Leeham News and Analysis
There's more to real news than a news release.
Bjorn’s Corner: Faster aircraft development. Part 2.
By Bjorn Fehrm
August 8, 2025, ©. Leeham News: We do a series about ideas on how the long development times for large airliners can be shortened. New projects talk about cutting development time and reaching certification and production faster than previous projects.
The series will discuss the typical development cycles for an FAA Part 25 aircraft, called a transport category aircraft, and what different ideas there are to reduce the development times.
We will use the Gantt plan in Figure 1 as a base for our discussions.
Figure 1. A generic Part 25 airliner development plan. Source: Leeham Co. Click to see better.
The development program
Before we discuss how to make the traditional airliner development program in a shorter time, let’s look at what phases there are and what will be done in each phase. We also include the rough number of people involved at the aircraft OEM and their suppliers in each phase in Figure 1.
Feasibility phase
Before the actual development of a new airliner starts, there is a continuous ongoing phase at established airliner OEMs called the Feasibility phase. An OEM conducts continuous studies in its current and potential future market segments. These studies are conducted by the OEM’s Future Airplanes group.
The feasibility phase examines interesting market trends and segment sizes, matching them with new technologies and configurations to identify market opportunities. As the studies show an opportunity for a new airliner project, a feasibility project gathers pace and will eventually pass over to a Conceptual design phase.
Conceptual design
The Conceptual design phase can be from a year to several years in length. Typically, it transforms the continuous research in the Feasibility phase to a more focused analysis of a few design candidates and how well these respond to the market segment needs that the airliner project shall satisfy.
The OEM’s Future Airplanes group is now complemented by Line R&D personnel and representatives from Sales and Marketing. The dialogue with key technology providers, especially for the engines, intensifies.
Preliminary design
After the Conceptual design, we proceed to the Preliminary design. Now key suppliers are chosen, and the performance and cost data for the airliner are developed so that performance guarantees and prices can be established. These are presented to the reference airline group and to potential launch customers.
Once the information is solid enough and the customer is ready to order the aircraft, an authority to offer and program launch is initiated. In the preliminary design phase, the team has expanded to hundreds of specialists in different disciplines.
Detailed design
After launch, the preliminary data undergoes various refinement steps, culminating in a design freeze. The freeze now allows for the large work of detailed design to begin. A modern airliner is made of about four million parts. Some of these are standard parts, such as rivets/fasteners, cables/tubes, and connectors, but most are custom designs.
Now, thousands of designers start the job of designing all structural parts in CAD, make stress and other simulations, and then enter all the design data for parts and systems in the PLM (Product Lifecycle Management) system. Subsuppliers start their design work to adapt existing system components or design new ones.
Prototype manyfacturing
Once the detailed design has converged to a release for the production stage, the parts for the aircraft prototypes are ordered based on the produced design data.
The assembly of the prototypes engages hundreds of people from mixed engineering and factory teams.
Testing and Certification
The testing of system components isolated and in rigs is going on as soon as their design are established. The certification work is continuous, spanning from conceptual design to various stages. Here, the timeline focuses on the testing and certification activities performed around the prototype aircraft, as well as their systems in the system rigs.
Production preparation, Production
As detailed design produces the initial designs on parts, they are analyzed for serial production suitability by production preparation engineers.
Final design data is issued to the supply chain, and the production parts are ordered when the testing of systems and prototypes has shown the designs to be ready for production.
Gradually, the serial production of the aircraft starts. The total throughflow times for serial production parts are in excess of two years.
Entry into service, Fleets support
Finally, the training and service departments prepare the hundreds of manuals and databases that must be onboarded with the aircraft at Entry Into Service, EIS.
We will go through these phases, examining the work done, the tools used, and the certification work and interaction that need to be completed. Then we will discuss the possibilities for accelerating the different phases and what’s needed for such acceleration.
Very interesting stuff to follow in this series- thank you, Björn.
Historically, airplane development took 4 years – less for derivatives, and a few months longer for new models. This was the case for 727, 737, 747, 757 and 767 (concurrently) and the 777, and many of the competing products from McDonnell Douglas. It was my clear impression that project management and awesome communication and coordination were at the heart of that record.
In 2002, I had a very clarifying conversation with a very smart, well-informed, hard-working financial analyst. He asked, “Do you really need 40 people in a coordination meeting?” I thought a minute and said, yeah, that sounds about right. He said, “I really don’t know, but GE has [several hundred] propulsion engineers, and Boeing has [about 2/3 as many] propulsion engineers, and maybe Boeing doesn’t need that many.” He said that, in his world the next word after “coordination” was “cost,” and those coordination costs are not really needed with “mature” products like airplanes. [Another long story there, of course.]
Then he said, “that’s what markets do well – manage coordination costs.” In that moment, I was as baffled by his world view, as he was of mine. I said, look; let’s build a new airplane with your market-based view for mature products. If you’re right, you [and people who look like you] will be very happy. If I’m right, we will all be very unhappy.
Two years later, Boeing launched the 787, under the ShareValue business model.
Now, 21 years later, I have to think that question is settled.
The point of this long story, is really a restatement of Scott Hamilton’s review today, of Ortberg’s first year. After one year, we are still not sure whether Ortberg is committed to an engineering problem-solving culture, or is Ortberg still embedded in the Shareholder Value cost-cutting world view?
Bjorn characterizes the Detailed Design phase as frozen. It isn’t, really. In the detailed design phase, the program will need to solve 20,000 problems [Mike Delaney’s number; my number was 40,000, but I defer to Delaney on this point.] Solving those problems early, with design changes that engage all stakeholders is the key to success.
The 4-year cycle times were dependent on the problem-solving culture and the degree of coordination and communication.
The wildly out-of-control cycle times we’ve seen since the late 90’s are the direct consequence of the weak problem-solving culture that you get when pretend your products are “mature” and you let markets try to solve all your problems.
The last problem solving method airplane developed at Boeing was the X-32. Two airplanes from scratch. The first flew in 3.5 years and the second (an airplane that hovers) completed flight test at 4.5 years. It was all completed under the live fire drill that was the merger. The engineering team put their heads down and focused on the job at hand. It couldn’t had have gone better from a development standpoint. Competition was winner take all so not exactly the commercial business model but those airplanes were built to a standard of safety that would have amazed the FAA if they were involved.
Stan:
Great perspective. I have no answers but framing the question is a start.
Even the best run, aka A380 had a couple of show stopper issues. Worst was the wiring. Ungh. Other was the tech issue of the wing ribs shimming, how do you foresee or catch that?
The A350 has gone well, but the lead into it was time consuming in the false starts on the A330 Mk 1-IV. And in the end Airbus did the A330Mk ? and while not a runway seller, its rate 4 (slowly moving up to 5 in a few years).
While not direct, Boeing and NG seem to have produced a prototype F-47 in a short time frame. LM did with the F-35 but then turning it into a viable product, still on going.
Are the testing methods and paperwork confirmation really outdated and a major part of the time delays?
“Boeing and NG seem to have produced a prototype F-47 in a short time frame.”
Just curious, how do you know they “have produced a[sic] prototype F-47 in q short time frame”?
Yes. Its a software prototype that ‘flies’ in a simulator only. The giveaway was the background of people who said they ‘flew’ the prototype. As there was no 2 seater and these pilots werent the bests of the best test pilot qualified, its just puffery.
More recently the McDD/Darpa X-36, a subscale demonstrator in tail-less research flying from 1997 was revealed to have been back doing development flying for F-47 program, which is also tail-less
McDD and BAE fighter proposal for the F-35 JSF program was also tail-less. But they didnt make the final two. So a lot of work was done on that program in early 1990s. Interestingly it was the STOVL variant that was considered unworkable and led to the failure
Chuckling at Stan’s comment about the out of control cycle times since the late ‘90’s… and that’s for mostly derivative airplanes, but the real reason is the lack of leadership. Back in the ‘70’s, 80’s, 90’s there were great leaders such as Alan Mulally and Joe Sutter.
One way manufacturers, especially Boeing, speed up the development process is by the use of ‘Exemptions to the Rule’
Meaning the CFR’s have very specific requirements and rules for airplane certification… for everything!
Sometimes meeting these rules requires much more detailed engineering, manufacturing and certification timelines to accomplish to the letter of the law. To skirt this and do it a different way, Boeing (and Airbus) will apply for these Exemptions – perfectly legal – both by the FAA and EASA to work around the law.
The manufacturer however must apply for and meet requirements on approval for an ALOS or Acceptable Level Of Safety – from the regulators for the specific CFR they are trying to skirt. Meaning we will grant this but first prove it’s still safe.
A good example of this is the current Max engine A/I work around until the problem can be solved.
I’m paraphrasing all this because this can all get very detailed.
One only needs to search the FAA website and see the massive amount of Exemptions granted especially to Boeing since the dawn of the jet age. It’s eye opening.
Good job on this series Bjorn!
To design a commercial airlines to certification and making money is no easy task. After preliminary design and selection of a new 12-18% better engine you get a go ahead and release the full design and test organisation chewing up 1$bn/year, normally you use research and development that has been ongoing for 15-30 years before it can be commercially used/certified. To this all manuals, tools, simulators and procedures need to be developed. Hence all waggling before design freeze cost alot, you have sales people meeting all the “right” customers that ask for the sky and the top salespeople on each continent must understand what to promise at what time and cost if at all.
Lots of time the design freeze is years in the making while $bn’s roll away. Hence the chief design engineer (specs, drawings) and chief project engineer (time and $) must get the management to agree to a freeze letting suppliers to work on issued specs, qualifications and 3D models.
@Stan Sorscher, thank you for your review.
Over the last 10-15 years I’ve been involved in 6Sigma / Lean programs. Programs that almost became religions with the usual simple examples & proving tests to silence any doubts. I raised my hand a few times but learned to be cooperative, enthousiastic.
Still I sometimes wonder if the “waste” to be cut out isn’t killing e.g. (my) typical 2 days reflection time;
Sleeping a night over it, sitting back reviewing against the original requirements, informal discussing with seasoned colleagues (in empty offices after 5PM..) checking asumptions..
Often preventing drama later on.
Here are a few elements of a strong problem-solving culture.
1) Emphasis on communication and coordination.
For instance: someone once told me that an engineering community builds trust when engineers tell each other what they need to know, and everyone expects others to them them what they need to know. Said differently – a network of relationships. Another way of saying it – when someone questions your design, you welcome that as a gift, not a criticism. This interaction is almost impossible in a globally dispersed design-manufacturing team.
2) Responding to challenge.
Years ago, Boeing’s core competency – which was very effective by the way – was to pile people and money on a problem until they crushed it. That’s fun to say, but I am not advocating that. A more practical cultural strategy is to make the best plan possible; track when you go off plan; reallocate resources to get back on plan. Reallocating resources means managing risk and getting everyone to accept risk for the overall success of the project. Again, a global supplier network is ill-suited to that strategy. Reallocating resources requires expertise, and authority, as well as time, people, money. If your business model disperses expertise, or compartmentalizes authority or paralyzes decision-making, or hoards resources, you have the wrong business model for complex heavily-engineered products.
3) If you do have a geographically dispersed team, you need to work very hard to integrate and co-locate as much activity as possible into targeted teams who have expertise, authority, and resources. The 787 program was exceptionally good at this.
Interestingly improvements of existing aircraft often do really well compared to all new aircraft, e.g. 737NG vs A320 and A330 vs 787.
New engines, mature system reliability, global MRO support and pilot availability do a lot.
Also for the A330-900 versus 787-9 competition. In the end operators take everything into account. Also free market attacking, one sided trade tarrifs, these days..
https://www.airliners.net/forum/viewtopic.php?t=1352139
Thanks for the link to that fine thread.
keesje:
The on major adder is the maturity of aerodynamics and little if any possible improvements. Tube and Wing with under slung engines.
There may be some improvements with BWB or TBW but even those are not huge (and not proven).
No where to go that gets a jump. CRFP does some but its not proven to be dramatic.
Put a new wing on an A320 or MAX and you match any exotic (or so close as to not make a difference).
So an upgraded aircraft, works. The MAX works, it was a single failure in MCAS 1.0 that was fatally flawed.
MAX matches the A320NEO. A 220 as an all new build is well done, good aircraft, but 10 – 15 % with new tube, new engines and a composite optimized wing.
Some gains in engines still as are composite wings but nothing that knocks it out of the park. RISE purports to be the path forward but its a one off commitment and to props. Someone jumps on it and it fails on its merits or the public aspect, huge costly program that is a loss.
Its not that Airbus or Boeing are conservative, its the downside of being wrong and both have a couple of failures of off the Market (A380) or a compromise Airlifter (A400) that would have been better as a jet (The C-390 has proven to have some chops in that regard).
Its not that A400 was wrong, that is a dilemma area, but longer range jets are better. And yes I am surprised, I did not think the C-390 would be much of a challenge to C-130 but its proving to be and amazing adaptation of an off the shelf power plant in a swept wing body. Original focus was Tanking but it of course also cross filled and will mainstay as a transport. If needed it can fly any long distance route. Not exactly a strategic airlifter but cross fills if desired.
Boeing of course has the huge cost issues of the 787 that were a bit worse than Airbus program mistake on the A380. MAX is fine now, but again the cost is as bad as a failed program.
“MAX matches the A320NEO.”
News flash: Airlines are migrating to aircraft that offers higher capacity, and longer range.
> A321neo overtakes original A320 as Airbus’s most-ordered model
The Max 10 matches the standard A321.
For the LR and XLR versions generally the capacity is much lower as seating is optimised for the longer haul routes, greater pitch etc. Aer Lingus goes from 212 seats to 184 on the LR
Longer range –> longer flying time –> more demand for premium seating.
Nobody needs lie-flat seats for a 2-hour flight.
Take a look at US carriers’ seating charts for transcon versus regular flights on a given aircraft type.
How come you regurgitate TW”s stale talking points? You are reading from the same PP?
Market shifts to premium travel. This is how successful airlines are making money while Spirit is in financial ruin.
“The Max 10 matches the standard A321.”
Please continue to tell BA there’s nothing wrong with their product, “everything is fine”!! 😂 Ignore the market at your own peril.
The MAX 10 can carry upto 230* passengers, only ten more than the MAX 9.
> The Airbus A321neo significantly outsells the Boeing 737 MAX 10. Airbus has secured over 7,000 orders for the A321neo, while Boeing has around 1,200 orders for the MAX 10. This translates to a sales ratio of roughly 7:1 in favor of the A321neo.”
AAL has a subfleet of transcontinental A321 that carries 102* passengers only! Don’t tell me that’s because the aircraft can’t carry more for those routes.
Airlines tailor their aircraft configurations to what their target markets need.
Delta, United & America each has their A321 with “lie-flat” seats and more premium seats, where $$$ is made. This’ll lower the overall number of passengers the aircraft can carry. Pls update yourself.