Artificial Intelligence Opportunities for State and Local DOTs: A Research Roadmap (2024)
Chapter: Summary
Summary
In recent years, significant advancements in artificial intelligence (AI) have profoundly impacted the field of transportation research. Most modern approaches use data-driven methods to find efficient solutions and draw key inferences. These methods promise real-time inference while achieving human-level accuracies. One of the most notable breakthroughs has been in the realm of autonomous vehicles. These vehicles can perceive their surroundings, make real-time decisions, and navigate complex urban environments. Moreover, AI has revolutionized traffic management and optimization. Through predictive analytics and real-time data processing, AI systems show promise in alleviating congestion, reducing travel times, and enhancing overall safety by alerting drivers to potential hazards. Additionally, AI-driven simulations are used for testing and improving transportation systems, saving time and resources that would otherwise be needed for physical tests. These recent advancements not only promise more efficient and safer transportation but also pave the way for innovative solutions in public transit, logistics, and urban planning, ultimately reshaping the way we move and interact with our environments.
However, to unleash the power of AI in transportation and use it for public good, we need every sector in transportation to strategically adopt methods and tools from AI. In this project, we specifically discussed possible steps for state and local departments of transportation (DOTs) to adopt AI in their pipeline. We explored a large corpus of literature to study the complexity and maturity at the intersection of AI and transportation. We also held discussions with DOT personnel through multiple workshops and interview sessions to understand their needs in infrastructure, workforce, and the desired solution space. As a result of this process, we have identified strategies that may help DOTs and facilitate the progressive inclusion of AI solutions.
This project was executed through six tasks. This includes a detailed literature review from the past 11 years of research at the intersection of AI and transportation. We also reviewed 100 research projects from the TRB database and summarized their trends and outcomes. The project team then performed outreach activities through interviews and workshops to understand the current state of practice, challenges, vulnerabilities, and research priorities at DOTs while focusing on AI-based applications and incorporation of AI in practice. Findings from the literature review and the outreach activities, along with the U.S. government’s current strategic plan for AI deployment, helped us to develop 11 research need statements that will help DOTs to assimilate the benefit of AI in their regular workforce. These statements were discussed with DOT personnel and NCHRP panel members for further refinements. Finally, we proposed a dissemination plan for the research need statements where we discussed possible partners, a tentative timeline, and potential challenges.
