You know that maintenance and repair costs, downtime, and productivity all have a major impact on the bottom line…and you’re always on the lookout for ways to have a positive impact on the bottom line. The good news is the smart use of predictive technologies can do just that!
What are Predictive Technologies?
Predictive technologies are tools that support PdM (Predictive Maintenance) and CBM (Condition-Based Maintenance) by:
- Providing critical, in-depth data about actual machine health condition
- Looking at equipment performance or deterioration and insights as to when and how equipment is likely to fail
- Discovering and analyzing patterns in data so that past behavior can help predict likely future behavior
And implementing PdM (Predictive Maintenance) at your facility can make a huge difference in factors like machine uptime, maintenance costs, and productivity. But before we go deeper into that, let’s talk about how predictive analysis actually works.
How Predictive Analysis Works
First, data is collected from machines using sensors. The data can then be plotted, have trend analysis performed on it, have advanced analysis performed on it, and extrapolated from to predict future trends.
From that data, set points can be established and when they are exceeded it can trigger maintenance to perform a very affordable equipment minor repairs, rather than deal with a costly catastrophic failure.
When fed to machine learning algorithms and the cloud, this data will identify patterns of behavior that have led to problems in the past. If real-time activity starts to follow one of those problematic patterns, the analysis system will be able to predict the potential outcome and alert appropriate factory personnel. This allows corrective action to be taken to avoid a serious loss of efficiency and any unexpected costs.
What Predictive Technologies Should I Consider?
There are many types of predictive technologies out there, but certain ones are able to provide more useful data. These include tools for:
- Vibration Analysis
- Infrared Thermography
- Motor Testing such as PdMA
Each of these four tools collects a different type of data that, taken together, provides a very clear picture of the condition of critical equipment in your facility.
For example, Motor Testing alone can tell you something is wrong with a winding, but combined with infrared thermography can point to the exact area where the main issue lies. Vibration analysis can indicate an issue with a bearing, while ultrasound can tell you if a bearing is over or under lubricated.
Each of these technologies provide data that is useful on its own, like the ability of ultrasound sensors to detect partial electrical discharge and vibration analysis to detect resonance conditions.
Benefits of using Predictive Technologies
The major benefit of using predictive technologies is the ability to go beyond time-scheduled maintenance to condition-based action. A condition-based approach can predict the likelihood of future failures by applying machine learning and data analytics to reduce asset failures. The reduction in asset failures also reduces both the downtime and repair costs associated with them.
Manufacturers get advance warning of problems, such as potential quality failures and/or unplanned downtime due to machine failure, allowing operators to take corrective action.
Remember that quality failures can translate into significant losses in the product, overhead labor, and time. Predictive analytics can enable your factory to identify the pattern behind quality failures and take corrective action faster. This minimizes impact and reduces the cost associated with waste.
For example, machine learning based on predictive technology data can predict a quality failure will occur in 15 minutes because line speed has dropped–based on the fact that when line speed has dropped in the past, products do not meet quality standards.
Factories are also turning to machine learning and predictive analytics to identify production trends, solve problems faster, and manage resources more efficiently. The ability to identify potential issues early on with predictive analytics enables factories to optimize their process and avoid the costs associated with material waste, high scrap rates, or downtime.
The Smart Way to Implement Predictive Analysis
I have seen a very enthusiastic evangelist “sell” the predictive technologies concept to the Plant Manager, and even in some cases to the stockholders. A common mistake is to “jump too far” in the front end, only to spend significant capital to invest in a major program without showing progressive benefits as the program builds. This can lead to discontent and the entire project being set aside as a poor investment.
There is a smart way to implement predictive analysis at your facility. First, begin with a limited budget and start reasonably small. Establish upper management buy in and include frequent, clear results over a several month period.
After everyone is comfortable with the results, then the Plant Manager (and often the stockholders) become the real driving force to expand the program as they see the cost reductions plant wide. Remember… enthusiasm is contagious.
Let the predictive program sell itself and allow those who control the budget to push the enthusiastic demand for a full blown program. The sky’s the limit and long term success is virtually guaranteed when the right people push it!
Applying predictive technologies enables manufacturers like you to identify problems at their very earliest stages so they can be dealt with before issues start to unfold and the cost to address them goes up. And Knower can help you with that, from helping you determine what the most critical equipment in your facility needs PdM, assisting you with selecting and installing the right predictive technologies, and guiding you through the analysis of the data that is produced. Our Reliability Team includes certified vibration analysts, tribologists, ultrasound technicians, and thermographers that offer turn-key condition-based maintenance services.