By Bjorn Fehrm and Henry Tam

August 29, 2025, ©. Leeham News: We do a series about ideas on how the long development times for large airliners can be shortened. New project talks 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 new Part 25 airliner development plan. Source: Leeham Co. Click to see better.

Market research in Feasibility studies

Market research is an important part of Feasibility studies, and it continues in different forms into conceptual and preliminary design.

In the feasibility phase, the study’s focus is on the overall market development for airliners, the trends that can be read out of history, what airlines are saying they are going to do or want to do, and the size and development of the market segments that the studies focus on.

There are a number of organizations that do market research for air transport. Many are serious and reliable, some are not.

I’m saying this as we (Leeham Consultancy) have been contracted several times to work with different OEMs in the early phases of projects, and we have found market research being bought by OEMs that can’t be classified as serious or professional.

Too often, the market research that is delivered is what the OEM wants to hear, not a realistic projection of the market. This is especially common when a project is investor-driven. To impress investors, the project contracts market research companies that deliver rosy prospects of the segment in question, not the reality. We will examine how excesses in forecasts for a segment enabled by new technology can be reality-checked.

First, let’s check the performance of an AI system. We have used ChatGPT Pro, which is a well-known AI agent. It did not perform well. There may be others that perform better.

AI agent performance

We put the following question to ChatGPT: What is the market for a new generation airliner to replace A321 and 737 MAX?

and got the following answer:

The market for a new generation airliner to replace the A321 and 737 MAX is driven by the need for more fuel efficiency, advanced aerodynamics, and modern cabin features. Key factors include increasing passenger demand for regional and medium-haul flights, environmental regulations pushing for greener aircraft, and airlines seeking operational cost savings.

Market Overview

• Demand: Growing demand for narrow-body aircraft, especially in regionally dense markets like Asia, Europe, and the Americas.
• Major Players:

o          Airbus, with the A321XLR and A321neoLR variants.

o          Boeing, with the 737 MAX series and potential future models.

• Opportunity: A new generation aircraft could target fuel efficiency improvements, lower emissions, and increased range.

Future Trends & Considerations

• Fuel Efficiency & Sustainability: Innovation in lightweight materials, engine technology (such as geared turbofans), and alternative fuels.
• Cabin Modernization: Enhanced passenger comfort and connectivity.
• Technological Advances: Integration of AI, predictive maintenance, and noise reduction.

 While there are few outright faults in the response, it’s disappointingly unspecific and weak. There is a total lack of any quantifying data (segment size, development, trends). In fact, there is nothing in the response that a program manager wouldn’t have known. He gets no wiser on what to do after asking ChatGPT.

We asked some further questions to get some more specific answers, same weak responses. Nothing useful.

We don’t go further with trying AI to give us facts and data. It suffices to say that the AI agents didn’t deliver anything useful. This is surprising, as there is a lot of very good public data on the subject. Airbus, Boeing, and Embraer all publish yearly global market forecasts for their segments, and these are very thorough. This year, they start with 2025 and forecast the development of the segments for 20 years.

All three are slightly biased to boost the segments where they participate and dominate, but this is easily checked by cross-referencing all three individually and with an independent research organization like Japanese JADC. Large consultancies like Oliver Wyman also conduct research, with a bit wider coverage, such as into the MRO market.

As said, there are companies that produce subpar studies, projecting sky-high market figures for segments that turn out to be empty balloons. Several recent Air Transport segments come to mind, like Personal jets, Electric aircraft, eVTOLs, etc.

Reality checks

There is a straightforward way to check such overoptimistic claims. Most of these are made in new segments, like smaller alternative propulsion segments.

The large OEMs’ research typically doesn’t cover smaller aircraft with fewer than 50 seats, but JADC starts at 15-seat turboprops. However, the reality check involves examining the history of actual deliveries for the segment or its closest equivalent.

Sources are aircraft database companies that track orders, deliveries, storage, scrapping, etc. The more credible cost money, and the most respected, Cirium, quite a lot. But for a $15 billion development of a new single-aisle airliner, such considerations are not a problem.

But there is a high-quality free source for aircraft up to 50-seat turboprops. It’s GAMA (General Aviation Manufacturers Association). Let’s use GAMA for a few reality checks:

• Alternative propulsion 9, 19, 30, or 50 seater aircraft projects. GAMA tracks the deliveries of the majority of OEMs in these segments (some that are not tracked are found below). Here is a table of the total worldwide deliveries for turboprops up to 19 seats from 2019 to 2024. I have deliberately made a 2024 sum under Daher, Pilatus, and Piper small turboprops, as many of these go to private use. Not a single company that has been in its segment for many years has passed 100, except for Pilatus, in one year. Compare these figures to the market claims by upstart companies, and one often wonders where these have got their market data from.

• Joby’s prospect for its 2021 SPAC merger said Booz Allen Hamilton estimates the UAM market by 2030 to be $500 billion, and Joby expects their production in 2025 to be around 200 to 400 eVTOLs. The 2021 acid test is to look at the helicopter production, as a VTOL is the alternative to a helicopter service (give or take, we are talking an acid test here). The world’s total civil helicopter production in 2021 was 860 units, and those that have performance closest to eVTOLs, the piston types were 181.

There are more examples, but these show the principle. There is a reason the nine-seat Britten Norman Islander is no longer being produced; the 19-seat Do228 was produced as one item during 2024, with De Havilland to deliver one or two Twin Otter during 2025 (no in 2024). The 48-seat ATR 42 (one of the few, in production 50-seat turboprops) collected five orders last year, and four were delivered.

Summary

Market research is an important area for a new air transport project. In the Feasibility phase, the focus is on market segment sizes, history, and the projections forward, the expected segment changes, and competition. As the program progresses, the market research gets more detailed with what airline type is the focus, what features are important, whether focus groups make sense, etc.

We will cover such activities as we progress through the program. The aim here was to check if AI could help with market research in this phase. ChatGPT can’t. The response was disappointing, especially as there is no lack of public information to amass and train on for the topic.