Not all failures occur with age.
This realization revolutionized maintenance philosophy when engineers working to improve the operations of the Boeing 747 in the 1950s published their data. Instead of assuming that assets failed on a schedule and the best use of maintenance resources is to predict when that failure will occur, maintenance leaders now focused on monitoring the conditions for failure and tying the consequence of failure to the maintenance practice.
It’s called Reliability Centered Maintenance (RCM), a framework to determine the most cost-effective maintenance practices for each asset by weighing its functional significance to the system and examining the data on its (and similar assets’) failure and maintenance history. The potential savings for businesses are huge: two researchers found that implementing an RCM program at a utility company reduced maintenance costs by 30 to 40 percent. After use by the aviation industry in the 1960s, its benefits have now caught on with companies of all types, including utility plants, theme parks and manufacturing facilities— and now, digitally connected utility plants.
Data plays a huge role in RCM strategy, powered by the ability of companies to integrate multiple data sources into one seamless dashboard for employees and communicating actionable insights from that data to the right devices for employees. The result: A reduction in labor costs, asset downtime and catastrophic failures, along with ultimately increasing the efficiency and ROI of plant assets and reducing failure sources.
The most recent standard for RCM, published as SAE JA1011, defines the minimum criteria a process must meet to be considered RCM. The criteria, which should be worked chronologically are:
- What is the item supposed to do and its associated performance standards?
- In what ways can it fail to provide the required functions?
- What are the events that cause each failure?
- What happens when each failure occurs?
- In what way does each failure matter?
- What systematic task can be performed proactively to prevent, or to diminish to a satisfactory degree, the consequences of the failure?
- What must be done if a suitable preventive task cannot be found?
This model allows you to prioritize how you perform maintenance based on what you can do to prevent failure and what the consequences of failure are. In a simplified form, it’s akin to the smoke alarms in your home alerting you (with incessant beeping) before the batteries have run out; a smoke alarm is critical to the functioning of your home and failure could be catastrophic, so therefore, you will opt for condition-based monitoring to have the machine identify and draw attention to a sign the failure is likely to occur (battery power running out), rather than a run-to-failure approach.
Rather than selecting a purely reactive or purely predictive approach for the facility, RCM allows you to mix and match the strategy to the asset. Here’s an example of optimizing the maintenance strategy for each asset:
|Strategy and Definition||Ideal for|
|Reactive: Maintenance technicians focus on running items to failure. No monitoring is required.||Only non-critical assets. This asset will fail during operations; keep safety risks for employees and spare part availability in mind.|
|Preventative: Parts are replaced on a fixed, time-based schedule.||Assets that have an established failure pattern. This may lead to unnecessary replacements and larger inventory costs.|
|Predictive/Condition-Based: Assets are monitored for conditions indicating failure is likely to occur.||Assets that are relatively simple systems where a small set of conditions can reliably predict failure.|
|Proactive: Assets are monitored for conditions, and diagnosed for possible causes of failure.||Assets that are critical to the business functioning. This takes a large upfront investment in data analysis and diagnostic modeling.|
The data of the digital plant plays a role in each of the above approaches, but most critically in the Proactive system and most strategically in the decision of which approach to use. Connecting your assets and components and powering them with a sophisticated analytics mechanism can help to decide the maintenance strategy and analyze data on current functioning. A non-critical asset experiencing random failures may be optimal for a reactive approach; critical assets that fail regularly within a certain amount of time need a preventative schedule.
When implementing digital plant technology, remember that the increases in ROI only happen when it is implemented correctly with enough resources, which includes setting up an efficient, mobile communications system with maintenance technicians.
When RCM Fails
“The first rule of any technology used in a business is that automation applied to an efficient operation will magnify the efficiency. The second is that automation applied to an inefficient operation will magnify the inefficiency.”– Bill Gates
Unfortunately, research from ReliabilityWeb, indicates that more than 70 percent of RCM programs fail to generate the desired cost savings or asset productivity results, possibly due to the fact that additional research shows only 5 percent of RCM programs are implemented properly. Our hypothesis: maintenance and operations teams fail to make the critical investment in data analysis and communication with employees when preparing to implement RCM. At many plants, frequent mergers and acquisitions for companies, as well as a lack of digitization of old paper systems, translate to a cost-prohibitive implementation of a digital maintenance and failure history.
Our recommendation is to start small by investing in analytics on real-time data to start delivering ROI, and use that to make a business case that investing in the data history and analysis required for RCM will generate increases in plant reliability, decreasing labor costs, and more significantly, decreasing costly plant or equipment downtime.