The promise of integrating internet of things technologies with enterprise asset management systems is that it can help companies act on real-time insights by tying together disparate systems, sensors and information.
All too often, companies aren't taking advantage of all the internet of things (IoT) data that's being captured in their enterprise asset management (EAM) systems, and the potential business results aren't realized. However, there are ways to get better answers from IoT data.
Organizations can start by organizing and managing the tsunami of sensor information flowing from IoT-connected assets, according to Mark Konya, principal industry consultant in the global energy and utilities practice at SAS Institute Inc., based in Cary, N.C.
Getting meaningful answers depends on state-of-the-art analytics tools that are capable of returning results quickly, Konya explained. Making this happen requires flexible deployment strategies, whether that involves embedding a model on the edge where response times are crucial or visualizing the health of asset fleets in a central facility to see how vehicles on the road are performing back at a warehouse or office.
"Better answers from IoT data requires all of these, plus a high level of integration between data management, analysis and deployment environments," Konya said. "Integration is the glue that holds it all together; the grand unifier that makes better answers possible from IoT data."
Tools can help get better results from IoT data in EAM systems
One tool that can do this is IFS' EAM functionality, which enables companies to easily track their assets. The goal of the IFS EAM system is to drive a better business result, according to Rick Veague, CTO of IFS, based in Itasca, Ill.
"The question is, are we using data to drive a better outcome that directly aligns with what the business is trying to achieve?" Veague said. "IoT brings a new twist to it because there are new technologies for capturing data, and the volume of data that we get from IoT is much greater than anything we've had in the past, so, potentially, that opens up new opportunities. The question is, what are those opportunities and how do you capitalize on them?"
From Veague's perspective, it's about how to operationalize that data -- that is, how to use that data for something useful.
Rick VeagueCTO, IFS North America
The problem is that companies are keeping the data that's generated from their IoT systems in silos, and it only exists within the realm of the machine or the asset itself, according to Veague. Even though that data is sometimes used in small ways to track maintenance activity on a machine, it isn't really tied to a more generalized business result.
"The goal is to actually find new, useful ways of using that data to drive business results," he said. "So that's where IFS comes in, and that's where the opportunity is -- getting the IoT data out of its silo and into not just the EAM system, but the broader business systems that can use that data."
The IFS IoT Business Connector captures IoT data as it accumulates, filters it and pulls it directly into the company's EAM systems, Veague said. Customers then use that to simplify and facilitate that flow of data.
"IoT generates large amounts of data that generally pile up in some front-end edge system," he said. "So our solution connects to those front-end IoT edge systems and pulls that data out. We apply what are called stream analytics, which is a way of essentially correlating that IoT data -- those readings -- to look for interesting things that indicate something a company wants to track."
Pest control company Anticimex Finland, a subsidiary of Anticimex International AB based in Stockholm, uses the IoT Business Connector for this purpose.
"By feeding sensor-captured IoT data from thousands of smart traps via the IFS IoT Business Connector into our business software IFS applications, we are able to react much quicker to service needs, and even work proactively to prevent problems before they arise," said Jussi Ylinen, managing director at Anticimex Finland.
Using IoT data, Anticimex can detect early on when the batteries in traps need to be changed, or if the traps must be emptied.
"Together with Microsoft Azure machine learning, we can find faulty trap batteries and analyze battery depletion patterns," Ylinen said. "We also leverage the IoT data as a foundation for optimizing the service technicians' travel routes within IFS' [EAM] application.
"We see clear benefits in harnessing IoT to offer more intelligent pest control solutions to provide premium customer service and, at the same time, save both time and money. This is true digital transformation for Anticimex."
SAIL on to better answers
Kevin Price, EAM product evangelist and strategist at the ERP vendor Infor, said to get better answers out of IoT data for EAM systems, enterprises should consider the acronym SAIL: sense, analyze, infer and learn.
"You can look at historical information in EAM, like the failure codes. You can look at data that's coming in from SCADA, [supervisory control and data acquisition -- a computer system for gathering and analyzing real-time data], from process logic, from building information, from building monitoring," he said.
The idea is to make sense of the data streaming in from the internet of things, according to Price. IoT data fed into EAM systems can be used to track asset reliability, leading to predictive maintenance, which can help predict failures. But all of that the data has to be analyzed, which can be difficult.
"Once you get the data, you need a reliability engineer -- someone who's been working on the engineering side -- or someone with tribal knowledge who has been working on a piece of equipment for 30 years," Price said. "Someone who knows that, when I get this alert, I see this happen and this is the action we need to take."
However, this isn't always easy because companies typically don't have that tribal knowledge about older pieces of equipment. That's where reliability engineers come in -- they can help you understand what to do with that analytical information, Price said.
Once the organization gets that data, it has to infer what to do with it; for example, if one particular thing happens, then a work order needs to be created.
"That type of infer stuff is not as difficult as the S [sense] and A [analyze] because you already know it -- you're just inferring what to do in that workstream, for example," Price said.
Learning means changing behavior
The L part of the SAIL acronym is very hard because learning from data to make the best business decisions often requires a change in company culture.
"You're changing [employee] behavior," Price said.
Price gave the example of a trucking company whose maintenance technicians were constantly replacing the starting components on the company's trucks. The technicians argued that when the drivers made deliveries in cold weather, they shouldn't shut off and restart their vehicles because it damaged the components.
"But the operators thought the technicians were stupid and said they should be able to shut off and restart their trucks in any weather," Price said.
So the engineers gathered and analyzed the data from the connected vehicles and conducted a telematics review. They determined that the operators were wrong, and they had to keep the engines running in cold weather and not shut them off and restart them -- a change in behavior that wasn't easy for the drivers to accept.
"We've been doing asset tracking and condition monitoring in our system for a very long time, but now there's so much data that's coming in that you have to take a step back," Price said.
"The most important thing is to ask why you're doing something. For example, are you trying to lower your energy bill? You can go on a path of gathering data and trying to do something, but it doesn't have that much of an impact if you're not focused on [delivering value to the business]."