John Dulchinos has been called a visionary. The VP of digital manufacturing for Jabil Inc. has more than 30 years of experience in developing digital solutions for manufacturing, including sensory robotics, additive manufacturing, and artificial intelligence. He attained his Master of Science degree in mechanical engineering and robotics in 1985, and he continues to lead in the key areas of robotics and 3-D printing. Always looking out on the horizon to see what's possible, he has a knack for anticipating the next direction of the industry and pioneering its formation. Dulchinos discussed his career with PMM correspondent Lisa Jo Lupo.
What got you interested in robotics?
Dulchinos: As I looked around, robotics was one of the industries where I thought mechanical engineers could have some exciting growth opportunities. My first job out of college was working for an injection molding company called Nypro in robotics. From there, I went to [Omron] Adept [Technologies]. I then joined Jabil five years ago to help them sort out what they wanted to do in automation, which was considered strategic in the long-term direction of the company. About a year into my job, my boss asked me to help define a strategy for additive manufacturing ... because we felt that, in the long run, it would have a very substantial impact on manufacturing. The 3-D printing industry has been around 30-35 years, almost exclusively focused on prototyping and new product development. But our focus is: How do we recast it to make true functional production parts?
Jabil does injection molding and 3-D printing. How do you decide which one to use?
Dulchinos: Of the things that Jabil produces, there may be only a few percent that are candidates for 3-D printing vs. injection molding. We first look at volume, which is really a big driver and, to me, probably the most significant determining point: If the volumes are greater than 50,000 units, it's generally rare that we would use 3-D printing; if volumes are in the 10,000 or 15,000 [range], they'd potentially be a candidate. The second step is the material; injection molding has a much broader palette of material today than 3-D printing. The third is the kind of finish and tolerance needed. 3-D printing doesn't produce at the same level of resolution, so if something has very fine detail, it will more than likely get injection molded.
How do 3-D printing design principles differ from those of traditional manufacturing?
Dulchinos: With every manufacturing process, injection molding included, there are constraints. To design a part for molding, you have to put draft angles on it, there are minimum wall thicknesses, there are undercuts you have to be careful of. What's really powerful about 3-D printing is that those [constraints] are much, much less, so you get much more design freedom. ... You can create some extremely complex geometry that essentially costs you no more to produce than simple geometry. It can allow you to take four or five parts you had to produce separately [in molding] and do them all in a single printing. For example, we took a 39-piece assembly and reduced it down to two parts.
Will 3-D printing replace injection molding?
Dulchinos: The way to think about it is the cost trade-off. Let's say you have a $100,000 mold. If you make one part, the mold cost you $100,000; at two parts, it cost $50,000; 10 parts, $10,000. There's a sliding scale, so when you get to millions of parts, the cost of the mold is essentially pennies or tenths of pennies per part. With 3-D printing, the material cost and machine depreciation are more per part, but there's no mold cost. So, let's say it costs $5 for that same part on a 3-D printer; at 10 parts the cost of each is still $5 — and when you get out to millions of units, it's still $5 apiece. But over the last couple years, we've seen a 10-times improvement in the unit volume where 3-D printing breaks even with molding. Every year it's getting faster and cheaper, so in three to five years, maybe we're looking at low hundreds of thousands of units, and maybe in 10 years, low millions of units in a break-even point. But it's not identical to injection molding, so it's not a perfect substitution.
What is the future for 3-D printing in the plastics industry?
Dulchinos: Today most engineers are not really ready to design for 3-D printing; they're still schooled on the limitations of manufacturing processes like molding. The way it will change is that as more designers come out of school and embrace the capabilities of 3-D design, we'll start looking at new applications from a system view: Let's design this part in the optimal shape. That will allow us to really take advantage of what 3-D printing can do. So, every year 3-D printing will fit a broader and broader set of applications.
How can processors get comfortable with 3-D printing?
Dulchinos: We're teaching our existing designers how to design for additive, what the capabilities are, what to think about an additive manufacturing solution vs. injection molding; we're teaching our engineers how to take advantage of the tool. One of the things that is really helpful for us is that we have a range of 3-D printers. We distribute the low-cost 3-D printers around our company so our engineers can get familiar with the technology and start to use it in their day-to-day jobs. Through that familiarity, they'll learn how to manage what 3-D printing offers, and a lot of that learning translates to more sophisticated, industrial printers. It's an iterative, long-term education process, but the good thing is the designers who have more recently come out of school have been exposed to the technology, so they're more ready to take advantage of its flexibility.
At Adept, you managed a strategic shift from making traditional industrial robots to mobile robots. Can you tell us about this?
Dulchinos: Adept was the largest U.S. robotics company; we were really the first to integrate sensing in the robot. Robots will become ultimately powerful for manufacturers when they can have the kind of sight, feel and tactile touch that people have that allow them to do fine, delicate things and have dexterity of motion. When I was at Adept, robots also were fixed, so we developed very flexible, mobile robots that could move around a factory. That helped for material-handling tasks; it really started to open up applications in warehouses, machine tending and logistics.
What are the benefits of digital manufacturing for the plastics industry?
Dulchinos: Ultimately, the combination of 3-D printing and automation will allow much more flexible, more distributed, localized manufacturing. What's great about 3-D printing is that I can create a CAD drawing today, send it to a 3-D printer and print the part. There's no lag involved, no expense to create a mold, so it's very scalable; you can be efficient in an operation with only a few printers. That will allow us to have smaller factories closer to the ultimate demand, producing customized products, and getting them to customers faster. When you want to produce in a region, all you need is 3-D printing capacity and raw materials.
Throughout your career you've been on the leading edge of digital manufacturing. What do you see as your primary contribution?
Dulchinos: If we are successful at what we're doing, I think we'll create a new way to manufacture parts. What's exciting about it is that it will be faster, more responsive to customers, and allow customers to have better and more targeted products — and it will do it in a more environmentally friendly and sustainable way with a lot less cost of shipping and transportation. What we're hoping to accomplish is to re-create manufacturing in a more digital approach that allows it to be more efficient and serve customers bett
