Issue link: https://iconnect007.uberflip.com/i/1502623
38 SMT007 MAGAZINE I JULY 2023 product. So, what did we do? We ripped a lot of the robotics out. When you're in manufac- turing, you don't have time to fail; you don't have time to investigate. It's all about, "How can I solve this problem right now?" is lab is different because we don't have that constraint. I don't have production. When we fail, we stop and ask, "Why don't we do it differently? What if we started over?" You can't do that in industry. You only get one shot. In this lab, we have a playground. In production, if I go out and buy a robot or automation, it must be right. at makes me very risk averse. In the lab we're allowed to be risk takers. If it didn't work, that's great, we'll try something different. Maybe we'll cut it down or cut this part out. We will see what worked and what didn't. e lab gives us the flexibility of time, and freedom to take more than one shot. Matties: We hear about Smart factories all the time. It's not a flip of the switch decision to be a Smart factory, but rather tiptoeing in with a Smart process approach. What strategies should PCB manufacturing be looking at in terms of automating a facility? Fundamentally, whether you're making appli- ances, vehicles, or making printed circuit boards, the problems you encounter are the same. e main general limitation is in getting one thing to talk to another and acting on that information. It doesn't really matter what you're making. Our goal in this lab is to find out what the general limitations are. If you have to get to the PCB-specific limitations, then find a PCB man- ufacturing lab where you can fail. If you don't get to fail, you'll be continually risk averse. Matties: Typically, when a company starts to look at robotics, they call in a robotics com- pany that arrives to sell robotics, not nec- essarily to solve a manufacturing problem, if you will. How do you address that? What advice do you give? at sounds like the adage, "Give someone a hammer and everything looks like a nail." When we say we have to automate, we automat- ically think robotics, but you need to look at the