Using Design Rule Checks (DRCs) to Find Errors in Wire Harness Designs

Video Transcript

Arik Vrobel We're going to be spending a little bit of time talking about design rule checks, or DRCs as part of the wire harness validation and development process,

Arik Vrobel 0:14
and we'll explore why the DRCs are essential to ensure that designs meet functional and compliance requirements and some of the complexities involved with them. So first of all, you guys are all wire harness domain experts. I don't know how well you can see this graphic here alive, but wire harnesses can be a huge mess.

Arik Vrobel 0:39
But we see here, hard to work with, hard to define and hard to manufacture,

Arik Vrobel 0:48
or they can be a beautiful piece of art and beautiful work of Art.

Arik Vrobel 0:53
So does anybody recognize this?

Arik Vrobel 0:57
This art display.

Arik Vrobel 1:02
It's took this photo a couple months ago at United terminal seven at LAX, so I was happy to see that lax invested in their wire harnesses and felt that they were work of art. They must have ran really good DRCs. So I'll

Arik Vrobel 1:22
start a little bit with my background. My name is Eric probel.

Arik Vrobel 1:27
I've been in the wire harness industry for over 30 years.

Arik Vrobel 1:34
I was the former owner and CEO of LCOM systems and elk, over time, became one of the leading suppliers of aerospace and defense, and in the latter years, became one of the leading suppliers of the new space market segment.

Arik Vrobel 2:00
We had around 500 people

Arik Vrobel 2:05
manufacturing our wire harnesses, and, you know, built a good reputation as an engineering partner to our customers, our customers, our key customers, appreciated that we

Arik Vrobel 2:21
helped them work through design issues that they were experiencing.

Arik Vrobel 2:29
I led a successful sale of LCOM in latter stage of 2021,

Arik Vrobel 2:35
to after Delphi,

Arik Vrobel 2:38
and then after unretiring, I mean, after retiring for a few months, I decided to UN retire and come back into this industry and tackle a problem that I was dealing with at LCOM pretty much every Day of my of my life,

Arik Vrobel 2:59
in my experience

Arik Vrobel 3:02
working with the customers we worked with, and some of them are here in this room.

Arik Vrobel 3:08
Our experience was that nearly 50% of wire harnesses that we were asked to manufacture

Arik Vrobel 3:15
had design related errors, so nearly one out of two had some sort of a design related error that that we felt was avoidable,

Arik Vrobel 3:29
but yet we didn't. Our customer didn't catch it in their design process,

Arik Vrobel 3:36
nor did we catch it in our engineering review process. We typically only found those errors as part of our product realization process, and this is what led to the evolution of cable tech. So I wasn't sure if what I saw represented the industry, and so we took a few polls, and I'll share them with you here. The first one, I don't think you can see this and read it in your experience with wireless design or manufacturing. How often have you encountered design errors due to improper component selection? So we kind of pulled this already and we had

Arik Vrobel 4:20
54%

Arik Vrobel 4:22
six say, quite often, over 30% of the time, over 1/3 of the time. And actually, more than a quarter said it happens very, very frequently, over half the time.

Arik Vrobel 4:33
We then asked them, you know, given that you've experienced this problem,

Arik Vrobel 4:40
which, which is the number one mistake that you typically see? And a great majority said there's bond mistakes or emissions,

Arik Vrobel 4:53
actually, 73%

Arik Vrobel 4:56
and the other to it, the next category, 23% said incorrect.

Arik Vrobel 5:00
Quick wire sizing, okay, which is also a bomb mistake

Arik Vrobel 5:06
zero. Nobody said inadequate protection. And there were some other problems that we found.

Arik Vrobel 5:14
So we said, you know, why is it that such a great number of errors relating to part selection, Bob, omissions, Bond, mistakes. Could it be related to data sheets?

Arik Vrobel 5:27
And we asked our users, do you struggle to create component libraries into your catalogs? We I talked to somebody here who's just getting a new software, and they've got to enter the entire library into their CAD database. And you know what's the problem, right? So why? There's too many parts and not enough time? It takes anywhere from 20 minutes per part to 40 minutes, depending on whether it's a complex connector that takes a lot of time.

Arik Vrobel 6:00
The Audi group said it's just really complicated. There's, you know, the technical data sheets and specs are not that easy to interpret and understand. Data is not standardized, and so it takes some time.

Arik Vrobel 6:13
And then final Paul is, how do you check the design flaws? Again, us, as a manufacturer, we almost always found it in production. But what do our other domain

Arik Vrobel 6:26
experts say? They said, as well, manual checks by the engineering team, right? Either the design engineering team or the manufacturing engineering team checks it and identifies the problems.

Arik Vrobel 6:39
Another group said, my experience, we find it in production. 25% said we find it in manufacturing.

Arik Vrobel 6:50
And a small section said we use automated software analysis tools such as DRCs. 17% successfully use DRCs. Okay, and of course, Xero said our designs don't have errors. So everybody agrees that at some level or another, their wire artist designs had errors.

Arik Vrobel 7:16
So a little bit about understanding what DRCs are.

Arik Vrobel 7:22
You know, DRCs are a critical set of automated tools that are used in a CAD or an E CAD environment to identify and potentially rectify design violations

Arik Vrobel 7:40
on predefined rules. So usually somebody predefines what the DRCs are going to do. It's either your CAD supplier or your engineering department or a combination of both.

Arik Vrobel 7:55
Their purpose, again, is to create an automated checking system to ensure that the design means physical integrity, the components fit correctly, they meet the electrical requirements, and they do not violate any kind of spatial, physical or regulatory controls.

Arik Vrobel 8:22
So the common types of DRCs that we see in wider harnesses, they're either electrical checks, physical checks

Arik Vrobel 8:31
and compliance checks.

Arik Vrobel 8:37
And then you know the role of the DRCs in enhancing the design is to help you, for those that are designing, refining your y of hardness configuration so that you can reduce weights, you can minimize cost, you can optimize the assembly process, and you can make The product more manufacturable, which addresses all of those.

Arik Vrobel 9:06
Okay, so what are the challenges

Arik Vrobel 9:10
that we see in DRC implementation?

Arik Vrobel 9:17
There's quite a few of them.

Arik Vrobel 9:20
You know, one is in

Arik Vrobel 9:22
general, wire harnesses have become more complex. There's more sensors, there's more power, there's more computing.

Arik Vrobel 9:35
Hello,

Arik Vrobel 9:38
ans, the wire harnesses become more complex. I mean, in my experience, we were dealing with connection systems that had hundreds of wires with 3040, 50 different connectors, and the complexities is very high in the software that was used to develop them and engineering skill

Arik Vrobel 9:59
it needed to be.

Arik Vrobel 10:00
Quite deep and have very experienced engineers. There's also, we see an integration issue between the systems. Wiring artists require an integration of the mechanical CAD design tools and the electrical CAD design tools, and not all systems are integrated extremely well.

Arik Vrobel 10:21
Spoke to one user here is using a system that's really a

Arik Vrobel 10:26
separated standalone between the schematic capture and the mechanical design, and so it's very difficult to integrate the two requirements together.

Arik Vrobel 10:41
There's results in a lot of manual process checks. So the tools themselves that the validation tool is not catching all of the potential design errors. The systems don't have the data capability for the design validation. The libraries don't have all of the information.

Arik Vrobel 11:04
There's an issue of scaling. So I mentioned complexity. As something becomes more complex, we've got to scale the DRCs. So if you've got a wire harness that has multiple bundle diameters and has different protective coverings and different layers of coverings that becomes more complicated. Again, it's more challenging to validate and to use the tools for the design wheelchairs. And then we've got the industry that's changing. You got update your DRCs. You got to update your library.

Arik Vrobel 11:38
Parts are changing. Specifications are changing, and the regulatory requirements are changing, and even the environmental requirements that your applications are being required to meet, they could change. So often you're being asked by your program team to use one design in one platform with another platform, right? And you just try to

Arik Vrobel 12:06
copy the same design, but it may have a different environmental

Arik Vrobel 12:11
standard that you've got to meet, and it's difficult to

Arik Vrobel 12:15
analyze and evaluate. And then there's a training and expertise the operators of the CAD systems need to be experts. They need to understand both the CAD system itself as well as the parts and the application. So there's a continuous learning and development of the expertise around design wheelchairs

Arik Vrobel 12:40
in terms of best practices, like, how do we do it? Well,

Arik Vrobel 12:44
so we recommend an early and continuous integration of the DRCs. Don't wait till the design is complete to

Arik Vrobel 12:55
run your DRCs. Use it as early as possible as you're building the desired process

Arik Vrobel 13:03
and throughout the entire design life cycle. So ecos repairs, sometimes the repair results in a different component. We've got to run it through the DRC again.

Arik Vrobel 13:19
You know, rather comprehensive DRCs that cover all aspects of the design

Arik Vrobel 13:27
and that are customized based on your specific project requirements. So if you're investing in a CAD tool and you're looking for good DRC coverage, you need to make sure that the CAD tool can support customized DRCs that you can create and evolve as your project requiring changes.

Arik Vrobel 13:52
We've got this concept of automated versus manual checks,

Arik Vrobel 13:57
right?

Arik Vrobel 13:59
Of course, automated is better. It speeds up the checking process, reduces the reliance on humans and, you know, engineers, to check their work.

Arik Vrobel 14:11
But it doesn't define, you know, it doesn't find everything. It's got some gaps, and so supplement that with manual reviews,

Arik Vrobel 14:22
in critical areas where newest judgment is required.

Arik Vrobel 14:29
Again, you need to make sure that your DRCs and your libraries are regularly updated. Okay, that's really important as standards change. Many of us use IPC 620 as the base workmanship. Some use space addendum, those standards change. So if they change, the DRCs may need to be updated, and you got to have the team that is able to analyze the change, and you re evaluate your DRCs and evaluate what needs to be,

Arik Vrobel 14:59
to be.

Arik Vrobel 15:00
Outdated,

Arik Vrobel 15:01
you should have a collaborative DRC process. So you need to make sure it's not one engineer that's writing the requirements. It's a group of engineers involve the manufacturing, people. You know, in my experience, we see the design and the manufacturing and the procurement and the quality run as separate organizations, we highly recommend that the definition of desirable checks in whatever system you use be done as a collaboration of all of those groups.

Arik Vrobel 15:38
And you need to make sure that there's constant documentation and traceability right. We want to see what DRC checks are finding right and validate them with what really happens on the production floor. I am certain that our customers that were doing business with my previous company, LCOM, were not all aware of the mistakes that we were finding, even though, of course, we would update them that we would find a mistake and that we would ask them for an eco I'm certain that they did not go back to their DRC rule checking and evaluate if that's something That should have been a parameter that the software should have defined, and we highly recommend that that sort of a closed loop system occur, that as we see things filter through manufacturing or found in quality, then they be brought back to the Design Group and the team that's responsible to configure your DRCs in your CAD and ensure that the Right compliance

Arik Vrobel 16:48
is adjusted there.

Arik Vrobel 16:59
So

Arik Vrobel 17:01
our company, cable Tech,

Arik Vrobel 17:04
we have created a design validation application, software based system that

Arik Vrobel 17:15
focuses on validating designs from different engineering systems, and say all of them by just taking an output of the design file and running in through our own independent algorithms for design wheel checks, which is something that continues to evolve. And so our approach to DRCs is

Arik Vrobel 17:43
a bit different than

Arik Vrobel 17:46
the

Arik Vrobel 17:47
engineering systems and the CAD systems, and it complements what they do. It's not in conflict, right? So

Arik Vrobel 17:56
first,

Arik Vrobel 17:58
we create our own algorithms for the desired rule checking

Arik Vrobel 18:04
that can predict

Arik Vrobel 18:06
potential issues, both on use specific cases as well as historical data that we're able to gather.

Arik Vrobel 18:17
They're automated and they're customizable, so we allow our users to write their own scripts and define their own rule based requirements, again, based on the industry standards

Arik Vrobel 18:34
or based on their particular application.

Arik Vrobel 18:37
Our system enables a continuous collaboration new. So as we said before, whether you're the design engineer, the manufacturing engineer,

Arik Vrobel 18:49
the manufacturing team, the quality team, whether you're the OEM that's designing or the contract manufacturing that's manufacturing, you're able to use the same platform and run the same change, and we capture those change and feedback to whoever's in the loop, and we maintain continuous data,

Arik Vrobel 19:10
right and insights about those changes, about what worked and what didn't work and how it was updated. And we do that holistically across industry. So we can improve our insights,

Arik Vrobel 19:26
as I mentioned, we can provide comprehensive reporting and documentation about what was changed, by whom, by what, and by you know, why and how, and is that consistent with our recommendation?

Arik Vrobel 19:41
And then you know, over time, our future developments is to integrate our DRCs directly with the CAD providers, right? So whether you're using a capital or Zoom can or cadonics,

Arik Vrobel 19:57
or some of the dissolved products or.

Arik Vrobel 20:00
You know, PTC,

Arik Vrobel 20:02
we're having ongoing discussion, discussions with them to integrate into their systems and provide

Arik Vrobel 20:11
our additional layer of desirable checking.

Arik Vrobel 20:20
So, you know, the way we see ourselves is sort of an integration

Arik Vrobel 20:26
or interconnection between the various engineering platforms that users use.

Arik Vrobel 20:34
Many of them are here, but it's not exclusive to them. And again, we're having dialog with many of them about integrating our solution over time,

Arik Vrobel 20:45
but we also cover what we call in here as the rest of the world. So in my experience, there's quite a lot of companies that I call the rest of the world that didn't use those

Arik Vrobel 20:58
more sophisticated design tools, and they just said, Let's depict what the wire artist looks like. Let's put the the

Arik Vrobel 21:06
wireless, the connection list on an Excel, of course, the valve, and let the manufacturer figure it out. Because we're not going to get it right in our design process, nor do we have enough time and we can figure it out in production. So even in those cases,

Arik Vrobel 21:24
our application can identify

Arik Vrobel 21:28
errors, a lot of them, mismatches, wrong selections of parts right at the top of our data.

Arik Vrobel 21:41
How do we do this? So we placed a real premium on a focus on the component library.

Arik Vrobel 21:52
We see that as the secret sauce. Okay? And a lot of you guys that are involved in designing realize the challenges of creating the component library, both in both in terms of bandwidth and in terms of just getting the right data into the system.

Arik Vrobel 22:11
And so,

Arik Vrobel 22:13
you know, the component library is really, really critical, as I said. It is the central repository of all of the designs, and it should include specification data, like the electrical characteristics,

Arik Vrobel 22:30
the physical dimensions, but often we don't see that it includes Part relationships,

Arik Vrobel 22:37
okay and compliance data and apart relationships. So something like certain connectors have locking mechanisms that they will only accept certain contacts, right? That has to be well defined in your library for a DRC to work well, right? And so the engineer setting up the system needs to make sure that those relationships are well defined and the compliance data, those are things that later on might be a requirement, like robots or some companies have to comply with conflict materials or even exporting right so having the intelligence around your library that can later on be run by different groups to ensure that we meet compliance to the industry requirements.

Arik Vrobel 23:33
Next, as you set the library up, you need to really consider its impact on the DRCs right, each component family is sort of unique,

Arik Vrobel 23:43
and so when we look at those fit relationships, is the engineer, and I'm assuming the engineers are setting up the libraries. Are they really understanding the spec sheet well enough to ensure how it mates with the other components? Are they looking at things like

Arik Vrobel 24:01
dissimilar metals, right?

Arik Vrobel 24:05
That's sometimes an issue. So there's got to be a,

Arik Vrobel 24:13
a an understanding of, you know, of the application itself,

Arik Vrobel 24:19
and then when you have a well managed library like it's much better. Things work a lot and better, right and speeds up the desired process

Arik Vrobel 24:30
to provide quick access to the approved parts and to tested components. It will reduce errors and iterations by ensuring that only the right parts are used. So remember the statistic the poll that I showed you, 73% of our users said there was an error in the file right and so you can eliminate so many errors by having a well managed library.

Arik Vrobel 24:56
And of course, as standards are changing, then.

Arik Vrobel 25:00
You've got the right data in the system, and you can ensure that it meets compliance with those standards.

Arik Vrobel 25:08
And then the challenge is doing all of that is, you know, even if you've created the right library, are you keeping occurring by looking at obsolescence,

Arik Vrobel 25:21
right

Arik Vrobel 25:23
parts become obsolete. Is anybody going in and updating the system,

Arik Vrobel 25:28
sometimes

Arik Vrobel 25:30
certain components go on a prohibited list, right? Is anybody going in and updating the library so that that doesn't go into the design process?

Arik Vrobel 25:42
Um, you know, be sure that it meets the global standards right again, as those standards are changing, are they referenced in the library?

Arik Vrobel 25:52
And are you continuously updating and integrating new components as they come up, so that the engineers have an understanding of

Arik Vrobel 26:05
available parts that they can use. So it's a continuous process. A library is a living thing, right? You've got to have, you know, continuously spend time to update and look at the outside world and see what's changing.

Arik Vrobel 26:22
And even if you do all of that really well, and I've got several of you guys sitting in the back, so you're not going to see these samples, but what we've seen, where you get your data matters,

Arik Vrobel 26:35
okay, so we have spent,

Arik Vrobel 26:40
you know, probably many years looking at supplier and manufacturer data at this point, and we see that data is not all the same, even when it's for the same parts. So I've shown you here some examples

Arik Vrobel 26:57
that we see between a manufacturer's pet which calls

Arik Vrobel 27:03
out a chrome l material

Arik Vrobel 27:06
and a supplier spec which calls

Arik Vrobel 27:09
out runs material for that same shell.

Arik Vrobel 27:14
Okay, so if you're pulling the information from the supplier website, you're going to set up your system incorrectly.

Arik Vrobel 27:24
I've got a few more examples, and again, some of you are here at the back. So you got to trust me,

Arik Vrobel 27:30
we have parts

Arik Vrobel 27:34
that define the standard. And I'll read what it says. It says connector designations depicted are for reference only and not to be used in Part Number developments,

Arik Vrobel 27:49
okay, but when you look at the supplier,

Arik Vrobel 27:53
this dash 12, right here is on the supplier website.

Arik Vrobel 28:00
What that part's not manufacturable by the manufacturer. So again, where you pull the data really, really matters.

Arik Vrobel 28:12
And another example, right? Same sort of thing, this W reference, it doesn't w1 it doesn't exist on the supplier sheet, but yet these guys tell you, you can buy it.

Arik Vrobel 28:28
So what's the reality?

Arik Vrobel 28:32
It very may well be, in this case, to the supplier, the manufacturer data sheet is wrong.

Arik Vrobel 28:40
Okay, and they've updated

Arik Vrobel 28:43
the supply but never updated their manufacturing machine. So how do you know what's true, right? And the way we do it

Arik Vrobel 28:52
is

Arik Vrobel 28:55
we

Arik Vrobel 28:57
compile data from different sources. So first, we've got engineers that look at spec sheets like anybody else does, and define the parts,

Arik Vrobel 29:12
families and figurations, right? But we've developed software tools to then generate a lot of data at scale for those entire families, so we basically built a huge attribute based library. Now that's not enough. We need to make sure it's correct. So we pull data from suppliers and APIs, and we normalize our own data to see if the parts that we've created in our library actually exist in the marketplace. If

Arik Vrobel 29:45
they don't exist in the marketplace, we don't want them in our library,

Arik Vrobel 29:49
but theoretically they could be manufactured,

Arik Vrobel 29:52
so we have to have a record or a capability to pull those if a manufacturer puts them in the design.

Arik Vrobel 30:01
But that, again, is not enough, right? Because if you buy, if you look at a specification sheet of one component supplier, let's say Glen ear,

Arik Vrobel 30:13
and you look at a similar parts, I'll use their Mighty Mouse family as a as an example, you look at a similar part by ethanol,

Arik Vrobel 30:26
it may be an equivalent part, but they represent their data attributes differently. They call them different names,

Arik Vrobel 30:35
right? And so the library itself, for it to be homogeneous,

Arik Vrobel 30:41
we need to transform the attributes to a common language. And that's really what we've gotten, is we're transforming to one normalized language. So when somebody calls and, you know, inside diameter, it's represented the same between the different manufacturers. So ultimately, when you're looking for a certain component in your design, you're not locked into one manufacturer. You know what another manufacturer components would be, and does it meet the requirements? Because the parts can meet the same spec, but not work. In my world, we used the mil spec. It's called v3, 8999,

Arik Vrobel 31:25
and an F and o connector and a T connector meet the same spec, but have variations in their form,

Arik Vrobel 31:34
right? And so in some applications, they may not, they may not work the same.

Arik Vrobel 31:40
And so those are the things that we capture. And we capture a data model that supports for fit in function and enables the integration of components and ultimately the understanding the alternates.

Arik Vrobel 31:56
As I said, our data mapping is defined by engineering domain experts,

Arik Vrobel 32:04
and then it's scaled using software tools. And that's how we can get to a both accurate library and and a deep one.

Arik Vrobel 32:18
We do that by creating sophisticated part generators

Arik Vrobel 32:25
that enable us to scale and ensure the quality of our data.

Arik Vrobel 32:31
We complement it with industry and government specifications

Arik Vrobel 32:37
that will ensure the compliance and the design in manufacturing so the industry specifications are integrated into our library data,

Arik Vrobel 32:47
and then what we've done is we've actually integrated it with real time supply chain information,

Arik Vrobel 32:56
so we've tied in via APIs to over 20 different distributors of electro mechanical components and interconnect components, and through what process you can have an understanding of whether The components that are selected are obsolete, are they available in stock? Because they could be not available, but still be active.

Arik Vrobel 33:28
What is the best lead time for those parts, and what's the price? The best pricing,

Arik Vrobel 33:34
right? So as you're designing, you have this data, and if you know that's that your system needs to meet

Arik Vrobel 33:43
$100

Arik Vrobel 33:44
the wire awareness needs to meet $100 and your bond cost $120

Arik Vrobel 33:48
you've got a problem that happened to me quite often. As a manufacturer, we had a target price my customer, I'm $100 but there was a component that was 1880 80 bucks, right? So how are we supposed to do that.

Arik Vrobel 34:02
This guy's laughing, and the rep that once resident,

Arik Vrobel 34:08
how are we supposed to do that? So we bring in the data and the knowledge early on in the process. And for those companies that are manufacturers, that are not designers, you get to see that on the front end of your quoting or your contract review process, right? We this all takes a minute to write.

Arik Vrobel 34:35
So the features of our software is, as I already said, we read user data from any design system. So whether you're using a zucchini or you're using a cadonics or capital, we'll read the output data of those systems.

Arik Vrobel 34:56
We are developing the largest component data.

Arik Vrobel 35:00
Database that exists in the industry, and it's optimized for this domain. And by large, I mean, we expect to have a million parts in our library by the end of this year.

Arik Vrobel 35:15
And a million parts, that's a lot, right? If you ask any of the systems that are out there. You know, if they give you 1000 or 2000 some some say they'll give you 10,000 parts. That's a lot. So our goal is to have the largest part library. Within two years, it will be a 10 million, which at that point encompasses most of the parts in our domain.

Arik Vrobel 35:41
I've already said, we integrate the engineering with supply chain so it keeps our data active. And when a part becomes obsolete,

Arik Vrobel 35:51
we immediately know it's become obsolete because we're integrated to the supply chain, or there's a stop ship

Arik Vrobel 35:59
on a certain component, we immediately know that there's a stop ship, and we can notify our users,

Arik Vrobel 36:06
and then, over time, we're going to use our machine learning algorithms to help you make better decisions about the selection so to understand what works well In the environment and to make recommendations about what works well, whether it's the specific industry or within your company, and based on your designers

Arik Vrobel 36:32
our system, this is a snapshot hard to see right now, very, very easy to use. You don't need to be an engineer. You could put your lowest skilled administrator person to upload

Arik Vrobel 36:46
a design file and and within one minute, you will get these insights, if you'd like to see it. We have demos going on at the show

Arik Vrobel 36:57
left.

Arik Vrobel 37:01
And then the ultimate goal under is to accelerate, accelerate time to market,

Arik Vrobel 37:09
you know, minimize the design time. So outside of doing it faster having less engineering resources required to design. So if it takes, you know, eight hours to design a complicated harness, we should be able to cut it in half,

Arik Vrobel 37:25
increase the product quality. So often, what I what I witnessed, is because mistakes were made in the design process, and by the time we found those mistakes was in production, many lives have passed, and at that point,

Arik Vrobel 37:42
there's a panic of getting the product out there. And so there are, there could be compromises about solving that problem, you know, patch it right, and that can result in latent defects. And ultimately, this will reduce cost. You

Arik Vrobel 38:04
so as a case study, we worked with a main

Arik Vrobel 38:10
industrial manufacturing equipment company that was working on a design for a high precision industrial machine.

Arik Vrobel 38:21
Had, I think, around 200 wires. Their challenge was to ensure that the wire harness could handle both the power and the complex signal pathways that they required for the control and operation on the machine, and while doing so, in maintaining ease of manufacturing right, and not only that, servicing and reprofits Later on, they use the DRCs to validate the specifications of each of the components against our part library, so that they can ensure compatibility and durability under the industrial conditions that they had to meet either checks included verifying current thermal ratings and the physical robustness of the connectors and wires and whether They even intermittent.

Arik Vrobel 39:18
And we found several

Arik Vrobel 39:21
component veg issues that could have had led to months of delays. There were five critical design errors and more than a dozen warnings that we identified.

Arik Vrobel 39:38
So what do we see in the future, right? And we're focused on design wheel checks. What does the future look like? First of all, we see advanced integration with AI machine learning, like learning from mistakes, analyzing past data, analyzing trends, analyzing.

Arik Vrobel 40:00
In

Arik Vrobel 40:04
the components themselves and how they're changing, and using those tools to improve the design process,

Arik Vrobel 40:14
we see the seamless integration between E cat and MCAT. So again, making those tools work better in covering for the gaps in those tools,

Arik Vrobel 40:25
customization and flexibility, like dialing in the tools the DRCs, to what the users really need and what they use. I spoke to a

Arik Vrobel 40:39
CAD company up in the hall a little while ago, and he said, we've got two rider DRCs, but our users turn them off

Arik Vrobel 40:48
right because they make all these noises that they don't really understand or want. It's kind of like a warning in your car. You just don't pay attention to it. So we want you to have the right DRCs that work for you and create value, and the ones that don't work shouldn't even be there as alerts, and we can really focus on that,

Arik Vrobel 41:13
integrate that with regulatory compliance and standardization. So understand what's happening in the world, understand what our customers are requiring, and ensuring that those are integrative.

Arik Vrobel 41:27
And lastly, the use of decentralization in the process right cloud based solutions that are secure, that secure the data, but that enable collaboration. A lot of our customers have designers working for multiple sites, so working remotely, and again, we've got definitely gaps between the OEMs and the manufacturers. We should use the same platform and see the changes or the issues that are real tight.