Artificial Intelligence (AI), mainly machine-learning algorithms, has been around for decades. What has changed since 2022 that made everyone suddenly want a piece of it? The breakthrough came with ChatGPT, which made AI accessible to a much broader audience. Even those without a technical background can now engage with AI in a meaningful way. While AI technologies are advancing rapidly, achieving these goals will require thoughtful planning. In this blog, I’ll go into two of many technology-related topics in transportation, AI and Digital Twins, and discuss:
What they are and what they promise.
Transportation agencies leveraging this technology.
Considerations to make the most of new technologies.
Projects you can take on today.
The State of AI and Digital Twins in Transportation
AI in Transportation Operations
Many Intelligent Transportation Systems (ITS) organizations and technology vendors are promoting AI as a way to improve almost everything. In transportation, for instance, AI promises to enhance efficiency (e.g., through traffic management), boost safety (e.g., with predictive maintenance and autonomous vehicles), and address equity (e.g., by optimizing services for underserved areas). However, regulation and legislation around AI in transportation is a work-in-progress in the US. Here’s what we know so far:
The Highly Automated Systems Safety Center of Excellence (HASS COE) published two reports on AI Assurance for Transportation and Understanding AI Risks in 2024.
The current AI dilemma is mostly focused on automated vehicles (AVs) and bias in decision-making AI algorithms involved.
Regulations governing AI in traffic management systems, such as the use of the 5.850-5.925 Band, is regulated by the Federal Communications Commission (FCC).
The regulations that safeguard personally identifiable information (PII) uphold Indigenous data sovereignty, secure confidential business information and protect intellectual property rights.
In the 2023 legislative session, at least 25 states, Puerto Rico and the District of Columbia introduced AI bills, and 18 states and Puerto Rico adopted resolutions or enacted legislation.
How Do You Determine What is Achievable?
The World Bank and the Energy Sector Management Assistance Program established a framework for achieving energy savings through ITS investments in smart cities. This framework outlines the necessary institutional, technical, and physical conditions for successful smart mobility investments. These steps help assess the risks and success of technology solutions and gauge public reaction:
Identify mobility problems and design solutions.
Deploy and operate these solutions.
Encourage user adoption and behavioral change.
Scale up and evolve the solutions over time.
Several agencies are already creating AI strategies or implementing solutions that can analyze and optimize service routes, improve safety, address transportation deserts, expand access for underserved communities or improve organizational processes.
Texas Department of Transportation is exploring using AI for monitoring and managing traffic, detecting incidents in real-time and streamlining project delivery. In December 2024, TxDOT released its 2025-2027 Artificial Intelligence Strategic Plan.
Washington Metropolitan Area Transit Authority launched the MetroAccess AI Pilot in 2024, introducing an intelligent digital assistant for booking one-way or round transit trips and providing support for incoming calls. This pilot aims to reduce extended wait times and enhance the overall customer experience.
Chicago Transit Authority partnered with Google Public Sector to create an AI-powered communication tool that makes it easier than ever for customers to provide feedback. The automated chatbot supports a variety of topics covering cleanliness, maintenance, ADA accommodations, safety & security, service disruptions and finding the next train or bus.
Los Angeles County Metropolitan Transportation Authority is the largest agency to install forward-facing cameras on some of their buses for transit-only lanes and bus stop enforcement.
New York City Transit implemented AI to improve scheduling and maintenance operations, helping to prevent bus breakdowns and enhance service reliability.
Bay Area Rapid Transit introduced data-driven passenger load charts that analyze ridership patterns, offering riders a clear snapshot of what to expect in terms of on-board train capacity during the COVID-19 pandemic.
Transport for London uses AI to improve traffic management and optimize bus routes. AI-driven analytics help Transport for London target areas with high demand but limited service, promoting transportation equity.
Hong Kong Mass Transit Railway leverages AI for real-time monitoring and service optimization, ensuring that less accessible regions receive adequate transit services.
The City of Bellevue is partnering with us through a Strengthening Mobility and Revolutionizing Transportation (SMART) Grant Real-time Traffic Signal Safety Intervention (RTSSI) Stage 1 Project. The project’s primary purpose is to conduct demonstrations of innovative technologies, including intelligent sensor-based infrastructure and smart technology traffic signals that are designed to improve the safety of vulnerable road users at Bellevue’s signalized intersections.
Digital Twins
Another technology gaining traction in the transportation industry is digital twins. However, similar to AI, regulations and legislation around digital twin technology are sparse. In any industry, a digital twin is a virtual representation of a system, updated from real-time data, and it uses simulation, machine learning and reasoning to help decision-making. In urban planning and transportation engineering, digital twins are emerging as a digital asset with the promise of connectivity, investment management and integration. Developing a comprehensive digital twin with these elements could help agencies and roadway owners modernize transportation systems and enhance efficient and resilient systems.
The City of Helena in Montana uses an Americans with Disabilities Act (ADA) planning software, to establish a detailed Light Detection and Ranging (LiDAR) scan of 300 miles of streets and leveraged artificial intelligence to automatically extract over 40,000 assets such as sidewalks, bike facilities, curb ramps, curbs, gutters, signage, road cross slopes, road striping measurements to populate the comprehensive digital twin environment. LiDAR is a remote sensing method that uses a pulsed laser to generate a 3D map of a given environment.
The Sarasota-Manatee Metropolitan Planning Organization uses a cloud software that leverages machine learning, natural language processing, and advanced data analytics, as an added digital infrastructure layer to close data gaps and to help expedite traffic safety analysis, reporting, and operations.
The City of Barcelona began implementing the internet of things (IoT) technologies in 2012 to improve quality of life for its citizens. This initiative includes sensors for managing parking, street lighting, and trash disposal services. The city has developed extensive smart infrastructure that enhances urban connectivity.
San Francisco’s Fiber to Housing program will connect 30,000 units of affordable housing with free, high-speed internet by July 2025.
The City of Chattanooga, Tennessee, partnered with the Oak Ridge National Laboratory to build a digital twin that pulls data from about 500 different sources, including traffic cameras, weather stations and emergency services. They used the digital twin to conduct experiments on reducing traffic congestion.
Several cities and regions in the U.S. are leading the way in developing and implementing digital infrastructure. Cities like New York City, Austin, Chicago, Seattle, Denver and Boston are investing in and expanding internet access, including fiber networks, 5G networks and public Wi-Fi. These cities are focusing on aspects such as connectivity, investments, management, national frameworks and integration. While they represent a range of approaches and strategies for advancing digital infrastructure, their overall maturity is still yet to be assessed.
How Do You Make the Most of These Technologies?
Building such infrastructure requires substantial investment, coordinated planning and ongoing maintenance. Success depends on alignment among various stakeholders and effective use of resources. Before you rush to implement AI and Digital Twin technology solutions, consider the following questions in your planning process.
Are You Addressing the Problem at its Root?
Distractions from the root causes of problems can result in solutions being proposed solely for the sake of technology. It’s crucial to focus on understanding the underlying issues before implementing technological fixes. One common promise of technology is the improvement of safety while reducing congestion, without addressing factors like car dependency, lack of contextual speed limit setting and physical limitations that contribute to both congestion and crashes. Simply adding another lane will not resolve safety nor congestion problems, whether through AVs or IoT solutions.
Is Your Workforce Prepared?
The most advanced technologies, AI languages and algorithms will be ineffective without proper workforce development. Transportation agencies, public works and transit authorities can only fully benefit from these advancements if there is an AI-trained workforce to utilize them. Data and technology literacy should be top priorities for workforce development across all public agencies, aiming to attract the next generation while also upgrading the skills of the existing workforce.
Innovative training methods, such as virtual reality and augmented reality, can significantly enhance skill development. While achieving these goals is challenging, it’s possible through targeted investments in education and training and collaboration between industry and educational institutions.
What Are the Risks and How Do We Mitigate Them?
There are significant risks associated with AI and any other intelligent technology, including user perceptions, increased distraction and partially automated driving limitations when navigating construction zones, inclement weather or responding to erratic behavior from drivers. Additionally, cloud-based and data analysis solutions rely heavily on accurate information dissemination and sensors. Malfunctions or inaccuracies in these components can lead to incorrect system responses or failures to detect obstacles. Technical failures are inevitable, so it's essential to prepare for redundancies in systems to ensure safety and accessibility.
One particularly concerning issue is the haste to resolve regulatory challenges, cybersecurity risks, and ethical considerations. Road authorities, responsible for the planning, construction, maintenance and management of roadways, often become overly enthusiastic about the promises made by vendors and technology companies. As a professional community in transportation, we have a responsibility to prioritize continuous improvement, robust testing and validation and strong cybersecurity measures, including data sharing with the private sector.
Achievable AI and Digital Twin Implementation Projects
Integrate Real-time Traffic Data for Routing Optimizations
Public agencies can integrate real-time traffic data into their traffic management systems. This can help dynamically adjust traffic signals, manage congestion and improve the flow of traffic throughout the city.
Optimized Routing for Emergency Vehicles: AI-driven route optimization can help emergency services (like ambulances and fire trucks) avoid traffic congestion and road closures, leading to faster response times.
Incident Management: Real-time traffic data can be used to quickly identify and respond to accidents or road incidents, ensuring that resources are deployed effectively and disruptions are minimized.
Educate and Engage
Public agencies can use traffic data and AI insights to keep people informed about traffic conditions, road closures and alternative routes, improving public trust and engagement.
Improve Transit Customer Experience
We face significant challenges with transit system efficiency and reliability. Alas, no AI model can automatically convert a parking lane into a bus lane or replace a bus driver. However, we can enhance efficiency and reliability through better driving assistance systems, transit signal priority and dedicated lanes—all achievable with investments in technology and AI.
Collaborate with Private Sector Technology Companies
Collaborating with technology companies specializing in traffic management and AI can accelerate advanced solutions and innovations to the public sector.
The future of transportation is rapidly evolving, and to truly harness the potential of AI and Digital Twin technologies, we must develop a deeper understanding of the underlying problem, advocate for regulatory clarity, prepare our workforce for the future, and work closely with stakeholders.
Cambridge Systematics (CS) is a recognized leader in the development of strategies and applications of ITS to improve operations performance—including highways, transit, shared mobility, active transportation, and freight—at the Federal, State, regional and local level. We have helped plan statewide, regional and local ITS deployments, such as freeway, arterial, and transit management systems; connected/autonomous vehicles; freight technology; road weather systems and incident management programs. Our services provide a structured framework in which a planning agency or regional organization could begin to ensure that the decisions they make today take into account the impact of emerging technologies operating in their region.
Let’s Discuss How to Plan for New Technologies in Transportation
