Forgive me for asking what may sound like a layman's question, but what drives (aside from cost/capital investment considerations or capacity management) Airbus, Boeing, GE, P&W, RR to outsource the production of certain components and parts vs. keeping production in-house? And once a component is outsourced, what dynamics keep the supplier in the supply chain, aside from pricing concessions?

Appreciate the feedback and looking forward to the community's knowledge on this subject.

They do it because it is very expensive to keep very specialized talent and tooling for a development program that runs only once. For example, FBW systems are outsourced to manufacturers that specialize across the industry for those components—Moog and Parker-Hannifin for two. BAE developed the FBW system for the G500/600. I can’t recall if it was Moog or Parker that did the C-Series. Division of labor, but designed to OEM specs and, often, on a risk-sharing cost basis.

The supplier has to have an FAA Parts Manufacture Authorization to produce the parts.

GF

Investment risk is a huge part of it. Starting a new production line costs billions. Taking all that risk yourself is rather wild, unless you're SpaceX.

Union wages. At least that's how it is at Lockheed. Even though they are making record profits.

Also outsourcing to companies in other nations can help sell planes in those countries.

The OEMs outsource as there is specialty experience at vendors. CFM, Boeing, and Airbus do not make valves for example. A plane needs a bunch of those. (Pratt owns a company that does make them, but they outsource as vendors are often less expensive). Take engines. Boeing, Airbus, and Bombardier couldn't make a modern engine. They need to rely on GE, Pratt, RR, Honeywell, or their joint ventures (CFM, as the Silvercrest shows SAFRAN needs to get more experience).

Mitsubishi was willing to invest a fortune in large scale CFRP technology to make 787 parts (wing). This allowed Boeing to move forward without investing billions more. The downside is Mitsubishi gets to make a profit on the wings. For the 777X, Boeing decided to develop a new process and keep that process of wing production in house.

Or take landing gear. Boeing no longer has an in house team. Last I looked, there are only 5 or so landing gear vendors left: Boeing, Honeywell, Fokker (yes, they're still around, and Fokker landing gear knows their stuff), ugh... there is one more (in Italy, IIRC) and Northrop (the old Douglas team was hired by Northrop). Lockheed is trying to form a team... But if you want good landing gear, you have to outsource.

Same with Avionics. Honeywell, UTX, BAE, Northrop, Garman (lower end mostly), Rockwell Collins, BAE, and Curtiss are the surviving vendors. To be blunk, I wouldn't go with 3 of them. Oh, I keep mentioning Northrop as they are like UTX, Curtiss, and Honeywell; a collection of specialty design shops that cater to the whole industry.

Why would Boeing or Airbus pay to design cable reels (escape devices that reel out a cable as you escape from the cockpit hatch), slides, rafts, life vests, fire extinguishers, galleys (including ovens and coffee pots), brakes, landing gear, engines, nacelles, floor panels, actuators (hydraulic and electric), tires, windows, seals, fasteners (rivits or bolts), the components of wiring (basic wire or connectors, they airframes do assemble wire bundles), insulation, bearings, heaters, smoke detection systems, seats, springs, or catering carts.

Or many of those parts (e.g., landing gear design), the need is so spaced out that the existing shops survive by modifying existing designs to sell for business jets, helicopters, military programs, and other commercial programs. A pure Boeing or Airbus company couldn't survive between launches. For example, UTX (Pratt parent company) makes FADACs for RR engines! Curtiss makes LVDTs and tries to sell subsystems built up. But if you just want an Aerospace qualified LVDT (e.g., what is the real flap position), Boeing isn't going to pay to develop one, they'll but it. They won't design a hydraulic pump. The companies that design and build hydraulic pumps sell 20X the volume to other industries and survive that way. Even MOOG sells a lot more outside of aerospace than they let on; e.g., they used defense technology to sell bottling (as in soft drinks) machinery.

For many products (e.g., Bearings), aerospace is a tiny part of the business. I've been part of *major* product launches were bearing manufacturers were so focused on getting the next F150 contract that there was a delay of a year in the program as the total profit in aerospace was too small that year to divert their R&D team away from the Ford F150 design. I ended up saying 'screw it' and designed my own bearings. They worked, were cheaper, but if there hadn't been a time deadline, the customer would never have accepted Lightsaber designed bearing and only accepted them due to knowing my reputation for looking into the details (e.g., which materials in a bearing have to be harder. Too many one off bearings make the mistake of inducing failure modes by not having the rollers sufficiently harder than the race as rookies wonder why race wear is just accepted...). They qualification... Timken wins many orders as they have off the shelf bearings qualified to extreme conditions. In aerospace, either the part is qualified for more than it sees in duty, or you must pay to test it to qualify it.

By the way, I'm annoyed how Silvercrest is diminishing a qualification failure in the Silvercrest. Failing a qualification test is HUGE. Changing qualification configuration invalidates prior qualification testing unless it is really minor (e.g., silver or gold plating electrical leads doesn't invalidate most other qualification testing). Many of the vendors have billions of dollars of completed qualification testing. Testing that must be completed to compete.

Much of this stems from the politics of WW2 (spread the wealth among states). Many new technologies were invented and the airframes need components. So Ohio or Wisconsin, will make the bearings, Pennsylvania, California, New York, and Ohio make the valves. Much of the small casting is done in Vermont, Wisconsin, Connecticut (UTX among others), or California.

Let's take casting. It is so much cheaper to precision cast a part than machine it from stock. Every precision casting shop keeps its way of doing the work proprietary. For example, only UTX and the Russian vendors are great at casting titanium. Boeing or Airbus couldn't... So they have to outsource.

Even the 737 doesn't need enough parts to keep a factory going year round. For example, I've been on the vendor factory floor where 2 months a year (about six months apart), all they do is make CFM-56 parts. But the rest of the year, the factory must be kept going by other product lines. About a month a year is A350 parts. About 2 months a year are business jet parts. It used to be a month a year was JT8D/MD-80 parts, but no more. The rest of the year is 'odds and ends' defense, aerospace, or high specialty parts that are made in such small batches, it isn't so efficient (or profitable), but it keeps the business going. This factory has only two machines required to make certain A350 parts. If Airbus made the parts themselves, a single machine might be occupied 2 months per year. What manufacturer designs for single point failure? By having other buyers pay for the pair of machines, Airbus pays much less overall. Heck, the factory is making medical parts a month or so per year.

Let's take springs. An aircraft might have a few hundred kg of springs on it. Eh... The value of aerospace springs is nothing compared to medical, automotive, or guns. Yes, guns. I've had to wait months when a Glock spring order went in ahead of my order at a vendor. I've had to wait months for an electrical motor when an Amazon robot order went ahead of mine. Aerospace is tiny compared to other industries. Aerospace survives off the economy of scales gained from other industries.

I end on that. To make a part economically needs economy of scale. It only takes a team of 5 to 8 to design most aerospace components. However, that team needs to keep skills sharp designing. A valve vendor will keep sharp designing for competitors, but also the power industry, medical, defense (e.g., submarine or tanker parts), or say a business jet. Economies of scale to design, build, qualify, and manufacture a part.

Now I've seen Boeing and Airbus qualify their own circuit breakers or other parts when it made economic sense. But usually, it doesn't. Boeing rid itself of the expense of a landing gear team between the 777 and 787 designs. Invest in the R&D with high returns and keep that R&D as secret as possible.

Lightsaber

lightsaber wrote:

The OEMs outsource as there is specialty experience at vendors...

Let's take casting. It is so much cheaper to precision cast a part than machine it from stock. Every precision casting shop keeps its way of doing the work proprietary. For example, only UTX and the Russian vendors are great at casting titanium. Boeing or Airbus couldn't... So they have to outsource.

Really appreciate the information here. Just a quick follow-up, I think there are instances where the OEM and the supplier will both produce the same part... in this case, does the supplier effectively license the technology/manufacturing process back to the OEM? Thinking about the statements by Airbus and Boeing on building out their aftermarket capabilities.

stant62 wrote:

lightsaber wrote:

The OEMs outsource as there is specialty experience at vendors...

Let's take casting. It is so much cheaper to precision cast a part than machine it from stock. Every precision casting shop keeps its way of doing the work proprietary. For example, only UTX and the Russian vendors are great at casting titanium. Boeing or Airbus couldn't... So they have to outsource.

Really appreciate the information here. Just a quick follow-up, I think there are instances where the OEM and the supplier will both produce the same part... in this case, does the supplier effectively license the technology/manufacturing process back to the OEM? Thinking about the statements by Airbus and Boeing on building out their aftermarket capabilities.

When I've seen dual production, it usually is due to:
1) The vendor having limited ability to ramp up production
2) The OEM designed the part, but isn't that efficient making the part

Sometimes a manufacturing license is required by the OEM.

Boeing and Airbus are trying to get more into the aftermarket. But lets take a standard part.

If the part costs $1 to make, the vendor will sell it for $4 to $5.
Airbus or Boing buys it, and then offers it to customers for $10 to $15. Usually the vendor is forced to sell it, by contract, for $15.

All that profit goes to keeping companies afloat, paying for downtime, etc.

Boeing and Airbus want more parts sold through them. But so does Pratt, CFM, Honeywell, etc. What Boeing and Airbus want is being pushed back by vendors. We'll see who wins this economic war.

lightsaber wrote:

Last I looked, there are only 5 or so landing gear vendors left: Boeing, Honeywell, Fokker (yes, they're still around, and Fokker landing gear knows their stuff), ugh... there is one more (in Italy, IIRC) and Northrop (the old Douglas team was hired by Northrop). Lockheed is trying to form a team... But if you want good landing gear, you have to outsource.

UTAS/Goodrich, Heroux-Devtek, and Safran (Messier-Dowty) all come to mind.


I think we're going to see a trend away from outsourcing of major assemblies (wings, fuselages, etc.) in light of the trouble Boeing and others have had with that business model.

gtae07 wrote:

lightsaber wrote:

Last I looked, there are only 5 or so landing gear vendors left: Boeing, Honeywell, Fokker (yes, they're still around, and Fokker landing gear knows their stuff), ugh... there is one more (in Italy, IIRC) and Northrop (the old Douglas team was hired by Northrop). Lockheed is trying to form a team... But if you want good landing gear, you have to outsource.

UTAS/Goodrich, Heroux-Devtek, and Safran (Messier-Dowty) all come to mind.


I think we're going to see a trend away from outsourcing of major assemblies (wings, fuselages, etc.) in light of the trouble Boeing and others have had with that business model.

I never understood outsourcing wings. There is too much intellectual property in a wing IMHO.

As to fuselages, that makes sense to outsource. Stuffing a barrel is for a company willing to hire a bunch of ex-submarine employees (where the technology came from). For whatever reason, the same mindset that allows a great airframer makes for poor barrel stuffing... Just my opinion.

In my opinion, the tail should be built in house too (due to the IP).

OK, a few more landing gear makers, but how many could do commercial gear? We're hitting a lull, I predict a consolidation among gear makers.

Lightsaber

lightsaber wrote:

gtae07 wrote:

lightsaber wrote:

Last I looked, there are only 5 or so landing gear vendors left: Boeing, Honeywell, Fokker (yes, they're still around, and Fokker landing gear knows their stuff), ugh... there is one more (in Italy, IIRC) and Northrop (the old Douglas team was hired by Northrop). Lockheed is trying to form a team... But if you want good landing gear, you have to outsource.

UTAS/Goodrich, Heroux-Devtek, and Safran (Messier-Dowty) all come to mind.


I think we're going to see a trend away from outsourcing of major assemblies (wings, fuselages, etc.) in light of the trouble Boeing and others have had with that business model.

I never understood outsourcing wings. There is too much intellectual property in a wing IMHO.

As to fuselages, that makes sense to outsource. Stuffing a barrel is for a company willing to hire a bunch of ex-submarine employees (where the technology came from). For whatever reason, the same mindset that allows a great airframer makes for poor barrel stuffing... Just my opinion.

In my opinion, the tail should be built in house too (due to the IP).

OK, a few more landing gear makers, but how many could do commercial gear? We're hitting a lull, I predict a consolidation among gear makers.

Lightsaber

To satisfy short term financial goals. Gotta meet Wall Street expectations for the quarter even if it means harming the company in the long term.

I worked at a company that brought in a new management team that never looked more than 3 months into the future. A financial shell game was played to make the balance sheet look good on paper. For the next year, things looked great when only the numbers were taken into account. Behind all that, there was a huge brain drain as talented folks left and core business IP was outsourced. The management team was eventually all fired but they already taken their bonuses and jumped out of the plane with golden parachutes. The company is still around but only as a shell of its former self.

Also need not forget the the ability for one geographic area to have enough skilled workers available to build all parts. By using OEMs more than one company is recruiting and training skilled workers over a range of locations.

Also by using OEMs, the manufacturer does not need to maintain stocks of parts and spares on their shelves which the OEM has (less of the manufacturers money sits on the shelf).

It makes a lot of sense to outsource sub-assemblies if you think about it. Boeing and the others design and assembly the airplane, but the components are built elsewhere. Could Boeing build it all in house, probably, but why do that when companies like Parker, Goodrich, and GE specialize in making the components they provide for the airframe. Parker for example manufactures hydraulic components for may airframers and gets contracted for an actuator that fits in x spot and moves y amount with z force. Then Parker designs and builds one to do that job using their expertise is hydraulic systems.

lightsaber wrote:

gtae07 wrote:

lightsaber wrote:

Last I looked, there are only 5 or so landing gear vendors left: Boeing, Honeywell, Fokker (yes, they're still around, and Fokker landing gear knows their stuff), ugh... there is one more (in Italy, IIRC) and Northrop (the old Douglas team was hired by Northrop). Lockheed is trying to form a team... But if you want good landing gear, you have to outsource.

UTAS/Goodrich, Heroux-Devtek, and Safran (Messier-Dowty) all come to mind.


I think we're going to see a trend away from outsourcing of major assemblies (wings, fuselages, etc.) in light of the trouble Boeing and others have had with that business model.

I never understood outsourcing wings. There is too much intellectual property in a wing IMHO.

As to fuselages, that makes sense to outsource. Stuffing a barrel is for a company willing to hire a bunch of ex-submarine employees (where the technology came from). For whatever reason, the same mindset that allows a great airframer makes for poor barrel stuffing... Just my opinion.

In my opinion, the tail should be built in house too (due to the IP).

OK, a few more landing gear makers, but how many could do commercial gear? We're hitting a lull, I predict a consolidation among gear makers.

Lightsaber

I think Liebherr (in Germany) also produce gear.

Anecdotally, there is the sad story of two excellent glider manufacturers. One was good at making wings, the other could do nice fuselages. They cooperated and produced one of the best gliders available today. Come certification and the wings partner decided not to disclose about the wings what was necessary because he wanted to protect his intellectual property. Since this plane was now uncertifiable, production ceased after only about 30 frames.

Some good answers to your questions above. As noted, it's efficient to allow suppliers to develop expertise, and then customers can utilize that expertise as required.

stant62 wrote:

And once a component is outsourced, what dynamics keep the supplier in the supply chain, aside from pricing concessions?

Pricing is certainly a big one. Aside from that, some major factors are Quality and On-Time Delivery.

You'll be willing to pay a higher price to a company that consistently produces high-quality parts and delivers them on-time. Not only that, but if a supplier has a good internal quality system, and has a low defect and scrap rate, then their costs will be lower, potentially allowing them to bid a lower price to the customer.

As an aerospace supplier, we get audited pretty frequently by our customers for various things, and also maintain our quality system in accordance with AS 9100 and such.

On the delivery side, it's not a good feeling to know that your parts are holding up an entire airplane production line, because you couldn't deliver on-time.

gtae07 wrote:

I think we're going to see a trend away from outsourcing of major assemblies (wings, fuselages, etc.) in light of the trouble Boeing and others have had with that business model.

Who else had major issues with outsourcing ( beyond some capitalist strife like seating the issue? )

Issue is/was located with Boeing. bit of a cultural thing.
The suppliers Boeing had issues with had a reasonable to good interaction history with other airframers.

Wiederling: If you don't mind, please expand on your post.

Which aspects of Boeing's culture do you think lead to difficulties?

Which suppliers do you think have had issues with Boeing, but not other integrators?

bluejuice wrote:

To satisfy short term financial goals. Gotta meet Wall Street expectations for the quarter even if it means harming the company in the long term..

If not quarterly, then what is an appropriate frequency for the firm owners (investors) to review management's performance? Objectively, the short-term matters more than the long-term (see: discounted cash flows). Focusing on the "long-term" is an easy way to breed procrastination.

lightsaber wrote:

The OEMs outsource as there is specialty experience at vendors. CFM, Boeing, and Airbus do not make valves for example. A plane needs a bunch of those. (Pratt owns a company that does make them, but they outsource as vendors are often less expensive). Take engines. Boeing, Airbus, and Bombardier couldn't make a modern engine. They need to rely on GE, Pratt, RR, Honeywell, or their joint ventures (CFM, as the Silvercrest shows SAFRAN needs to get more experience).

I don't necessarily agree that Airbus or Boeing couldn't build a modern engine. At the very least, the OEMs could vertically integrate by acquiring an engine supplier. I just see no economic incentive for them to do so.

Vendors can achieve better economies of scale by marketing products across multiple industries. GE and RR also sell their engine cores to oil & gas, naval, and power generation customers. If I recall, about a third of the CF6 fleet are non-aviation engines. Airbus and Boeing have no presence in those industries and non-aviation sales would likely wither away under the ownership of an airframe OEM.

The same goes for many other suppliers like avionics, cabin fittings, controllers, etc.