Technologies.

Outputs: The expected outputs of this project include:

  1. A data management and visualization platform (website/software)
  2. A real-time city monitoring system (website/software)
  3. A paper on real-time transportation object detecting and tracking
  4. A paper on real-time traffic condition prediction
  5. A paper on traffic risk prediction
  6. A dynamic simulation environment for the city of Austin
  7. A paper on the proposed digital twin system.

 

Outcomes/Impacts:

The expected outcome of this project is a fully functional digital twin for city of Austin, incorporating real-time transportation information, active fires and fire risks, air condition, and noise information. This digital twin for city of Austin is able to archive historical data, display real-time (or near-real-time) information, generate predictions for the future (e.g., traffic conditions and risks, smoke paths and air quality drops due to fires), and perform simulations for events, emergencies, and new traffic patterns. The information provided by the digital twin can be visualized in a 2D map (e.g., traffic condition) or a 3D city model (fire smoke). The digital twin can also serve as a city monitoring system for accidents and emergencies to trigger early responses. These functions help the city in transportation management, risk response, and policy evaluation, contributing to a sustainable and smart city. For residents, the information provided by the digital twin can guide route planning, travel mode selection, and evacuation decision.

The proposed digital twin project will have long-lasting benefits for city of Austin on effective city management, better resource allocation, more efficient transportation operation, and more proactive responses to risks, all of which are critical for building and maintaining a smart and sustainable city. The digital twin will also generate predictions for traffic condition and fire smoke spread, which enables the city to make corresponding transportation management strategy and evacuation instructions.

The project is particularly for the city of Austin, but it also serves as a robust case study and data source for researchers while setting new practical standards for integrated, data-driven governance that other municipalities can adapt. The data parsing algorithm, the visualization program, and the predicting model are generalizable. The project will also offer unique datasets for academic research such as transit patterns, traffic monitoring, and transportation management. It sets a practical precedent for other cities to enhance efficiency of the transportation system and reduce GHG emissions. The project will provide comprehensive real-world dataset, both static and dynamic, to support equitable distribution of utilities and services. These models would inspire more AI based risk detecting and predicting method for better risk management in the future.