Digital Twin and Automation of Pipeline Data for Predictive Maintenance and Risk Analysis incorporating Bayesian network

Pipeline infrastructure is critical for global energy transportation, yet aging systems face increasing degradation risks from corrosion, fatigue, and environmental factors. Recent catastrophic failures have demonstrated severe safety, environmental, and economic consequences of inadequate integrity management, with annual losses reaching billions of dollars. Despite significant advances in inspection technologies, including intelligent inspection tools and sensor networks, three fundamental challenges remain unresolved: lack of interpretability in machine learning approaches, difficulty quantifying uncertainties in defect growth predictions, and the challenge of optimizing maintenance decisions with incomplete information. This research develops a Valuation Bayesian Network (VBN) integrated with Digital Twin technology as an automated decision-support tool for pipeline integrity management. The VBN approach provides a principled foundation for uncertainty quantification, data fusion, state estimation, prediction, and maintenance planning while maintaining interpretability. The framework incorporates probabilistic degradation and failure models, logic and relational models, and surrogate models to address key performance indicators, failure risks, and remaining life estimation. The proposed Digital Twin architecture comprises four interconnected layers. The physical layer encompasses pipeline infrastructure with distributed sensors and expert knowledge. The simulation layer employs VBN-based probabilistic models capturing time-dependent degradation processes, where state variables, including defect depth and material properties, are modeled through Markov transition models with parameters learned from historical inspection records, pipeline failure databases, and expert elicitation. The data fusion layer performs Bayesian updating by integrating inspection data, operational history, and expert knowledge, enabling adaptive parameter calibration to capture site-specific degradation patterns. The decision layer implements maintenance optimization using value of information analysis and supports counterfactual reasoning for intervention scenarios. This framework achieves full uncertainty propagation from measurement noise through remaining life predictions, ultimately enhancing operational safety by predicting high-risk defect development and enabling timely maintenance interventions.

Language

• English

Project

• Status: Active
• Funding: $60,000.00
• Contract Numbers:

  69A3552348323

• Sponsor Organizations:

  Office of the Assistant Secretary for Research and Technology

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

  Howard University

  2400 6th Street, NW
  Washington, DC  United States  20059
• Project Managers:

  Bruner, Britain

• Performing Organizations:

  University of Maryland, College Park

  Department of Civil and Environmental Engineering
  College Park, MD  United States  20742
• Principal Investigators:

  Attoh-Okine, Nii

• Start Date: 20260102
• Expected Completion Date: 20260930
• Actual Completion Date: 0
• USDOT Program: University Transportation Centers Program

Subject/Index Terms

• TRT Terms: Data fusion; Decision support systems; Digital twins; Maintenance management; Pipelines; Risk analysis
• Subject Areas: Data and Information Technology; Maintenance and Preservation; Pipelines; Planning and Forecasting;

Filing Info

• Accession Number: 01976553
• Record Type: Research project
• Source Agency: Research and Education for Promoting Safety (REPS) University Transportation Center
• Contract Numbers: 69A3552348323
• Files: UTC, RIP
• Created Date: Jan 19 2026 4:27PM