In Part 1 of this series, we covered the evolution of Predictive Service and in Part 2 we covered the five key benefits of using Flo-Tech’s Alert Management System.
In Part 3 we’ll discover how Predictive Service reduces device downtime by 80 percent (or more).
As soon as a jet engine starts, all kinds of data are monitored. By monitoring various types of data, an airline carrier can optimize the performance of its jetliners. On the other hand, if an airplane is sitting on the ground—needing repairs—that airline carrier is losing money.
To reduce downtime, airline carriers leverage predictive technology
If problems are detected that could cause a component to fail in the very near future, predictive technology tells the airline carrier to schedule and dispatch a maintenance person before the plane lands. When the plane lands, the tech is already at the gate with the right parts, making the necessary component repair proactively—getting that jetliner back up in the air as soon as possible.
A jet engine component fails at 30,000 feet. Or a CEO’s printer fails. Either way, the event is a crisis.
A crisis easily eliminated by using Predictive Service.
What is Predictive Service?
Predictive Service takes real-time historical data along with device information and predicts when a device component (or something else) will fail.
Flo-Tech’s extensive service experience lets us understand failure rates like:
- On average, how often a particular make/model will require service (e.g., every 40,000 pages)
- On average, how long certain assemblies within a make/model will last (this is referred to as the assembly’s yield; e.g., the fuser lasts 90,000 pages)
Knowing these things along with the service history of a specific unit allows Flo-Tech to build a formulaic approach to determining when service will most likely be needed (e.g., the percent likelihood of failure based on known failure rates and print volume).
Predictive Service also allows Flo-Tech to dispatch a service technician to resolve the issue—before the device fails.
Here’s a good way to think about Predictive Service
The average copier has anywhere from 100–200 counters in it. Think of a counter as a gas gauge.
These gas gauges can register “full” and “empty.”
Now, take those hundreds of counters and multiply them by the number of devices you’re monitoring, and you’ll get into the tens of thousands quickly; that’s because you’re looking at so much information—an impossible task.
What we’re doing with Predictive Service is looking at all those gas gauges with computers—not people.
Predictive Service can tell when you’re going to reach empty based on how much you’re “driving” or, in this case, how much you’re printing.
Predictive Service is about learning patterns
Take a component like the swing plate.
HP doesn’t publish a yield saying the swing plate needs to be replaced every million pages, but Predictive Service discovers that yield over time. That component is then added to the device’s profile of what gets replaced.
So this principle can be applied to a component that doesn’t even have a counter; that’s because we know from history that the average swing plate is going to last one million pages.
Predictive Service then goes to the next level
We know which inventory items the device needs, based on the device’s history.
Does it need
- A fuser?
- A transfer roller?
- A swing plate?
- A combination of components?
Predictive Service is able to look into a technician’s real-time parts inventory to see if that tech has the components needed to make that repair. So Predictive Service is not creating another activity—it’s making sure that the tech has all the right qualifications and parts when arriving at the customer’s office.
In short, Predictive Service gives us detailed specifics so the device is fixed right—the first time.
Predictive Service in action
Let’s say there are two identical devices that have the same model number with the same manufacturer-stated yields.
One device is in a law firm where they use heavy and thick paper, which would wear more heavily on the device; the other model is in a human resource department using regular copy paper.
Predictive Service has the intelligence to know that the device with the heavy paper will break more often, because that paper wears down the components faster than in the scenario with the HR department using regular copy paper.
Over time, Predictive Service sees the differences in performance and will adjust accordingly on an ongoing basis.
Here’s a graph visualizing Predictive Service.
How is Predictive Service different from AMS?
Imagine three users all using one group printer. Mary printed to the group printer and had a jam.
She removed her jam, finished her job and walked away without placing a service call. She didn’t think anything of it because copiers copy, print and jam.
Joe did the same thing an hour later, and Fred did the same thing the next day.
No one’s connecting the dots and saying, “Hey’ something’s starting to fail in this device.”
AMS connects those dots, notices a pattern and generates a service call while dispatching a technician.
Predictive Service, on the other hand, knows how many pages a device is running per day. It knows what components need to be replaced and at what intervals.
Predictive Service equates that to how many days we’re estimating before that component fails and making sure that the tech notices it and schedules a service call—prior to the estimated failure date.
Response time without Predictive Service
You’re in the office. Your boss says, “I need seven copies collated and stapled.”
You press print and your device displays an error code—and you can’t fix it.
So you call the help desk, place a call to the service provider and wait for your technician to respond, based on your contract’s SLA (service level agreement) response time.
We’ll use a typical four-hour response time for this example.
The device failed at 9:00 a.m., and that’s when you placed the service call.
The technician arrives at 1:00 p.m. When the tech arrives, the tech might—or might not—have the parts needed to repair the device.
After working on it for one hour, the tech has fixed the device, and you’re back up and running.
Your downtime? Five hours.
Response time with Predictive Service
Looking at the Predictive Service application, the tech notices that a device has a component that is likely to fail within the next few days. The tech then services that device before it goes down.
The downtime for that device? One hour.
The difference between five hours’ downtime and one hour of downtime?
With Predictive Service, your company runs more efficiently and with fewer disruptions.
Here are several key benefits that can be realized when your organization uses Predictive Service:
- Devices rarely go down without warning
- Downtime is minimized to repair time only
- Emergency calls are avoided
- Devices are fixed the first time the tech arrives—avoiding unnecessary follow-ups
- Service calls can be group-scheduled based on location, saving time
- Devices needing service match the company’s criteria and goals—not what the tech thinks
- Overall customer involvement is reduced by (up to) 40 percent
Subscribe to Flo-Tech’s Blog
When you receive Flo-Tech’s blog updates, you and more than 3,200 other smart CIOs will get the latest managed print service articles delivered right to your inbox.
- How to manage your print costs (even on a tight budget)
- Why optimizing your work flow is easier than you think
- How to secure your print environment
- And more