Condition Monitoring Optimization: Going Beyond Traditional CML Optimization

Introduction

The goal of inspection is to accurately assess the condition of an asset and to reduce the current and future uncertainty of the damage state that can occur from corrosion. An accurate condition assessment provides the information required to calculate an asset’s probability of failure and define future inspection requirements. Current inspection programs rely heavily on inspection data from condition monitoring locations (CMLs); however, many facilities struggle to properly place the correct number of CMLs and struggle to optimize CML inspection activities based on their assets’ damage rate and risk.

Traditional CML optimization programs are often used to determine the minimum number of CMLs needed to accurately monitor active damage mechanisms and identify susceptibilities prior to an asset failure. These optimization programs are typically concerned with eliminating CMLs by intelligently defining inspection scope, techniques, and intervals that prevent unexpected failures. The desired result is the identification of inspection program deficiencies and specific changes required to effectively assess risk with confidence.

This article describes condition monitoring optimization as a data-driven methodology by which inspection scope, techniques, and intervals are intelligently determined and dynamically updated to maximize reliability and return on investment (ROI) as new information becomes available. Using this methodology, data collected through inspection is used to improve confidence in the asset damage state and to determine situations in which additional data is required, inspection adds little or no value, or corrective maintenance is the appropriate action.

Condition Monitoring Optimization

Condition monitoring optimization goes beyond traditional CML optimization, which is often limited in the breadth of analysis, or can overemphasize a subset of the overall objective. For example, facilities may focus solely on the elimination of CMLs, simply select effective inspection techniques, or just quantify the damage state of the asset within a specified level of uncertainty. Not only is CML optimization inconsistently defined and applied throughout the industry, but most efforts also do not consider the ramifications of major changes in the context of neither entire mechanical integrity (MI) and risk-based inspection (RBI) programs nor their impact on overall facility reliability performance.

This article is the first of a two-part series on condition monitoring optimization. In this article, we will:

1. Review probabilistic models for determining the end of useful life for individuals CMLs
2. Provide a description of potential methodologies for performing condition monitoring optimization
3. Demonstrate the validity of the approach using a real-world case study

In our second article, we will explore additional complex scenarios including limited inspection history, inadequate CML placement, and poorly chosen inspection techniques.

Probabilistic Models for CML Remaining Useful Life