top of page

Experience driven vs. technology driven maintenance

Experience is good, Technology better

Industry 4.0 for maintenance means driving forward the use of new technologies

Executive Summary: Humans have impressive senses and experienced maintenance operators can hear/smell/sense upcoming failures of machines. One of the problems is that human experience is subjective, hard to teach and hard to document - you have an experience driven maintenance department. The technology driven department uses modern technologies (e.g. ultrasound, oil analysis) to gather objective data of assets, which is the basis to define when parts need to be changed. The data is used to improve the system over time and the knowledge is documented for the next generation to come. You know the numbers of an oil sample which is considered "normal" and you know the ultrasound reading which is considered "normal" (to just name a few). You only need to teach the use of the technology, but the know how is "digitized" forever (Technology-Driven-Maintenance).


Full Article

"I can hear when something is wrong with my machines" is something one does very often hear from maintenance operators in factories around the world. And indeed, it is impressive what humans can achieve with their senses alone. Beeing responsible for maintaining machines for a decade or so, maintenance operators build up impressive experience. The system you run is experience driven with all its advantages and disadvantages (more about that later).

Why would you even introduce new technologies to help monitor asset's condition, if the current team has everything under control?

Here are 8 reasons why you should build up capabilities to use the available technologies. I am certainly biased towards ultrasound cause it offers the best value for money, but all other technologies (e.g. vibration monitoring, oil analysis, thermography) are adding a lot of value too.

1.) Humans are subjective

The judgement if a certain noise/smell/shaking is ok or not varies from human to human. Our senses are very complex and we don't derive a number like "45 Decibel is the limit" to hold ourself accountable to a "threshold", nor can we communicate our knowledge fully to others.

2.) Human senses are not made for machine monitoring

Human senses are a great addition to monitor machine conditions. One can see the dirty filter, one can see the oil dripping down. But the "language" of machines is a different one. The most relevant emissions of machines are ones we humans can not hear, feel and see. There is temperature, which can be visualized with infrared cameras. There are problematic frequencies above the range we can hear (ultrasound). There are vibrations which we can not analyze (If you are an expert: Did you try an FFT with your bear hands?). Did you taste an oil sample? You get the idea ;)

3.) Know How Transfer

It is very hard to make the knowledge available for the next generation. You can try to document as much as possible, but to be honest: If your most experienced operator retires/leaves the company, a lot of knowledge leaves as well.

4.) It takes a long time to build up

Building up specific knowledge about your machines health status takes a lot of time, mostly decades. You don't need to build up everything from scratch. You don't need to see a machine fail again and again to connect the "squeaky noise with a potential failure". It is well known how the ultrasound frequencies of a faulty bearing are and it also known what makes up a "healthy oil" (to just name a few).

5.) Data Driven Improvement

Ideally, one changes spare parts exactly before they break. This way, the spare parts are utilized as much as possible. If you have implemented a certain technology, you have data. One can use this data to extend the replacement cycle again and again. It's very hard to remember "how loud it was last time" and maybe, "let it be a bit louder, because we changed too early last time".

6.) Efficiency and Scale

Humans can't monitor hundreds, or even thousand of different assets with there "bare senses". With the help of technology, this can be done in an efficient and cost-effective way. Technologies are impoving as well and the support of Artificial Intelligence makes the analysis process of data more efficient every day. Combine ultrasound with AI and one engineer can analyze thousand of assets per day (this is already possible today).

7.) Competitive advantage

There is a lot of generic knowledge available (see 4.)), but the real competitive advantage is achieved with specific data from your own assets when combined with the generic data available in public. So, the sooner you start to "digitize" your assets health status, the greater your advantage. It's hard to speed up real data over time, so better to start now.

8.) The P-F Curve

The "P-F Courve" describes how early you detect a potential failure. It is well known, that ultrasound is the technology, which can detect potential failures very early. Human senses are more "late stage" (e.g. if you already hear the "rattling noise", the failure is not far away). So technology gives you much more time to plan replacements accordingly.



Rated 0 out of 5 stars.
No ratings yet

Add a rating
bottom of page