Issue link: https://iconnect007.uberflip.com/i/647109
72 SMT Magazine • March 2016 see that their process is going out of control they can do something about it. Otherwise if you have to wait hours to get a result what's the point of collecting all of this data? Then there are a lot of scraps. So you want to be able to query quickly and get data so you can make de- cisions and for that you're going to have to rely on real big data solutions. Matties: Understanding what data you really need is key, too. You have to look at the right data to make the right decisions. Manor: Yeah, and also you can archive some of the data. Not all of the data needs to be in real- time. There's some queries that somebody can sit in the office and if he has to wait ten minutes for the result, that's fine. But if he's doing some big investigation and trying to drill down but there's some cockpits and some management dashboards that need to be in real-time to see whether he's making bad boards, those things you need to know so you can stop. The idea is that some of this data is really collected on the line level and then maybe the line manager can go with his tablet to the line, see the overall equipment efficiency and the de- fects per millions, get all of these classic metrics, and see he's okay. The manager might get some- thing for the enterprise and the full site and he can see that. Somebody on the corporate side, which might have six different sites, can see a comparison between the site in China and the site in Mexico and he can run some metrics to see who is better. That is probably not needed in real-time. You have to differentiate between who is seeing the data and how well it's aggregated and that will really give different people the ability to intersect with the data and also do it remote- ly. You could sit at home, see your dashboard and see that nine six is down from 9:00 in the morning and say, "Okay, you know, I have to do something about it." You don't have to be physically at the factory and look at the traffic light like we have to do today. Matties: Now in sectors like automotive, big com- panies, the tier ones, can afford to do this right away. They are 4.0. They have to be, right? They have the money, the funds, and the resources to do it. It's the tier two and below who probably can't get to it quite as quickly. Manor: We're sitting here and our message says, "Enable Industry 4.0 today without changing your world." One of our main messages is you don't have to go to a huge IT project with mil- lions of dollars and implement this massive MES system. We might be able to help you get to Industry 4.0 by just bridging in some of those systems with our IoM, getting the data and sup- plying you the data and you do everything else. Tier twos cannot come up with a big project that costs $3 million; it will never fly. No one shut down their whole operation for three months to do a massive upgrade. Matties: And they don't want to go out and buy new equipment. It's not in the forecast. Manor: Exactly, so we have to keep their exist- ing infrastructure, their existing IT systems and slowly start bridging in these elements, making it Industry 4.0 compatible. And that's something we're going to try to focus on. The key message here is we don't want to change your world. We don't want to make you have to buy a lot of new machines. The idea is to focus on data acquisi - tion and the ability to make decisions autono- mously, in addition to the machines, with some clever hardware that can do this bridging. This will be a tier two, tier three enabler. Matties: What sort of gains can they expect from this type of investment? Manor: I can tell you two things. First of all, from a planning perspective, this allows our production plant to be much more efficient. The machines today are all very flexible. They're ac- curate and they're fast. But people don't really leverage the capabilities of the machines that well. The overall equipment efficiency in the industry is probably about 50% and we can do much better. If we look at the airline industry, the planes fly 16–18 hours a day. They have been able to re- ally maximize the usage of the airplanes. If you compare that to our industry, the machines are MEntor graPhICS' orEn Manor IntErvIEw