Data is the new oil. All of you might have heard this metaphor in one or another context. But, what does this actually mean. How can we create value by collecting some ones and zeros? In the context of this podcast, we have already spoken with a variety of guests about collecting data using modern sensors. Today, we want to focus on how we can create value from this data. To discuss this, we invited a wonderful guest. Together with her team, she is an expert in understanding machine data and turning this knowledge into value for enterprises – especially the ones from the industrial sector. Let’s welcome Lisa Erdmann. Lisa is Managing Director of Industrial Analytics. 

Guest: Lisa Erdmann, Managing Director, Industrial Analytics
Moderator: Thomas Reinhardt, Director Corporate Campaigns ＆ Customer Communication, Infineon

Date of publication: 15 July 2024

***

In Europe allone more than 116 million chillers are being operated which consume more than 58 TWh per year. Potentially, up to 30% of this consumption could be saved - which we can also confirm from some first own projects that we have conducted in the past.

***

Moderator:

Hi everyone. Welcome to a new episode of the #MakeIoTwork podcast. My name is Thomas Reinhardt, I am your host, and I am excited to have the opportunity sharing this podcast with all of you.

Data is the new oil. All of you might have heard this metaphor in one or another context. But, what does this actually mean. How can we create value by collecting some ones and zeros?
In the context of this podcast, I've already spoken with a variety of guests about collecting data using modern sensors. Today, we want to focus on how we can create value from this data. To discuss this, I invited a wonderful guest. Together with her team, she is an expert in understanding machine data and turning this knowledge into value for enterprises – especially the ones from the industrial sector.

Let’s welcome Lisa Erdmann. Lisa is Managing Director of Industrial Analytics. Welcome Lisa – what a pleasure having you.

Guest:

Thanks for inviting me. I’m really excited to be here.

Moderator:

You are working for Industrial Analytics - can you tell us a bit more about the company and what you are doing?

Guest:

Industrial Analytics was acquired almost 2 years ago in 2022 by Infineon. We are a small venture of engineers and software developers that bring more software capabilities to Infineon and finally help to serve our customers with more advanced, more complete solutions that go well beyond the "pure hardware". 

In recent years, we have recognized that our customers see lots of potential in applying analytics - and I think this is true across all segments but esp. also for the industry where we are in. To be more specific, every industrial company along the value chain is currently dealing with topics like reducing energy consumption and the CO2 footprint, to tackle challenges like a shortage of qualified engineers or to simply use machinery in a more sustainable, more efficient way.

And this is where we from Industrial Analytics want to support our customers and partners with.

Moderator:

Okay, this sounds thrilling. To make it more tangible, could you elaborate a bit more what kind of solutions you are building.

Guest:

Our objective is to build solutions that can precisely determine the health status of industrial equipment - e.g. when it's breaking down or when the next maintenance is needed; but we are also building solutions for energy management or performance optimization.

All of these are pretty hot topics where we see a lot of potential. If I may give an example: we are currently building a solution to optimize the energy consumption of chillers - cooling equipment used for air conditioners and industrial processes. Only in Europe more than 116 million chillers are being operated which consume more than 58 TWh per year. Potentially, up to 30% of this consumption could be saved - which we can also confirm from some first own projects that we have conducted.

Moderator:

That is a remarkable number. And sounds like a significant lever in driving decarbonization together with our customers and partners. Can you tell us a bit more about your approach to optimize energy consumption or to predict equipment failure?

Guest:

Our company slogan is "engineering knowledge build into AI technology" and that summarizes pretty well what we are doing: we combine the knowledge of our experienced engineers with AI approaches. More concretely, we are not simply using statistical models which assume huge amount of data which finally allow a kind of pattern recognition. Instead, we use an approach that we call "hybrid AI". First, we build physics-based models that incorporate the ontology of components, equipment or even processes and know how these work in a healthy state. Then, we apply machine learning and feed in data - this allows us to work with much less data and to get much faster to results. Furthermore, our results are much more reliable - in expert terms: we reduce so called "false positives"- which means that are models know much better if the equipment is really failing or if the anomaly is still in the range of a usual operation.

Moderator:

Okay, let’s make it more concrete again. Can you share another example on what kind of use cases or processes you are currently working on?

Guest:

Yes sure. Recently, we have started at our own facilities at Infineon to explore more use cases. For example, we have started at one facility to equip a specific type of pump with our service to predict breakdowns 48 hours before they actually occur. This brings various benefits: for example, the reduction of costly downtimes, the shift from a fixed maintenance schedule to a more flexible one, but also less scrap in production.

Moderator:

Great, so you are actually applying the AI-based solutions in our very own manufacturing capacities. That is great. And, what is it that you have learned from these kinds of projects? What are some of the biggest challenges your customers face when making use of AI?

Guest:

Even if everyone is talking about AI or trends like predictive or prescriptive maintenance, I have the feeling that it's still a way to go - and there are various reasons for it. For example, the equipment is being operated for decades and hence they lack certain data points since sensors are missing. Or it's not easy to access the respective data since they are distributed across various IT and OT systems. But this is for sure solvable - the major challenge is to generate insights from the data so that they get actionable and create the value, that one was looking for. And this is exactly where we are focusing on - to make sense of data based on our machine knowledge. More concretely, we are offering dedicated analytics building blocks that can be integrated wherever customers or partners what them to be. We call this product "Analytics Core Element" - in short "ACE".

Moderator:

Looking forward, how do you see AI and predictive maintenance evolving, and what role will Industrial Analytics play in shaping this future?

Guest:

Our vision is to not only analyze mechanical data, which is commonly used for analysis, but also to harness the potential of data from the electrical world, such as data derived from semiconductors which, by now, are present in every machine or component anyway. These topics are currently still in the research phase but offer tremendous potential. In this way, complexity and costs could be significantly reduced, thus making predictive maintenance accessible for entirely new applications or industries, for example, for a promising market like renewables, including PV, solar, wind power, or EV charging.

Moderator:

Thank you so much, Lisa, for this exciting journey into the power of data in AI.

This brings us to the end of this episode.

Dear listeners: If you want to learn more about us and our world of IoT, visit our website www.infineon.io. If you're currently listening to us on Spotify or Apple Podcast, we'd love for you to subscribe to our podcast and leave a little review. Now it only remains for me to wish you a good time. Take care and see you next time.