Inspection and Repair Decisions for Hydraulic Structures Under Symmetric Deterioration

Abstract

In this report, we will focus on minimizing the cost due to inspection and repair of hydraulic structures. The optimization is based on symmetric properties of the underlying physical deterioration process: i.e. the damages per time-unit are exchangeable and the probabilities of preventive repair and failure are obtained by conditioning on the average amount of deterioration with regard to a finite or an infinite time-horizon. By introducing a prior for the average deterioration per time-unit we can account for uncertainty in the decision problems. Advantages of our Bayesian approach are that we base our mathematical models on an observable quantity, namely the damage, and, surprisingly, that our results are just sums of products which can be easily evaluated. Two examples from the field of hydraulic engineering are studied: determining a preventive repair interval when a safety norm is given and a cost-optimal periodic inspection rate if there is a possibility for a preventive repair during each inspection. Optimization; Symmetric deterioration; Bayesian approach; Hydraulic structures; Exchangeability.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 1992
Accession Number
ADA253621

Entities

People

  • Jan M. Van Noortwijk

Organizations

  • University of California, Berkeley

Tags

Communities of Interest

  • Human Systems

DTIC Thesaurus Topics

  • Bayesian Networks
  • Dirichlet Integral
  • Electrical Engineering
  • Engineering
  • Inspection
  • Integrals
  • Intervals
  • Maintenance
  • Markov Chains
  • Mathematical Models
  • Military Research
  • Models
  • Preventive Maintenance
  • Probability
  • Probability Density Functions
  • Structural Engineering
  • Time Intervals

Fields of Study

  • Engineering

Readers

  • Facility/Structural Engineering.
  • Statistical inference.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms