The ability to perform predictive maintenance on a machine may not be ubiquitous today, but it will soon become a standard offering on packaging and processing equipment.
“If an OEM chooses not to provide predictive maintenance, they run the risk of long-term obsolescence,” said Blake Griffin, principal analyst at Interaction analysis, a global market research firm focused on the entire automation value chain. “[Predictive Maintenance] is an unstoppable trend.
Griffin presented the findings of a new PMMI white paper “Packaging and Predictive Maintenance”, during the Executive Leadership Conference today, April 19, 2021. Predictive maintenance is defined by Interact Analysis as a means of monitoring a machine, or a component of a machine, to determine when it is likely to fail and taking action to repair it. shut down, thus avoiding downtime.
“It’s a technology built around data and network architectures, but the definition we gave it [includes the] using technology to collect asset data, such as temperature or vibration levels, and perform analysis on that data to predict when the asset needs repair to eliminate the risk of failure,” Griffin said. . “It’s always about the bottom line from the user’s perspective. If you can improve equipment efficiency, you can improve profitability as a business. It is about avoiding and optimizing risk, these are the two main value propositions of predictive maintenance. »
It’s also a way for OEMs to reconnect with manufacturers to increase customer satisfaction with service, Griffin said, citing another PMMI report from a few years ago that found the customer satisfaction with OEM service had declined from 42% to 21% between 2015 and 2019. due to an aging workforce carrying a lot of native knowledge retiring and new, less experienced talent retiring. arrive. And predictive maintenance is essential in machines where operation cannot be replaced by manual labor, such as filling and metering, forming, filling, and sealing, and labeling, decorating, and coding.
So, with 45% of CPGs already piloting or using predictive maintenance technology and 29.4% evaluating it, now is the time for machine builders to offer it as a service.
Source: PMMI report on packaging and predictive maintenance.
There are two different approaches to predictive maintenance: sensor-based and software-based. The first version uses wireless smart sensors to communicate asset health. Once the sensors are deployed, edge computing devices could be linked to the system to transmit useful data to the cloud to perform the analyses. The software approach, on the other hand, exploits the intelligent devices already installed on the machine, such as drives and the PLC.
“If you have a lot of sensing components around the machine, that gives good insight into the behavior of the machine,” Griffin said. “Additionally, safety stops can indicate the stress levels the machine is facing.” The software option will need to be flexible enough to accommodate the emerging Internet of Things (IoT) platforms that many automation vendors are deploying, but in proprietary formats. “Don’t lock yourself into a solution by being too dependent on a solution. Think about the context of the whole operation, not just the context of your machine. »
Ultimately, OEMs will want to make money on a predictive maintenance service, but there are often roadblocks to that. For example, their customers can restrict remote access. There is also a fear of cannibalizing revenue associated with service level agreements (SLAs) or fear of cannibalizing replacement revenue. “SLAs just need to be redesigned around an availability factor. If you guarantee 95% uptime, you price based on the downtime that affected the operation, and anything above 95% is charged a premium,” Griffin said.
It’s also a way to differentiate businesses, and as end users see the value in it, they’ll lift those remote access restrictions. Additionally, as predictive maintenance becomes pervasive in the industry, OEMs may develop more complementary services or offer completely new business models such as machine as a service.