Electronica 24: Harnessing AI for Digital Transformation in Electronics and Semiconductors

Unlock the potential of AI in manufacturing as I sit down with Augusto Vilarinho from Critical Manufacturing to explore the future of digital transformation in electronics and semiconductors. Augusto shares his expertise on the paradigm shift towards integrated solutions that enhance connectivity and data analysis, offering a comprehensive approach to harnessing AI's full power. You'll gain valuable insights into the strategic necessity of shifting focus from mere capacity expansion to prioritizing efficiency and performance as companies navigate economic challenges.

Join us for a thought-provoking discussion on the convergence of the electronics and semiconductor industries. Discover how a unified supply chain is becoming the norm, challenging companies to adapt and innovate to remain competitive. Augusto and I delve into the strategic benefits of investing in advanced capabilities and efficiency improvements rather than just increasing production capacity. Find out how partnering with experts can offer a distinct advantage in this rapidly evolving landscape.

learn more at https://www.criticalmanufacturing.com/

EMS@C-Level at electronica 2024 was hosted by IPC (https://www.ipc.org/)

EMS@C-Level is hosted by global inspection leaders Koh Young (https://www.kohyoung.com) and Global Electronics Association (https://www.electronics.org)

You can see video versions of all of the EMS@C-Level pods on our YouTube playlist.

Transcript

Speaker 1 0:08

Hello, I'm Philip Stoughton. I am an Electronica 2024 on the IPC booth and I'm joined by Aldo so from Critical Manufacturing. Thanks so much for joining me. I think we're at this fascinating, pivotal phase in digital transformation at the moment where we've just done a huge amount of foundational work but we haven't got across the line and I wonder if AI is one of the things that's pushing us closer to that line of actually having a digital dividend. Tell me where you feel the benefits and the pressures are in the system.

Speaker 2 0:39

That's true At least. It's a subject that everyone is talking about.

Speaker 1 0:43

Yeah, of course it's getting attention and for critical fact.

Speaker 2 0:47

We are not an exception, but we look at it also a little bit from a more holistic perspective. So what does this mean? This means that we have been there since people started to talk about industry 4.0, smart manufacturing, digital transformation, to talk about industry 4.0, smart manufacturing, digital transformation, and there is certainly a path to get to, let's say, advanced capabilities such as AI, and to that subject, we look at it from an execution perspective, basically with the critical manufacturing MES. That is basically.

Speaker 2 1:23

Its core now is basically the execution tracking, the traceability, the quality control and so on, but we have to add two additional variables to get to the AI portion and a very important variable is also the connectivity, yeah, where we also see and I've been seeing in the last year, especially in the electronics industry, of course, more as seeing in the last year, especially in the electronics industry, of course, more as well in the semiconductor industry as this show is also electronics and semiconductor and we have been seeing also a substantial level of investments and evolution in terms of connectivity. So we have the execution portion that is there since many, many years. We have the connectivity and the enhanced connectivity capabilities, not only for collecting data but also for bidirectional communication, of basically using the traceability results or the results that are being connected to feed them back to the machines and to the equipment in a kind of a real-time and intelligent way.

Speaker 2 2:24

But we have a third pillar that is equally important to get to the AI portion, which is the capacity and the capability to absorb and to basically give a context to all the data that in the meantime, we are collecting from all the IoT devices and the activity that we are collecting and many systems and many solutions are not prepared for that.

Speaker 2 2:46

They are prepared for the execution, they are prepared for the connectivity, but you need to have also a place where you basically have the contextualized information from the machine data collection, from the execution portion, which is basically the manufacturing orders, the products, the quality data and so on, and contextualize it. And once this data is contextualized, it can be used in AI Because you have the information and you can have real case and real examples of detecting trends or, basically, with the information that is being collected from multiple sources located in one single place, apply, for example, machine learning algorithms and evolve and grow with utilization of these algorithms to get to a result that then is ingested back in the operations.

Speaker 2 3:37

So we see it from this holistic perspective the execution, the connectivity, but also the data and analysis portion, and one cannot survive without the other.

Speaker 1 3:48

No, if you can do all three, it puts you in a unique position. Do you think? That's where critical manufacturing differs from a lot of the legacy MES systems that we're used to? It's more ready for this non-siloed More holistic approach.

Speaker 2 4:03

Indeed, because since sometime, we are perceiving that this offer should be more than just an MES. Yeah, so we built our name around MES. We are a renowned MES provider, but in the meantime, we have also to consider these other two aspects.

Speaker 1 4:23

The connectivity.

Speaker 2 4:24

that, if you look at it from a siloed perspective, there are connectivity solutions, there are MES solutions, but the combination of both with one single solution, it's an asset right Because you get one single system to configure and to manage. And now, with the capacity or capability that we are delivering under the brand name of Data Platform, we are getting to these data solutions the ability to capture, absorb all this data that is coming from a classical model put it inside the same infrastructure and contextualize it through what we call also a common data model.

Speaker 2 5:02

And this common data model is basically capturing data from sources that are not standardized, that are coming from different legacy machines from our modern machines. Mes, give it a context, transform it in a way that is kind of standardized and made linear, for then applying intelligent algorithms that we are also bringing as part of our offer. And this entire architecture is what is really the difference and what makes Green Common Effecting special a lot.

Speaker 1 5:34

Yeah, I think what's fascinating is, as you say, we have done all that work and you know bringing it all together is the only way we get value from it. A lot of people talk about data, but if that data doesn't deliver insight, it can't deliver value, and we have to get to the value stage. There has to be a digital dividend for particularly EMS companies to invest, because they're on fairly narrow margins. They're not going to invest in a product that isn't going to deliver efficiency, quality performance, all those different things. Are you seeing a willingness to invest in those kind of things, particularly in a time where capex is lower than it has been? My sense is they're keener to invest in performance and efficiency enhancements than they are in capacity, because they don't need more capacity.

Speaker 2 6:25

Indeed and in EMS companies, we see this quite frequently because we saw a few years ago that EMS companies. We see this quite frequently because we saw a few years ago that EMS companies perceived that the digital transformation and the industry 5.0, call it whatever you want was a need. It was required for them to go beyond, and to basically continue to be competitive and basically survive in this digital world. But what we also saw is that because of this, eventually because of what you just mentioned that our margins are low that there is always some issues related with capital investment capital investments versus operational investments, and so on.

Speaker 2 7:14

And maybe some of them.

Speaker 2 7:18

I don't want to be, let's say, unpolite to some organizations, but some of them falling into the trap of creating silos and by basically limiting the investment that was required and is being required. They invested only in a connectivity solution or only in a tracking and tracing solution, or only in a data lake or something to put the data. And then the things need to be combined together and we see that at this point the investments that they did, that were limited, that were reduced because of circumstances right, are now not really bringing to kind of a payback of the return on investment, because there was the need of having a holistic and strategic perspective that things need to be together, because it's very nice to talk about AI these days, but AI requires data, but requires data with quality, with adequate context, and it's not only with the machine information or with the traceability information of a certain product that you can apply AI?

Speaker 1 8:26

Yeah, and then you have the data. You have reliable data, you have contextualized it, you can do confirmation, learning and all those important things on it that deliver the dividend that you need? What does 2025 look like for critical manufacturing? Whatever you've, got coming along in in terms of new iterations, new products and new solutions.

Speaker 2 8:48

It looks bright actually. So we are of course coming to the end of 2024 with last-minute changes and last-minute challenges. Like every company, the outlook looks good at the moment In 2025, we just had a couple of weeks ago our strategic gathering where we basically stopped to think about what is to come.

Speaker 2 9:15

And there are some important strategic decisions that were taken to basically keep our level of competitiveness, keep ourselves, let's say, in the top level, at least from a technology and innovation perspective, and, of course, with this it's a paradigm we will have a successful year of 2025, for sure.

Speaker 2 9:36

Another important aspect relates to what we have been doing already since some time is focusing, focusing in the segments that we consider strategic for us and basically electronics and semiconductors, focusing in the segments that we consider strategic for us, and basically electronics and semiconductors since, again, we are in a combined event, electronics and semiconductors are strategic segments for us and we will keep creating functionality features, out-of-the-box capabilities that only require to be configured specifically for these market sectors.

Speaker 2 10:08

I think what's exciting is those market segments are getting closer and we're starting to see a little bit of overlap, so for those players that are in there the opportunity to have someone that's got expertise in both sectors is really great and especially because if you go back to the supply chain of electronics by adding semiconductor, you have the semiconductor front-end facilities, the semiconductor back-end facilities that are near to an electronics assembly house and then the electronics assembly houses with their safety lines and processes.

Speaker 2 10:45

Nowadays, in many occasions, this is seen as one single supply chain, so no more isolated silos and so on, and if you look at these, sub-segments that are a little bit more.

Speaker 1 10:59

I would dare to say advanced in terms of technology.

Speaker 2 11:03

The other ones are really perceiving that they need to catch up with that to remain equally competitive.

Speaker 1 11:09

So that's where we are at the moment and where we see the growth, yeah, and I think, as I say, there are good opportunities there when you're not investing in capacity.

Speaker 2 11:19

It's important to invest in capability efficiency performance and working with companies like Critical Manufacturing allows you to do that.

Speaker 1 11:26

Thank you so much for your time, augusto. It's a pleasure to talk to you, always a pleasure.