TAM – AI asset inventory and condition extraction to keep data up to date

Once you are done editing, you may close this tab.
Note: Please save any changes made to the fields above before adding new champions, supporting organizations, or potential panel members to this statement.
Champion(s)

Steve Wilcox

New York State DOT

[email protected]

Others Supporting Problem Statement
Please add at least one supporting organization.
Potential Panel Members
Please add at least one potential panel member.

TAM - AI asset inventory and condition extraction to keep data up to date


Funding

Please provide a funding amount.

Research Period

Please provide a research period.

Description

Please provide a brief description of the project.

Literature Search Summary

NCHRP 23-16: Implementing and Leveraging Machine Learning at State Departments of Transportation: Available here.


Objectives

 Emerging technologies hold the promise of transforming asset data collection for transportation asset management such as the use of drones for inspections, LiDAR field data collection, continuous monitoring of real-time sensor data, and more. While the technology has been transforming, MAP-21 and the Fast Act jump started at many agencies in attaining an inventory of infrastructure assets and transportation data. At the same time, accessibility and affordability to collect high volumes of asset inventory data, such as LiDAR point cloud data, present the problem of how agencies can visualize and manage such large amounts of data and integrate the many layers for each transportation asset management plan. Now that the need for such data is federally recognized, further research is needed to understand what the latest technologies for asset analysis can offer an agency as well as how frequently that information needs generated.

Research is needed in the following areas:

  • Address the adoption and practical application of these technologies and the rapid pace of technological advancement.
  • What level of extraction detail and frequency interval is needed to support TAM at both the state and local levels and how can the condition assessment be applied to the performance measures of both pavement and non-pavement assets?
  • Further investigate what tools are capable of visualizing asset extraction layers, as well as presenting such data to all stakeholders in powerful GIS formats with standardized TAM graphics for universal interpretation.
  • As the accuracy of techniques such as mobile LiDAR improves, the ability of traditional asset management data begins to converge with the accuracy needed for engineering purposes such as design and construction. If asset data at the network level becomes sufficiently accurate redundant data collection can be eliminated. The research should consider any refinements that would need to occur in network level asset management data collection to make the data useful for compliance (i.e. ADA), safety (i.e. bridge clearances) or engineering purposes.

Urgency and Potential Benefits

Please describe the urgency and potential benefits of the project.

Implementation Considerations

Please provide implementation considerations for the project.

Champion(s)

Steve Wilcox

New York State DOT

[email protected]


Others Supporting Problem Statement

Please add at least one supporting organization.

Potential Panel Members

Please add at least one potential panel member.

Person Submitting Statement

Please add information about the person submitting the statement.

Notes




Cancel