Issue link: https://iconnect007.uberflip.com/i/390433
46 SMT Magazine • October 2014 RA: One of the things that help manufacturers cope with the high rate of change, as you mentioned, is automating the processes in their manufacturing lines. From your perspective, which processes in the electronics manufacturing can most easily be automated? JD: Contract manufacturing is more challenging than a typical OEM in the sense that there is a broad diversity of products, a broad diversity of customers, and a real lack of certainty for the business. For instance, if I am an OEM mak- ing printers, I've got a pretty good understanding of my long-term roadmap, and I can make decisions based on that. For contract manufacturers like Jabil, while we have great customer relationships that span decades, there is that level of uncertainty that at any point in time, the business can go away. The wide range of products that we produce cre- ates a lot more diversity than a typical OEM will have to deal with. So those baseline assumptions do make this a challenge for contract manufacturers to put automation in. It makes more difficult than it would for typical OEMs. That's kind of a foundation. So, starting with that, you kind of like look at where the easiest, best opportunities for automation. And typically, what really make an automated system expensive and complex are product- specific tooling, feeders, and characteristics that really revolve around the geometry of the product. If you work with that assumption, it typical- ly means that that kind of generalized assembly is going to be further down the line in terms of what would make sense to automate. You have to back away and think where the easier oppor- tunities are—things like machine loading and unloading are good applications for robots to- day. We've got a lot of robots installed at Jabil loading and unloading molding machines, and loading CNC tools. From an ROI standpoint, they typically ROI well because we take the ex- pensive capital equipment like a molding ma- chine, and then put in a robot, making it more productive, and bringing more consistency to the process, which improves the yield. Another area where we can get a lot of value is in complex processes: polishing, painting, coating, grinding. These processes, where you do high-dexterity motions—and you need to do them consistently and fast—typically a robot can do a much better job of than a person can. What you end up with is not just labor savings, but the quality and the yield improvements that automation drives. Those are the great ap- plications. Then you work to the specific fastening tasks. Screw driving, for example, is a great application that a robot can do with a high degree of consistency. We can monitor the torque characteristics, therefore we can ensure that there's a bottoming out of the screw at the right torque level. The feedback of screw driv - ers tends to fatigue your arms, so it's hard to do that consistently manually over extended periods of time. That's a very good application for robots. And then finally, things like point-to-point sol- der, high-precision test—again, they are difficult to do manually, and they require high levels of quality because of the precision and the pro- cess requirements. Mechanizing that with some levels of automation typically brings some im- provements in terms of quality. FLeXibLe manuFacturing continues ArTiCLE