It has been my experience that when applying the proper predictive tools strategies to critical assets, it gets confusing about where to apply capital expense, training, and people. Here are 3 factors to help determine which predictive maintenance tool to use first:

1. Always start with the asset life cycle plan (6 months, 6 years, etc).
2. Consider criticality of the asset in relation to throughput and quality of the product or process.
3. Evaluate the time needed to train resources and expect some competency with the tool(s).

After deciding if you will operate the asset for a short time or the long haul, decide which tool can be most useful. Most of the time ultrasound and infrared have the shortest time needed to train personnel to a beginner level of competency. Also, the costs are much lower than vibration or motor circuit analysis. However, there are instances where the benefits of vibration and motor circuit analysis will outweigh the training and delayed timing.

Keep in mind other tools, like hi-speed video, fiber scopes, and laser alignment, all have their place when planning for long-term life cycle of critical assets.

For a deeper dive into predictive maintenance and the challenges associated with the Industrial Internet of Things, check out our guide:

How valuable is it to make precise asset failure predictions?
1. Can we make more precise calls around Failure Probability.
a. You found an anomaly – How long before it fails?
b. Do I need another predictive tool to validate what I found?
c. Do I need to call in outside help?
d. Can we get by until the weekend? End of Month? Etc…

EXAMPLE:
Problem identified
Broken Gear Teeth.
The use of a continuous Vibration monitoring and and proper Alarm levels allowed the problem to be identified early. At the top of P-F curve.(Predictive Failure Curve). Giving the throughput Team 8 – 10 months lead time.
No unscheduled downtime occurred.
4 days of scheduled downtime.
800 proactive labor hours were required
to make repairs.

If the problem had not been identified and the
press failed, it would have resulted in 10 days
of unscheduled downtime, and 2640 reactive
labor hours to make repairs.
Cost Avoidance: $545,086.00

Note: Information is based on Data from a
similar press failure that occurred in 2012.

The early detection of a problem, allowed the production team to keep the asset operating through a critical launch, saving the company $545,000.00.

Maintenance Teams struggle with inconsistent, and varying levels of data collection. Also, inconsistent and varying levels of talent and data gathering, from site to site.

By using newer Big Data Tools like Maximo & Predictive Maintenance Quality with Vibration, Infrared, and Ultrasound, we can develop precision data trending and patterning.

“We are only valuable to the Manufacturing process if we are focused on Throughput and Quality”. When we fail to deliver uptime & equipment reliability to the production process, we are no longer needed. More importantly, we are perceived as a high cost silo”.

Let’s change Maintenance to a positive cost silo!
Dave Reiber – Business Solutions Architect

I’m looking forward to co-presenting a webinar on Thursday, Dec. 3rd on how the cloud can deliver the benefits of predictive maintenance sooner, and I invite you to join.

Abstract: Asset-intensive industries are becoming increasingly instrumented and connected, motivated by the immediate need to improve efficiency or via interest in the potential of the Internet of Things. These trends enable capture and analysis of volumes of detailed operational and performance data. However, transformation of these data into meaningful insight that help operations, manufacturing, production, maintenance, and procurement make well-informed decisions remains a challenge. Provisioning the IT environment, identification of meaningful asset performance data, development of accurate predictive models, and delivery of asset health reports in real time have depended upon combined and coordinated efforts of IT, data scientists and business intelligence professionals.

Cloud technologies, pre-built analytics, industry- and asset- predictive models, and customizable user interfaces significantly improve the ability for organizations to operationalize the inherent value of asset performance data. Via real-time reporting, lines of business can use this information to assess asset health – historical, current and future – to make better decisions to proactively manage critical assets to help improve asset availability, reduce maintenance costs, and improve throughput.

Register here!

Speakers:
Dave Reiber, CMRP, Business Solutions Architect, Perficient
Dan Bigos, Predictive Maintenance Solutions Marketing, IBM Analytics

Join Perficients’ Dave Reiber for a webcast, sponsored by Plant Engineering Magazine, to talk about the fast-changing landscape of Asset Management. The title for the talk is “Predictive Maintenance Strategy Without CMMS.” He will discuss an altered view of what is going on, in the age of “Internet of Things.” Traditional methods become ineffective or inadequate. New data requirements and regulatory mandates force us into a deeper dive into Strategic Analysis.

This webcast takes a closer look at what we will see, and what will be expected of us as Maintenance Professionals. It is important to note, that the Future Maintenance Professional & the Future Maintenance Asset, significantly change the current landscape of how things get done. Webcast will address how a Future Asset should behave, which tools are available now, and what should be expected from our Maintenance Professionals moving forward.

DATE – Thursday, November 19, 2015 at 2 p.m. ET/1 p.m. CT/11 a.m. PT
Register – https://event.webcasts.com/starthere.jsp?ei=1055682