By Bjorn Fehrm (pictured) and Henry Tam

August 22, 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 new Part 25 airliner development plan. Source: Leeham Co. Click to see better.

Feasibility studies

We start our dive into the different program phases in Figure 1 with the Feasibility stage. We are pleased that Henry Tam, an experienced Airliner development program manager, has agreed to co-write the series. Here is his background:

Henry Tam contributed to a series of articles on certification of 14 CFR Part 23 aircraft on Leeham News in 2021.  He has worked at both mature and startup aircraft OEMs and was the Head of Program for a regional jet and an electric aircraft program. He is a co-mentor at the Sustainable Aero Lab.

Here are Henry’s thoughts around the Feasibility phase:

Feasibility studies

During the feasibility phase, a company must examine multiple dimensions of a potential product: the market needs, potential product configuration, potential cost, potential enablers such as technologies, and potential return on investment. The word “potential” is repeated multiple times to emphasize that these numbers are preliminary and will be refined over time.

Let’s assume we want to make a next-generation single-aisle aircraft to narrow the scope of this series.  What range, payload, economics, customer experience, etc., are needed?  How big is this market?  Who and what are we competing with?  What are we good at?

Once the team has an idea of what they want to focus on, technical folks can start some design work.  Some can be done parametrically.  Others may require more work.  Sometimes the team needs to speak with suppliers to see what is available or in the pipeline as well.  This is especially important regarding engines. It would also be good for the engineers to check the design against key airworthiness, operational, and airport requirements.

At this point, the team may not have a lot of confidence in some of the systems, but they should have some confidence about things like stability and controllability.  This is an iterative process and requires various teams to work together.

Once the study team has shareable data, the sales and marketing team may bring some of these concepts to customers to assess their interest.  More sensitive topics may be discussed in one-on-one or small group settings.  For more generic questions, the OEM may set up advisory boards or conduct focus groups to discuss and debate the needs.  Sometimes, customers challenge each other. Other times, a few customers, as a group, provide strong feedback to the OEM.

It is always a good idea to get feedback from customers.  A product needs to appeal to multiple customers to reduce risks for the business case. This feedback could then drive requirement updates.

There might be more groundwork if the new product relies on novel materials & processes, new technologies, or non-traditional arrangements. For example, do we need to qualify new composite materials and their manufacturing processes?  Is the complexity of a fly-by-wire (FBW) system, which can control the airliner in a safer way, worth extra effort in design and certification?

Company priorities

Unfortunately, not all projects get a green light.  New products may need to compete with other initiatives or corporate priorities. For instance, a commercial aircraft OEM may want to sell off a product area like business jets if it has both.  The marketing, sales process, production, supply chain, and customer support could be very different. Finance is another consideration. If a conglomerate has multiple viable projects, executives may choose the one with the highest return at a reasonable risk.  This project selection process often aligns with budget or governance cycles.

Keep in mind, risks are important considerations.  If risks are not evaluated and mitigated up front, they could have serious consequences.  As an example, the Safran Silvercrest engine was selected by Dassault and Cessna for their new business jets, the 5X and Citation Hemisphere, respectively. The engine development ran into some roadblocks. Dassault cancelled the 5X and launched the 6X with a Pratt & Whitney engine.  Cessna suspended the Hemisphere development.

Consider a program where there is a 90% chance of successfully delivering a new product – where success for this discussion is defined as meeting cost, time, and product performance – without new enabling technologies. If a new product requires five new enabling technologies, each with a 10% chance of failure (from significant delays, increased costs, or unmet customer requirements, etc.), the chance of a successful program decreases significantly. This is one of the reasons why companies try to derisk, if possible, ahead of time instead of running technology development simultaneously. Delaying a program during execution phases can be expensive.

By the end of the Feasibility phase, there should be a list of market requirements, an initial design of the plane (not just artistic renderings), a list of key technologies required, and a rough business case to justify the next round of investment.  The concept could get a green light for the next phase, be shelved, or require additional work, just like any business proposal.

Speeding up the Feasibility phase

There are potential methods to accelerate these iterations.  For instance, conducting market research, formulating market requirements, and obtaining stakeholder agreement typically require significant time and resources. Emerging tools, such as AI, can process large amounts of historical data and offer recommendations. These technologies may also assist in drafting some of the requirements.  For the recommendations to be reliable, however, the tools should provide clear explanations of the market analysis and the reasoning for each requirement.

We will, in the next Corner, look at some aspects of AI supporting a market analysis.

Optimizing aircraft design is not a new topic.  A few popular aircraft design textbooks offer processes and data for initial sizing, some of which have already been implemented as software tools to streamline analyses. Running these designs through an optimizer can help find improved solutions. Fortunately, this workflow is not new, as one of my classmates in grad school was already working on a multidisciplinary optimization tool for aircraft design two decades ago.

Emerging tools may be used to review hundreds of existing airplanes to help identify a better starting point. Engineers will still need to understand and refine the initial concept to make sure that it is explainable and implementable.

New tools can also help put together a business case.  If the design is relatively traditional, an aircraft OEM with a few programs under its belt can make predictions using datasets from previous programs. The challenge often comes from data, or the lack of it. If a company did not track, as an example, work hours with the right granularity, it would be difficult to fix the dataset after the fact. This type of analysis, often done by humans, could potentially be done more quickly with new tools. Yet, explainability is very important because there are tons of assumptions at this stage.

The ability to join these workflows and datasets together could also be beneficial. Design teams conduct trade-off studies regularly. Some of these cross-functional studies are time-consuming.  The ability to complete trade-off studies quickly, reliably, and coherently could help converge on a design and its business case better, cheaper, and faster.

For start-ups, this is a more challenging problem. These companies do not have in-house data.  They may rely on subject matter experts (SMEs) to help generate estimates. Nevertheless, some of these SMEs may not realize the assumptions and preconditions behind his/her reference point, resulting in biased estimates. Could AI help solve this problem? Perhaps. I could see mature OEMs feeding their data into the machine and using it for some of the workflows previously mentioned.  Yet, it is unlikely that these mature companies would allow other companies to have access to their datasets or AI tools trained using these datasets. Could start-ups use synthetic data to train the AI? At this moment, the use of synthetic data is still a field with active research.