Replacing Third Parties With A ‘Skunkworks’?

The original “Skunk Works,” which gave credence to what is now little more than a business idiom, referred to a division of Lockheed Martin responsible exceptionally future-thinking aerospace design. Today, some companies, like Primus Green Energy, have mixed that philosophy with one called “dogfooding.”

Mnet 30665 Pge Plant

The original “Skunk Works,” which gave credence to what is now little more than a business idiom, referred to a division of Lockheed Martin responsible exceptionally future-thinking aerospace design. It began in the 1940’s, and operated separate from Lockheed’s primary business, working on new designs, particularly for secret operations from the CIA and the like.

Today, the term “skunkworks” refers to a smaller group within an organization that is, generally, a research and development (R&D) operation that does not necessarily work within the confines of what will make the business the most money. Today, one of the most famous examples is Google X, which has developed self-driving cars, the Google Glass augmented reality glasses and now wants to deliver internet to poor countries via balloons.

In a difficult economic time, many of these types of projects, particularly among manufacturers, are shuttered in favor of working on projects that directly relate to improving the bottom line. And much of the capacity — whether it’s manufacturing itself or R&D — is shifted away from internal parties to third parties, who, in the past, have offered greater flexibility or lower costs.

Primus Green Energy (PGE) is a relatively new company working to develop alternative drop-in fuels produced from available domestic sources, such as natural gas and biomass. And since the company opened up shop in 2007, they’ve operated in a different way from almost all of its competitors, and different from many manufacturers. They operate, essentially, as one large skunkworks, eschewing third parties in favor of engineering in-house, manufacturing parts with their own people, all before implementing the technology in a demonstration plant of their own.

Typically, a company would focus on only one element of that pipeline, perhaps building the fuel conversion plants, like an oil company might, or simply building the reactors and selling them to fuel processors. This editor’s colleagues have a running joke, or observation, that you know a manufacturer’s product is good when they use it in their own plant—when a conveyor maker uses those same conveyors to move goods, for example.

Some call it “dogfooding.” Either way, PGE seems to be exemplifying the benefits of doing most everything in-house, wildly cutting costs and doing more to protect their intellectual property, which truly does separate them from the competition.

PGE’s process, essentially takes a variety of feedstocks — whether that’s biomass, natural gas or coal — and turns it into drop-in gasoline. In between, there’s a conversion from the original feedstock into syngas, which is catalyzed into methanol, and then again into dimethyl ether. A third reactor turns that into gasoline, while a fourth cleans it up and condenses it into the final product: 93 octane, high-quality gasoline that could be pumped straight into a car.

That process, a series of chemical reactions that are well-understood, is not necessarily PGE’s primary advantage, however. George Boyajian, the vice president of business development, says, “With the advent of new catalysts, new control systems and better understanding of chemistry, we’ve made an inventive step here, and we’ve made a much more continuous and integrated process that reduces the number of parts and complexity of the system. We’re able to convert more of the molecules into gasoline than [our competitors] are.”

To make that possible, they had to take control of the entire process — from building the reactors themselves to the installations in a demonstration plant. Other companies, Boyajian says, would outsource that engineering to a different firm, and building the plant to yet another, but he and his colleagues instead worked to keep that entire lifecycle within their company, hiring for engineering, fabrication and construction teams alike. They have welders to build vessels, a machining center that can cut parts, chemistry labs to evaluate catalysts and an assembly team to put everything together.

Sam Golan, PGE’s chief operating officer, says there are three main benefits to this kind of in-house, or skunkworks, type of development: IP protection, costs and speed of development. He says, “If you took our idea and you outsourced most of it, as most of the companies are doing, you definitely would not be able to finish in 14 months. We know many companies who are doing kind of what we do, but it took them three years and at three times the cost. We knew we had to build a very strong team that could take on the whole cycle.”

He offers, as a counterexample, Boeing’s recent troubles in outsourcing an unprecedented amount of the engineering and manufacturing for its 787 Dreamliner. There have been year-long delays and billions in cost over-runs because the company simply couldn’t keep a handle on all the various parties. Golan says that when there’s a known budget and strict constraints, efficiency is everything. And amid those constraints, he argues, in-house is the only way to go.

By keeping everything in one building, so to speak, PGE can have daily technology meetings to optimize previous designs and look over data from test implementations. They can pivot on new data or a new catalyst in a day, drastically reducing the cycle time of evaluation and innovation. In-house engineering keeps intellectual property (IP) closer to the chest, as the saying goes. Some of PGE’s executives have had issues with IP theft in the past, but Golan says their plan is all about “elimination,” reducing the vectors of theft. He says, “We just don’t give it to the outside, so we will not probably face it sometime in the future.”

Boyajian adds: “When you’re doing a first-of-a-kind (product), you want to understand everything about it. If you give it to somebody else, you really don’t know what it is you’ve got when it’s delivered.”

The company also worked hard to become an ASME-certified code shop, which allows them to build any type and size of high-pressure vessel. Before that development, they purchased a vessel that cost $100,000 and took nearly five months to arrive. Once the code shop was online, they were able to build the next vessel for less than $30,000 and in less than two weeks.

That incredibly quick turnaround, in conjunction with the company’s engineering team, allows them to reduce cost and delivery time by more than 70 percent. This means they can iterate on new designs, from design to fabrication to being implemented in a demonstration plant, in a matter of weeks rather than months, or even a year. It’s an incredible asset to a company that is working in a constrained atmosphere.

This is not to say that Boyajian, Golan and the rest of PGE think third parties make for bad business, or are unnecessary. Boyajian says, “Where we can buy services that bolt-on to our core, we do that.” And that core consists of, essentially, the single inventive step within the four-step process that helps them convert more molecules of feedstock into gasoline. “To focus on everything and you focus on nothing.”

Boyajian seems to argue that a similar company structure could work well for other companies, even if they’re in a more traditional manufacturing situation. By doing all the work themselves, from engineering to implementation, he says the company knows exactly what they’re looking for in a new design, and exactly what they’re getting when it’s finished. It’s a strong break from the typical model seen now in manufacturing, but perhaps one that merits more thought.

Golan adds: “When you outsource, you lose it. When you send a supplier some characteristic and he plans it, and you get it as a product, then it’s just a black box.” And when it comes to “dogfooding,” it’s best to keep mysteries to a minimum.

More in Industry 4.0