How asset management software enables prescriptive maintenance programs

They understand that RxM is an extension of, not a replacement for, current maintenance best practices. Condition monitoring, which detects asset degradation, will continue to fuel predictive maintenance (PdM), which drives preemptive actions in time to avert failure. Adding RxM reinforces the analytics to prescribe recommendations that drive improved outcomes.

The appeal of augmenting proven practices with prescriptive analytics and maintenance is twofold: to widen the pool of available asset intelligence, and to have evidence-based correction options prescribed with greater specificity. Therefore, maintenance analysis and planning time is reduced, actions are more precisely timed, outcomes are more predictable and effective, and unplanned downtime is more avoidable than ever before.

• Read "EAM and ERP in the era of the IIoT"

Ultimately, building prescriptive capabilities into EAM solutions enables more-predictable reliability. It makes maintenance and reliability professionals, who work hard to avoid downtime and are the heroes to operations when they bring failed equipment back online, also heroes to the C-suite because limiting business interruptions increases revenue and profitability

To make this case, EAM solution providers are working alongside digitalization partners to drive up the number, scope, and value of validated RxM proofs of concept.

New technologies invite innovative opportunities

RxM’s strength is that it learns from – and bases maintenance recommendations on – the holistic asset status: physical, environmental, operational, and historical. This capability would not be possible without the emergence of the industrial internet of things (IIoT) and Industry 4.0.

The IIoT is driving an explosive growth of data from operations and maintenance, monitoring and inspection, engineering, process safety, and related systems. It is also making dark data such as maintenance logs, warranty claims, emails, videos, and engineering drawings accessible with help from artificial intelligence (AI) and machine learning (ML). The result is a more complete picture of asset health.

Prescriptive analytics enable higher learning from the data. AI/ML algorithms, data trending and modeling, advanced pattern recognition, failure prognostics and diagnostics, digital twins, and augmented reality (AR) help to better detect and analyze asset wear, predict the time to failure, and suggest the timing and course of action for maintenance, repair, or replacement. The AI/ML component propels continuous improvement of the outcomes.

These developments are already paying off. For example, there are automation technologies with embedded sensors and EAM integration that can virtually request maintenance for themselves. They can inform operators about how and why the equipment is degrading, what best, specific course of action is needed, and what time frame is optimal to avoid failure.

What does RxM mean for EAM?

Prescriptive capabilities heighten the function and effectiveness of EAM software. RxM elevates asset intelligence, which improves the value of information displayed in EAM dashboards, diagnostic reports, and mobile devices and optimizes the quality of the recommended actions. This newfound cognitive support has the potential to be game-changing for maintenance teams and beyond by:

[pullquote]

• Reducing the time spent on manual inspections, condition data collection, and analysis by having machine sensors from multiple sources track live conditions and alert when a threshold (or combination of thresholds, or some other decision rule) is crossed
• Improving work planning by having distilled predictive and prescriptive analytics prescribe evidence-based options and timing for corrective action, minimizing the effort to research the condition, cause, history, and potential solutions and avoiding the need to visit the asset to personally assess its condition
• Improving work scheduling due to better failure predictions and prognostics, a more-defined window of opportunity, and better coordination of labor, parts, and tools availability
• Reducing the mean time to repair (MTTR), mean time between failure (MTBF), and cost of failure by facilitating specific, timely, data-driven maintenance recommendations
• Minimizing the risk of rework by equipping the technician with a thorough work order and all the information and materials needed for safe and successful work completion
• Optimizing inventory and avoiding critical part stockouts by having the analytics prescribe inventory requirements with ample lead time for replenishment, whether by purchase requisition, stock transfer request, or 3D-printing order
• Optimizing equipment warranties by having the analytics flag failing assets covered under warranty and prescribe the steps to submit a warranty claim for service, replacement, or compensation
• Read "How mobility is reshaping asset management best practices"
• Avoiding perpetuating the risk of an aging workforce, whose wealth of tacit knowledge is not found in any system, by using AI/ML to bring dark data online, automate learning, fuel predictive and prescriptive analytics, and help the organization transition from intuitive to evidence-based decision making
• Improving design and engineering decisions by learning from patterns and failure signatures and developing better ways to avoid or mitigate adverse conditions
• Improving asset investment choices by leveraging prognostics and asset and vendor histories in the analytics, prescribing reliable and effective product choices, and strategically timing the acquisition and installation
• Supporting knowledge-sharing opportunities with third-party maintenance and reliability service providers, whose valuable data on similar customers using similar equipment in similar industries under similar operating conditions could improve the analytics
• Enabling knowledge-sharing opportunities with OEMs, whose wealth of information about how their assets are used, common indications of failure, and how to optimize reliability, performance, and longevity could improve the analytics

Get the most out of the technology

As good as it sounds, RxM faces challenges similar to PdM. For example, if there is bad data in the EAM, it could affect the analytics coming out. The deluge of new sensor-based condition data is not in question, but some EAM systems contain uncleansed data from legacy systems or spreadsheets. Consider this when setting up the prescriptive data sources and rules.

Similarly, like PdM, user adoption influences RxM’s success. If technicians keep crucial data in personal spreadsheets, or if they fail to log important details when closing out work orders, or if root cause failure analyses (RCFA) are not conducted, then the full potential of RxM will not materialize because the best prescriptions cannot be deduced from the available data.

Additionally, program funding could lapse if the C-suite does not see the ROI. Start with the most critical asset and pain point, create a proof of concept, push the ROI findings up the chain of command, and repeat this process over and over to ensure sustained buy-in.

Remember that it is RxM’s nature that the quantity, quality, and intelligence of actionable EAM information will increase as more data is added and the AI/ML-fueled analytics become more intuitive and sophisticated. As such, it provides a sound and continuously improving validation of the value of digital transformation.