Reliability Centered Maintenance (RCM) and the Next Evolution of Data Modeling

An Introduction to Reliability Centered Maintenance

Reliability Centered Maintenance (RCM) is a popular asset management strategy for prioritizing maintenance tasks for non-fixed equipment. For heavy processing industries, knowing where to focus maintenance efforts is a crucial aspect to ensuring asset reliability. By knowing where to prioritize resources, you can efficiently reduce unplanned failures while also realizing the most value for the cost.

But how can you easily get this information? With thousands of assets to monitor, getting the data right can be challenging.

This quest for the “right data” has led to the creation of many different methodologies. Reliability Centered Maintenance (RCM), for example is a proven method for prioritizing maintenance strategies based upon asset risk. While RCM does provide efficiencies, it also leaves room for more finely tuned data, which leads us to the next evolution in reliability data modeling: Quantitative Reliability Optimization (QRO).

Let’s take a look at how maintenance strategies have evolved with RCM and how it will continue to evolve with QRO.

What Is Reliability Centered Maintenance?

Reliability Centered Maintenance (RCM) is a maintenance and asset management strategy for facilities with high consequences of failure. RCM delivers a set of criticality-based, proactive tasks, focused on sustaining functionality of systems and equipment, to improve reliability and safety for asset intensive industries.

An RCM program is also a key aspect to comply with the U.S. Occupational Safety and Health Administration (OSHA) Process Safety Management (PSM) regulation. This regulation was created to ensure safety for facilities that contain highly hazardous chemicals, and it requires process facilities with these chemicals to have a documented plan that specifically addresses the Maintenance of the assets containing these highly hazardous chemicals.

Reliability Centered Maintenance: An Evolution in Maintenance

RCM was introduced in the airline industry in the 1960s. Since that time, several other industries—such as nuclear, oil and gas, and aerospace—have adopted the practice.

Prior to RCM, organizations relied on reactive maintenance as a primary asset management practice. A reactive approach is highly inefficient because it requires a large number of resources (and hence, high spending) to maintain the maintenance program. It also does not have significant impact on improving an asset’s reliability, and, in some cases, may increase rates of failure by introducing infant mortality.

When RCM was introduced, it was a game changer because it transformed the way asset intensive industries performed maintenance. It did this by enabling a maintenance program to be driven by risk, not time. By using risk as a guide, companies could have much more efficient maintenance strategies by gaining the ability to prioritize risk and appropriate mitigation tasks, thereby increasing availability and better utilizing resources.

What Does the Future of Reliability Look Like?

Over the years, RCM has been a valuable method that has helped complex asset-based systems effectively maintain asset reliability through cost-effective strategies. Now, decades later, advancements in data acquisition, warehousing, modeling, and analytics are creating opportunities to improve upon the RCM model. The next leap in reliability will further improve availability while continuing to reduce maintenance spend.

An RCM program is the first step after a reactive program in the evolution of maintenance program maturity. However, when it comes to further optimizing and improving reliability performance, Reliability Centered Maintenance can be limiting depending on the objective. Specific limitations include the following:

• RCM does not calculate absolute risk but rather relative risk, using that to drive maintenance priorities rather than an objective cost/benefit analysis.
• RCM models are typically very conservative around the calculation of probability for failure and consequence of failure, due to the fact that they are predominately based on likelihoods that the event or failure mode would occur that often stem from SME knowledge.
• RCM analysis are static by nature and typically do not incorporate the condition data of the asset to update the risk calculations (especially the Probability of Failure calculations) in order to drive maintenance priorities accordingly.
• RCM calculations occur on asset-by-asset basis and generally do not relate to the overall performance of the system, unit, or facility.
• Although RCM program architectures are generally similar (criticality analysis, FMEA, etc.) they are subject to interpretation regarding how to the analysis is structured.
• RCM does not help quantify the value of data collection or help with sensitivity analysis of required data for calculations beyond manual Iteration of values from the user.
• RCM cannot be used to optimize an entire system, unit, or facility’s reliability strategy based on availability, cost, and resource constraints.

Whether you are just starting to implement an RCM program or are already using a matured program, you have the necessary tools to make the next leap in reliability. We believe this leap is being made possible through Quantitative Reliability Optimization (QRO). Quantitative Reliability Optimization (QRO) is a method that pushes RCM to the next level by unlocking its capabilities through the use of dynamic data analysis. Using actual asset data, QRO provides detailed insight when it comes to understanding when an asset will fail, identifying the impact each asset has on the larger system, and knowing how and when to use resources improve the results you care about.

QRO is an approach to reliability modeling which connects every relevant reliability data point to one integrated model that enables users to do things such as:

• Optimize, in near real-time, all maintenance spend based on short/mid/long-term reliability targets.
• Understand the economic value of every maintenance activity based on dynamic condition models (such as Probability of Failure) that update in a near real-time manner as new data is introduced in order to quantify where an asset is on its P-F Curve to determine when tasks should be performed.
• Understand the quantifiable impact each asset has on availability
• Understand the economic value of every piece of data that is currently being gathered or possibly gathered in the future.
• Simulation analysis to evaluate the impacts of maintenance and operational activities, including the implications of moving a turnaround, feedstock pricing changes, or various capital projects.
• Drive effective economic decisions in the event of reliability based operating excursions.

Learn more about Quantitative Reliability Optimization (QRO).