The National Academies

NCHRP 07-34 [Pending]

Artificial Intelligence for Transportation Systems Management and Operations Applications

  Project Data
Funds: $450,000
Contract Time: 30 months
Staff Responsibility: Dr. Zuxuan Deng



As contemporary transportation systems get more complex, it becomes more challenging for decision makers to consider the large number of intertwined factors needed to optimize systemwide processes and performance. For example, when an on-roadway vehicle crash occurs, operational systems such as dynamic vehicle routing and variable speed limits may need to be activated and used to provide timely and effective improvement in traffic incident management performance. These systems are examples of transportation systems management and operations (TSMO) strategies. According to the Federal Highway Administration (FHWA), “TSMO is defined as an integrated set of strategies to optimize the performance of existing infrastructure through the implementation of multimodal and intermodal, cross-jurisdictional systems, services, and projects designed to preserve capacity and improve security, safety, and reliability of the transportation system.” Decision support systems (DSSs), which are primarily computer-based information systems used to sort, rank, or choose alternatives, have been developed to improve TSMO. However, conventional DSSs are usually built on a set of expert rules that might not be able to provide customized and optimal solutions. Meanwhile, artificial intelligence (AI) offers potential to revolutionize many facets of our daily lives, including transportation. AI has the capacity to process multiple-sourced, large-scale, real-time data to model system behaviors, predict traffic conditions and evaluate system performance, which aligns with the key functions of DSSs. Research is needed to support state departments of transportation (DOTs) in selecting and deploying the right AI technologies in DSSs for TSMO applications. 




The objective of this research is to develop a guide, including implementation roadmaps, to help state DOTs and other transportation agencies in developing and deploying next-generation, data-driven, and AI-enabled DSSs for TSMO applications.


A key emphasis should be on identifying areas where AI technologies can improve DSSs for TSMO applications and providing detailed implementation steps, resource needs, and assessing reliability and scalability of AI-based solutions. 


Accomplishment of the project objective will require completion of the following tasks, at a minimum. 





PHASE I – Planning 


Task 1a. Conduct a literature review and state-of-the-practice review of DSSs for TSMO and identify major gaps that impede agencies from achieving their strategic goals with DSSs. The review shall include current practice at state, regional, and local agencies. 


Task 1b. Identify existing AI use cases in TSMO and identify potential areas for successful AI technologies with an emphasis on how AI can address the major gaps identified in Task 1a. Identify AI technologies available to solve specific TSMO functionalities. The research team should review related published and unpublished research conducted through the NCHRP, the Federal Highway Administration (FHWA), and other national, international, state, and pooled-fund sponsored research. 


Task 1c. Develop a technical memorandum to summarize the findings from Tasks 1a and 1b. 


Task 1d. Recommend a set of AI-enabled DSSs for TSMO applications with wide implementation potential in terms of spatial contexts, agency capacities, and data needs. Conduct a webinar with the panel to choose specific TSMO applications for further investigation in subsequent tasks.   


Task 2. Focusing on the TSMO applications approved by the panel in Task 1d, propose an outline of the guide for transportation agencies to leverage AI on operational activities, resources, system needs, and decision-making, and offer insight into the development and deployment of AI-enabled DSSs for TSMO applications. The proposed guide must be generic and flexible for implementation at different spatiotemporal scales to ensure transferability.  


Task 3. Prepare Interim Report No. 1 documenting the results of Tasks 1 through 2 and provide an updated plan for the remainder of the research no later than 6 months after contract award. The updated plan must describe the process and rationale for the work proposed for Phases II and III. 




PHASE II – Execution  


Task 4. Based on approved Interim Report No. 1, develop the proposed guide, covering at least the following aspects:

  • Data requirements, e.g., what are the data needs for AI applications? What data to collect and what infrastructure (e.g., sensor) investment decisions need to be made? How to address data management and governance challenges?
  • Infrastructure needs, e.g., what sensors are suitable for different functionalities to collect data for AI-enabled TSMO applications? What communication means are required to support reliable information transmission for AI-enabled DSSs for TSMO applications?
  • Computational resources, e.g., what are the requirements for computational power and data storage/management capacity? 
  •  Algorithm transparency and interpretability, e.g., how was AI used in the process? Which AI algorithms and AI tools are used? What assumptions are used in the AI algorithms? What is the interpretability of AI algorithms?
  • Validation and testing, e.g., what validation and testing mechanisms should be implemented to ensure reliability, scalability, and transferability?
  • \Performance evaluation, e.g., how best to evaluate the effectiveness of AI and how to compute return-on-investment?
  • Operational challenges, e.g., cybersecurity threats and other challenges such as equity and ethical concerns including but not limited to data privacy and algorithm bias.
  •  Workforce requirements, e.g., what type of training is required for the agencies?
  • Roadmaps for implementation and deployment.

Task 5. Conduct case studies to demonstrate the application of the roadmaps with one or more agencies for the AI-enabled DSSs for TSMO applications that were identified in Task 1d. In the case studies, the contractor should, at a minimum, evaluate readiness and identify gaps in data availability, infrastructure, technical resources, and workforce capabilities, and develop strategies for closing these gaps. 

Task 6. Prepare Interim Report No. 2 that documents Tasks 4 through 5 and provide an updated work plan for the remainder of the research no later than 18 months after approval of Phase I. The updated plan must describe the process and rationale for the work proposed for Phase III. 


Meet with the NCHRP to review the report and obtain approval for subsequent tasks. 



PHASE III – Final Products 


Task 7. Refine and update the guide based on comments from NCHRP project panel. 


Task 8. Present the draft guide and the roadmaps to appropriate AASHTO technical committees for comments and propose any revisions to NCHRP. Revise the draft guide after consideration of review comments.


Task 9. Prepare the final deliverables including the final guide and roadmaps, and the following: 

a.    A conduct of research that documents the entire research effort and any lessons learned;  

b.    Media and communication material (e.g., presentations, 2-pager executive level flyer, graphics, graphic interchange formats (GIFs), and press releases); and 

c.  A stand-alone technical memorandum titled “Implementation of Research Findings and Products.”



STATUS: A research contractor has been selected for the project.  The contracting process is underway.


To create a link to this page, use this URL: http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=5315