Technology-Focused Strategies for Optimizing Asset Maintenance
In 2014, when the ISO 55000 series of Asset Management standards was launched, organisations around the world finally had a ratified, international standard covering the management of assets. Since that time, asset management has been discussed more often and in greater depth by several organisations and entities.
Nevertheless, the data collection technologies to effectively manage and maintain these assets have often lagged behind. A 2016 oil and gas industry study by General Electric Corporation determined that 60%1 of facility operators found gathering asset-related data to be a “major challenge.” Ironically, the same study reported that operators who used a predictive, data-based approach experienced 36%2 less unplanned downtime.
The volume of asset-related data that is currently available to provide insight is staggering. Analysis of this data is vital, as it is the foundation for predictive algorithms that enable facilities to detect “assets of interest” that can be viewed and analysed by an expert. Once an analyst converts the alert from an automated algorithm into actionable recommendations, maintenance can be scheduled at the right time.
A Problem Rooted in Inefficient Practices
Historically, the entire process has been fractured from start to finish, with the overall approach to data collection and processing occurring like this:
- Experts or maintenance personnel visited a plant and collected a variety of asset data from a permanently installed sensor or from a sensor and acquisition system that the expert carries.
- Experts then analysed the data to determine the enterprise’s equipment asset health and prepare reports.
- Maintenance engineers received the reports, by which time the asset health data collected was potentially already out of date.
- In the worst-case scenario, the plant had already experienced one or more equipment failures, throwing asset maintenance teams into firefighting mode and potentially impacting the corporate bottom line.
Now, let’s assume the team has moved past this rudimentary approach and is using advanced technology to collect the data. Even if they were trying to extract and interpret data for standard predictive maintenance efforts, they still would be likely to experience impediments. Why? The odds would be very good that their analysis tools and methods lacked the capacity to process the massive volumes of data modern machines can generate through automated data collection. A more effective approach must be deployed to enable large volumes of data to be collected and analysed.
Wireless Monitoring Proves its Value
If the assets are wirelessly monitored, the solution provides even more flexibility, as information from facilities — no matter how remote — can be transmitted for expert processing. Even in widely dispersed plants in isolated locations, the growth of remote Internet connectivity has made it possible for assets to transmit information. Furthermore, this capability is only going to increase. By 2027, an estimated 267 million3 active asset trackers will be in use worldwide for industrial automation and other uses.
Whether a facility is located in a remote wilderness or a massive chemical complex with advanced technological capabilities, automated wireless monitoring offers proven value. Specialists who analyse the data can determine not only the current state but also historical trends. That information can then become part of a “feedback loop.” Not only does that approach allow maintenance schedules to be optimised based on each wave of input; but it also informs future predictions that help machine maintenance technicians avert downtime.
This approach not only ensures uptime but also prevents “over maintenance,” where maintenance is performed on a more aggressive schedule than the asset requires. Finally, it provides maintenance teams and decision-makers with important insight into potential “run to failure” assets — those for which allowing them to fail is more cost-effective than maintaining them aggressively.
Kris Deckers
Referenced:
2 Ibid
