Tapping predictive maintenance with mobile, IIoT
Organizations can improve asset reliability with predictive and condition-based maintenance based on asset health insights from operational data and analytics.
The ability to make repairs before they’re really needed has long been an ambition for plant managers. For organizations that manage large numbers of complicated assets, predictive maintenance accomplishes a number of aims: reduced downtime, lower maintenance and replacement costs, greater productivity, safety and even improved sustainability and resilience. Through predictive maintenance, organizations can improve asset reliability with condition-based maintenance based on asset health insights from operational data and analytics, as opposed to routine scheduling or reactive repairs.
For larger organizations, this can be easier said than done. A generation plant, power plant or refinery may have hundreds or thousands of assets, from production equipment to safety equipment to the HVAC systems running through their facilities.
However, it’s now possible for sensors, beacons and cameras to monitor all those assets, and with a strong hybrid cloud digital infrastructure you can access whatever data or software you need wherever you are. Today, AI can assign holistic asset health scores and even help determine the likelihood of breakdowns and outages. However, if these insights are restricted to the realm of data scientists producing insights at headquarters, their utility is limited. By the time said insights make it out to technician in the field, they may be irrelevant. There may be new priorities or breakdowns that matter more.
For the derived insights from operational data and analytics to be useful, they need to be applied quickly. This can only happen if predictive maintenance is accessible in the field. After all, technicians in the field are the people we rely to keep the power on, to fix broken parts and keep utilities, factories and refineries operating smoothly. Particularly as assets grow in sophistication and complexity, technicians need to be able to take all the information they might need to know about it everywhere they go. In short, predictive maintenance needs to be mobile.
Bringing predictive maintenance to the edge
Mobile technology makes predictive maintenance more actionable. When technicians can access a sophisticated enterprise asset management system in the field, they waste less time traveling back and forth researching repairs. By enabling a disconnected mode, organizations can make it possible for technicians to access operational data, scheduling optimization, asset health scores and even guided repairs. For companies with assets in the field, this ability is even more critical.
With mobile enterprise asset management (EAM), users can make it possible for technicians to read schematics, update data and perform inspections even in the most remote locations. Native mobile features – like tap and touch or taking smartphone pictures – promote efficiency. Virtually everyone knows how to use a smartphone. Technicians can even use voice-to-text functions and access GPS to better navigate to locations.
Field technicians can take a robust “toolkit” with them anywhere, which can make it easier to manage asset health overall, leading to a more resilient, profitable and sustainable organization.
Democratizing access to AI
In many ways, the breakthroughs in mobile EAM technology aimed at technicians represents an important step toward the greater democratization of AI. Traditionally, the massive data output from a company’s many assets would have required data expertise to analyze, and AI-based expertise to build models and understand how the models were making decisions. Thanks to a number of technological developments, from hybrid cloud that enables software and data to be stored and run anywhere, to the greater connectivity afforded by 5G, the benefits of AI are much easier to realize.
In the last few years, the technician of the future has come into much sharper focus. Thanks to more accessible applications of AI, a technician can now take a smartphone picture of an asset, and AI will identify and annotate likely faults. The AI can tap into the organization’s operational data and analyze all the similar parts that had breakages and help the technician figure out what the likely problem is. Virtual assistants can guide the technician through repairs and take them through tasks step-by-step. Augmented reality (AR) also can be used to connect technicians with experts to walk them through the right fix, the first time. Technicians can even create a digital twin of the asset to study its ins and outs. Even better, they can perform all of these tasks on a mobile device.
The mobile revolution has always been about accessibility. The computing power found in the first smartphones had technically existed for decades in government supercomputers and at NASA. What made smartphones a paradigm shift was it put computing power into people’s hands.
As mobile technology improves even more, other even more advanced technologies like AI become more accessible as well. By putting AI into the hands of their workforce, organizations can make predictive maintenance a reality, leading to a more healthy and profitable organization overall.
Joe Berti is VP product management, AI applications, with IBM Cloud and Cognitive Software. IBM is a CFE Media content partner.
Original content can be found at Plant Engineering.