What is Condition-Based Maintenance & How Can it Be Leveraged as Part of a Holistic Asset Performance Management Strategy?

According to research from MarketsandMarkets, the condition-based maintenance market is expected to grow from $10.6B in 2024 up to $47.8B by 2029. This increase is expected due to energy organizations working to meet production and energy transition requirements by better monitoring assets to help drive more efficient operations.

So, what are the key elements of condition-based maintenance? What are the potential benefits for your organization? How is condition-based maintenance evolving?

Let’s check it out.

Condition-based maintenance is a preventative maintenance strategy that focuses on the actual condition of the asset rather than a predetermined maintenance schedule. Condition-based maintenance is often times included as part of a predictive maintenance strategy, that relies on various datapoints to determine a true condition of the asset(s) being monitored.

Today, condition-based maintenance programs can draw insights from more data than ever before. This can include data inputs from sensors, operators, images, robotics, and satellite imagery based on industry. Organizations are able to use this aggregated data for subject matter experts (SMEs) to make decisions, leverage machine learning (ML) for simple analysis of asset changes, or artificial intelligence (AI) in the form of predictive analytics to develop an overall risk profile of assets being monitored with condition-based programs.

Before highlighting how condition-based maintenance differs from predictive or prescriptive maintenance, it’s important to cover how the definition of condition-based maintenance is changing across the asset-intensive landscape.

There are several emerging key trends in condition-based maintenance, including:

While for some industries emissions may not be a key priority, we’re seeing a shift in how organizations use condition-based maintenance as a way to help deliver on energy transition goals. With condition-based maintenance technology becoming more robust, organizations are now starting to think about how to measure and track how asset performance impacts emissions outputs.

In the past, organizations required operators to physically perform readings, inspect an asset, or analyze oil or other elements within production. Condition-based monitoring is now expanding into use cases that include cameras, robotics, infrared imaging and ultrasounds to maintain a consistent view of asset health nearly 24/7.

Organizations performing condition-based maintenance are now looking to combine physics-based models with ML models to co-mingle data to determine best maintenance path. This area, which will be discussed in the next section, is beginning to blur the lines between condition-based and predictive maintenance.

While these emerging trends are exciting, it’s critical that organizations do not lose sight of the maintenance fundamentals that help to drive condition-based outcomes. This includes structured asset hierarchies, defined asset criticalities, and ultimately an infrastructure strategy that is able to handle increased data load.

While both condition-based and predictive maintenance are focused on reducing asset failures, increasing production, and extending asset lifespans, there are key differences.

Each of these approaches have the potential to bring immense value to organizations. In fact, some of the most highly performant asset-intensive organizations employ both condition-based and predictive maintenance depending on asset criticality or importance to production.

There are several benefits to condition-based maintenance. As a leader in the asset performance management space, GE Vernova has seen the following benefits by using a condition-based approach:

Furthermore, industries across the energy ecosystem can reap the benefits of a holistic condition-based maintenance program. Listed below are some examples of operators using GE Vernova’s APM Health condition-based maintenance solution to see tangible outcomes.

It’s clear that a condition-based approach to maintenance has the potential to deliver real impact to an organization. However, it’s also evident from GE Vernova’s deep experience in these areas, that there are some key elements that could create potential friction that will need to be mitigated:

When sensor coverage or other instrumentation is spotty, organizations may get noise or inconsistencies from data. To help mitigate, it is recommended to start with a criticality analysis of assets and align these programs to the top 25% of assets.

With the increasing requirements of data that needs to be leveraged in condition-based maintenance, ultimately there is a chance siloes will be identified or created. In order to help manage this, organizations require an ingestion strategy that spans IT, OT, and ET systems.

As operators look to get investments to perform advanced maintenance, it is often difficult to determine the right use case. This leads to the potential of not delivering on the promised outcomes. To address this, it is critical that OT and IT teams are aligned on the scope of the program, technology required and expected return.

Many in asset intensive organizations have deep subject matter expertise and have been performing maintenance at a high-level for many years, this may lead to hesitancy to adopt new technology. To avoid this concern, it is important to keep SMEs involved in the program, create a clear adoption plan and document value of the condition-based program.

Successful condition-based maintenance programs typically begin with defining the scope and criticality that will be included in the program, often these are asset classes that impact safety, cost or production goals. Once selected, it’s important to validate the data readiness of the selected asset classes, including aligning the assets to proper asset hierarchies to allow for clean data to flow. After the data is determined to be validated, organizations can then integrate existing tools or deploy new tools to connect plant-level data. Once completed, users can then create proper tie-ins to other systems such as CMMS or EAM and begin to fine-tune maintenance practices based on plant data.

Once condition-based maintenance is deemed successful at the plant level, organizations have the opportunity to scale programs across the enterprise. From the first plant, organizations can develop a playbook that can be repeated and aligned with corporate asset strategies, such as Failure Modes and Effects Analysis (FMEA) and Reliability Centered Maintenance (RCM). This can enable users to expand the coverage of condition-based programs to more assets and allow for the ability to benchmark data from individual plants against one another to determine best maintenance practices.

Condition-based maintenance, when done properly, has the ability to make a major difference for asset-intensive organizations. From helping to align SMEs to assets, to deploying AI/ML for advance insights, condition-based maintenance helps to generate a foundation for organizations to get the most out of their assets.