Machine Learning Maintenance Cost Forecast for Better Infrastructure LCCA

Traditional life cycle cost analysis (LCCA) involves considering multiple design/project alternatives to deliver the same outcome, and then comparing the net present value (NPV) of each alternative using estimated costs. An issue with LCCA/NPV analysis is that one must assume that once a project alternative has been chosen, the cash flows are set, and the alternative is irreversible. Using financial technology (FinTech) tools, specifically data analysis, along with remote-sensing data for structural health monitoring (SHM), it is possible to develop an intelligent forecasting tool that can better estimate infrastructure maintenance costs in real-time. This project aims to develop and test a Machine Learning (ML)/FinTech-based forecasting tool that can be used to provide an intelligent forecast of maintenance costs, allowing for a more accurate and actionable LCCA based on real-time data.

Language

• English

Project

• Status: Active
• Contract Numbers:

  69A3552348339

• Sponsor Organizations:

  Center for Durable and Resilient Transportation Infrastructure

  University of Texas, Arlington
  Arlington, TX  United States  76019

  Office of the Assistant Secretary for Research and Technology

  University Transportation Centers Program
  Department of Transportation
  Washington, DC  United States  20590
• Performing Organizations:

  Missouri University of Science & Technology, Rolla

  Department of Engineering
  202 University Center
  Rolla, MO    65409
• Start Date: 20230901
• Expected Completion Date: 0
• Actual Completion Date: 0

Subject/Index Terms

• TRT Terms: Costs; Data analysis; Forecasting; Life cycle analysis; Machine learning; Maintenance management
• Subject Areas: Finance; Maintenance and Preservation; Planning and Forecasting; Transportation (General);

Filing Info

• Accession Number: 01909698
• Record Type: Research project
• Source Agency: Center for Durable and Resilient Transportation Infrastructure
• Contract Numbers: 69A3552348339
• Files: UTC, RIP
• Created Date: Feb 23 2024 4:35PM