Not long ago, it was hard to imagine a field service world in which providers could predict exactly when equipment would have issues. But today, we’re closer than ever to perfecting the art of proactive maintenance, says David Ovadia, director of product marketing for GE Digital.
Proactive (or preventive) maintenance is based on a combination of historical data and real-time information from equipment sensors, artificial intelligence and machine learning, and can monitor patterns to provide predictive insight into equipment issues. This can lead to better inventory control, lower service costs, fewer truck rolls through increased remote repairs, and happier, more loyal customers.
“You have to know what to listen to, what to look for. [Through preventive maintenance] you can reduce downtime, which has great benefits. You will know how equipment fails. You will know which equipment is older and is more likely to fail,” Ovadia says.
At a rudimentary level, preventive maintenance has been around for some time, Ovadia points out, but thanks to advances in technology like artificial intelligence and the growing trend toward servitization, it’s getting more powerful than ever.
The Origin and Evolution of Proactive Maintenance
Historically, there have been three types of maintenance strategies, according to Ovadia. First and most basic is reactive maintenance, which is designed to fix devices when they break. Second, there’s time cycle-based preventive maintenance (PM), which forecasts equipment failure using a time-based performance curve, Ovadia says.
Failures tend to be very low during the first two-thirds of a hardware device’s expected life, while most items of the same type will wear out at the same time. The solution is to plan preventive maintenance a short while after the maintenance curve starts moving up, but before it rises sharply. And, finally, there is also condition-based predictive maintenance, which works like time-based maintenance while also factoring device condition into the preventive maintenance schedule equation.
Given a number of different market forces and trends today, however, these three maintenance strategies have been pushed to their limit, continually maturing to meet customers’ needs.
For one, there’s the servitization trend. Twenty years ago, more than a quarter of specialized tech equipment came from Asia, but that figure has more than doubled to 53 percent today, with much of that coming from China. As hardware has become more commoditized, margins have eroded sharply, so technology firms have altered their focus from hardware sales to sales of software and service, according to Ovadia, giving rise to a greater focus on maintenance.
Perhaps an even more important factor driving the evolution of proactive maintenance that unplanned downtime is extremely costly, Ovadia says. Parts may not be readily available and may need to be expedited even if they are. If parts are difficult to get, they may result in significant downtime, meaning a very unhappy customer, Ovidia points out. The cost is even more severe for industries operating at capacity because their customers have mission-critical systems that must be operational or they lose business. These challenges reveal the serious limitations of reactive maintenance.
Using historical information to plan predictive maintenance—both time-based and condition-based—avoids many of these problems. But now, real-time information takes these services to the next level. Using the combination of Internet of Things (IoT) sensors, artificial intelligence (AI), and machine learning means customers can proactively anticipate equipment failure, either preventing it altogether or taking advantage of planned downtime, according to Ovidia.
Innovating Maintenance, One Data Source at a Time
To help manufacturers harness the power of proactive maintenance, ServiceMax has integrated with GE Digital’s Asset Performance Management (APM) to provide an end-to-end system for failure prediction and proactive maintenance. With this solution, after-sales service departments can transform from reactive break-fix and planned maintenance service offerings to condition-based maintenance rooted in equipment condition and predictive indicators of potential failures.
The ServiceMax APM solution includes maintenance workflow support tools for exception-based issue management as well as triage support, reliability analytics, and reporting capabilities. Service providers that use it also receive predictive analysis via the web, enabling them to assess the issue remotely, with 30 to 80 percent of repairs (depending on the type of equipment) performed remotely. Even if a truck roll is warranted, the technician knows in advance what the likely problem is and which part(s) and tool(s) will be needed to fix it.
As a result, downtime is avoided, inventory needs are less urgent, and the customer is delighted, meaning the customer is more likely to stay with the service providers at service renewal time and when the equipment reaches the end of its useful life.
As one customer put it: “I need preventive maintenance because I cannot live with the uncertainties of an unplanned failure in the user’s device, something that could have been avoided that may ultimately cost someone their life. I need to be certain that my customer’s faith in me and my team is well-founded, that we won’t let them down through negligence, disorganization, or inefficiency.”
This topic was presented as a part of ServiceMax’s Maximize Chicago conference in October 2019. The session titled “What Proactive Maintenance Strategy is Right for You in an IoT World?” was co-presented with Kathryn Narayan and Lacy Cotton-Hodgson from ServiceMax.