Bridges

Cover of Improved Bridge Deterioration Models, Predictive Tools and Costs
Improved Bridge Deterioration Models, Predictive Tools and Costs
  • Publication no: AP-R487-15
  • ISBN: 978-1-925294-46-0
  • Published: 26 June 2015

The management of Australia and New Zealand’s bridge stock involves the development of forward works programs for periods up to, and longer than, ten years into the future. The scheduling of maintenance, rehabilitation and replacement, forms part of these forward works programs; however, it is often difficult to estimate when these costs will be incurred and plan accordingly.

Bridge deterioration modelling and prediction tools have the potential to improve the scheduling of maintenance and replacement works for bridges. In this report, state-of-the-art deterioration modelling techniques are reviewed highlighting the advantages and limitations of each approach. The findings indicate that currently deterioration modelling is potentially difficult to implement in the Australian and New Zealand context, given the lack of integration of the data sets with existing bridge asset management tools.

Based on the results of a comprehensive survey of Australian and New Zealand road agencies, the status quo regarding bridge asset management practices is discussed and evaluated. This includes the number of condition states, inspection frequency, structural inventory classification, and the collection of environmental data and historical cost data. In addition, issues such as intervention levels, utilisation of condition data, quality assurance and quality control of inspection data and engineering judgement were also reviewed. The report identifies differences in approaches between member road agencies and makes recommendations regarding minimum data sets required for deterioration modelling.

A vision statement which covers the specific issues that need to be progressed has been developed to assist road agencies to work together towards harmonized common goals for inspection practise, inventory information and bridge management approach. The achievement of this vision will enable the development and implementation of a national deterioration model for bridge asset management.

  • Summary
  • 1. Introduction
    • 1.1. Background
    • 1.2. Aims
    • 1.3. Project Scope
    • 1.4. Outline
  • 2. Literature Review of Deterioration Models
    • 2.1. Deterioration Modelling and Predictive Tools
      • 2.1.1. What are Deterioration Models and Tools?
      • 2.1.2. Condition State, Exposure Classification and Inspection Frequency
      • 2.1.3. Maintenance, Events and Disasters, and Influence on Historic Data
      • 2.1.4. Validation and Calibration
      • 2.1.5. Failure Criteria Selection
    • 2.2. Types of Deterioration Models
    • 2.3. Deterministic Models
      • 2.3.1. Average Time to Failure
      • 2.3.2. Linear Deterioration Models
      • 2.3.3. Regression Models
    • 2.4. Stochastic Models
      • 2.4.1. Markov Models
      • 2.4.2. Semi-Markov Models
      • 2.4.3. Weibull Survival Model
      • 2.4.4. Hybrid Markov-Weibull Models
      • 2.4.5. Stochastic Gamma Process Deterioration Model
    • 2.5. Artificial Intelligence Models
      • 2.5.1. Artificial Neural Network (ANN) Backward Prediction Model (BPM)
      • 2.5.2. Case-based Reasoning Models
    • 2.6. Other Deterioration Models
      • 2.6.1. The Reliability-based Mechanistic Models
      • 2.6.2. Ordered Probit Method
  • 3. Implementation of Deterioration Modelling in Bridge Asset Management
    • 3.1. Current Use of Deterioration Modelling in Australia and New Zealand
      • 3.1.1. Review of Jurisdictional Practice
      • 3.1.2. Implementation in the Australian and New Zealand Context
    • 3.2. Potential Use of Deterioration Modelling in Bridge Asset Management
  • 4. Deterioration Modelling and Asset Management
    • 4.1. Bridge Asset Management
    • 4.2. Bridge Management Using Performance Models
    • 4.3. Network Strategy Development
      • 4.3.1. Methodology
      • 4.3.2. Method Variables and Outputs
      • 4.3.3. Model Implementation
    • 4.4. Time to Intervention Prioritisation
      • 4.4.1. Methodology
      • 4.4.2. Method Variables and Outputs
      • 4.4.3. Model Implementation
    • 4.5. Advanced Models and Requirements
  • 5. Current Data Collection Practices
    • 5.1. Australian and New Zealand Practices
      • 5.1.1. Data Sources for Bridge Asset Management (BAM) Systems Input
      • 5.1.2. Bridge Asset Management (BAM) Systems and Data History
      • 5.1.3. Inspection Frequency
      • 5.1.4. Structural Inventory Classification
      • 5.1.5. Condition State Classifications
      • 5.1.6. Identification/definition of Critical Structural Components
      • 5.1.7. Intervention Levels and Required Actions
      • 5.1.8. Environmental Data
      • 5.1.9. Cost Data
      • 5.1.10. Utilisation of Condition Data
      • 5.1.11. Risk Prioritisation (WhichBridge)
      • 5.1.12. Quality Assurance and Quality Control
      • 5.1.13. Engineering Judgement
    • 5.2. International Practice
      • 5.2.1. Inspection Frequencies
      • 5.2.2. Structural Inventory Classification
      • 5.2.3. Condition State Classification
      • 5.2.4. Condition Assessment Criteria
      • 5.2.5. Intervention Levels and Required Actions
      • 5.2.6. Environmental Data
      • 5.2.7. Cost Data
      • 5.2.8. Quality Assurance (QA) and Quality Control (QC)
    • 5.3. Discussion
      • 5.3.1. Condition State Classification
      • 5.3.2. Inspection Frequencies
      • 5.3.3. Structural Inventory Classification
      • 5.3.4. Intervention Levels
      • 5.3.5. Environmental Data
      • 5.3.6. Cost Data
      • 5.3.7. Quality Assurance and Quality Control
      • 5.3.8. Need for Deterioration Modelling
      • 5.3.9. Willingness to Change
      • 5.3.10. Engineering Judgement
  • 6. Data Specifications for use in Deterioration Modelling
    • 6.1. Potential Outcomes of Deterioration Modelling
    • 6.2. Effective Data Sets
      • 6.2.1. Data Requirements
      • 6.2.2. Structural Inventory Classification
    • 6.3. Optimum Condition Assessment Criteria
      • 6.3.1. Number of Condition States
      • 6.3.2. Condition State Criteria
    • 6.4. Classification of Related Expenditure
      • 6.4.1. Maintenance and Refurbishment
      • 6.4.2. Structures Valuation
    • 6.5. Capturing of Environmental Data
    • 6.6. Risk-based Inspection Frequencies
    • 6.7. Risk Score Prioritisation
    • 6.8. Quality Assurance (QA) and Quality Control (QC)
    • 6.9. Intervention Levels and Actions
    • 6.10. Mapping Existing Data to a New Scale
  • 7. Engineering Judgement-based Model Framework
    • 7.1. Engineering Judgement-based Model Framework
      • 7.1.1. Engineering Judgement and Risk
      • 7.1.2. Selection of Experts
      • 7.1.3. Engineering Judgement in Asset Management
      • 7.1.4. Model Framework
      • 7.1.5. Manual Engineering Process for Maintenance and Replacement
    • 7.2. Case Study: NZTA Coating Deterioration Modelling
  • How was it developed?
  • How does it work?
  • What are the limitations?
  • Where does the engineering judgement fit in?
  • 8. Vision
  • 9. Conclusions and Recommendations
  • References
  • Appendix A Survey Questionnaire
  • Appendix B Summary of Survey Responses
  • B.1 Department of State Growth (DSG)
  • B.2 ACT Government – Roads and Transport
  • B.3 Department of Planning, Transport and Infrastructure (DPTI)
  • B.4 Main Roads Western Australia (MRWA)
  • B.5 NZ Transport Agency (NZTA)
  • B.6 Roads and Maritime Services (RMS)
  • B.7 Queensland Department of Transport and Main Roads (TMR)
  • B.8 Roads Corporation Victoria (VicRoads)