The development of artificial intelligence has been advancing rapidly in recent years, and the emergence of ChatGPT has pushed the development of artificial intelligence to new heights. When artificial intelligence is combined with the aviation industry, people’s first reaction may be unmanned aerial vehicles. There is still a long way to go to prepare air passengers to board an unmanned aerial vehicle, but artificial intelligence has the potential to completely change the business practices of the aviation industry: aircraft manufacturers and airlines have invested a lot of resources in artificial intelligence technology and actively explored various application scenarios.
Significant impact
There is latest evidence to suggest that multiple industries are investing more funds in adopting artificial intelligence. Global cloud enterprise software company S has released a report stating that its revenue increased by 55% year-on-year in the first half of this year, mainly due to the high demand for artificial intelligence technology in the market. IFS stated that despite the policies of Qualcomm and the market, companies are still investing in artificial intelligence technology. Last year, IS had several transactions and collaborations with aviation companies, including China Airlines and Emirates Airlines.
Tim Clark, the CEO of Emirates Airlines, conveyed a message to the aviation industry about artificial intelligence: “Many people care about what artificial intelligence does and doesn’t do. But if you are a company that relies heavily on process flow and productivity, and such a powerful thing emerges, you must study how to use it to improve work.” In Clark’s view, artificial intelligence will have a significant impact on the aviation industry.
A good example of airlines using artificial intelligence technology is Alaska Airlines. The company took advantage of the reduced flight volume during the epidemic prevention and control period to test some new travel plans. During a 6-month period, Alaska Airlines implemented an artificial intelligence program called Flyways, which tests the original flight, current weather conditions, aircraft weight, and other factors to determine the most efficient route and determine the best route.
It is understood that Alaska Airlines’ artificial intelligence program has tested all possible routes, collected range and fuel usage data, and used this data to adjust the focus of subsequent work, all for real-time planning of the most effective flight routes.
Alaska Airlines pilot and head of corporate development, Pasha Salih, said that aggregating large amounts of data is exactly what machines excel at, and “Flyways may be the most exciting thing I have ever encountered in the field of aviation technology since I can remember.
During the 6-month pilot period, Flyways reduced the average flight time of Alaska Airlines flights by 5 minutes. Although 5 minutes may not seem like much, it is equivalent to saving up to 480000 gallons of aviation fuel, which is a significant benefit for Alaska Airlines as it strives to fulfill its commitment to achieve carbon neutrality by 2040.
It happens that there is a similar case. Etihad Airways will soon provide passengers with another way to book flights – using artificial intelligence technology. This Abu Dhabi based airline has established a partnership with Astra Technologies, which will allow passengers to book their flights through its chat application BOTIM. Thanks to this new technology, Etihad Airways passengers can book flights by simply entering the basic information of their desired flight service, and BOTIM will complete the booking on behalf of the passenger.
People expect artificial intelligence in the aviation industry to have a huge impact, as it can reduce the workload of aviation professionals or enhance their abilities in complex situations, including supporting the work of air traffic controllers, pilots, and cybersecurity officers. People also predict that artificial intelligence will improve security and network resilience by providing new conflict detection, traffic consultation, and solutions. It will better utilize aviation data, make more accurate predictions, improve productivity, and increase the utilization of scarce resources such as airspace and runways.
Change the game rules
Erich Lanjan, founder and CEO of GridRaster, said, “Artificial intelligence has helped manufacturing, repair, and maintenance (MRO) users better collaborate with robots.” GridRaster is a software company specializing in the development of augmented reality technologies, such as augmented reality and virtual reality, which use artificial intelligence and spatial mapping software. The company provides such tools for the aerospace and automotive industries and collaborates with several top defense contractors in the United States.
ChatGPT and other similar language learning artificial intelligence models excel at immediately providing relevant and accurate outputs based on large amounts of information. It can tell people how to build and certify an aircraft, provide recommendations for improving the aerodynamic performance of the fuselage, and even generate maintenance schedules for specific aircraft.
Although ChatGPT and other generative AI models can access all the information publicly available on the Internet, they cannot access the valuable private intellectual property libraries of companies. Allowing AI to access these highly protected information will open up a Xintiandi for AI in the entire industry.
Lan Zhan said that when the aviation industry begins to verticalize this technology and integrate it with its own intellectual property for internal use, the maximum value of generative artificial intelligence models like ChatGPT will be realized. For real applications, some companies will pay for artificial intelligence models like ChatGPT so that you can leverage this massive learning model and train it with proprietary data.
Although ChatGPT is a language learning model that only outputs text, generative artificial intelligence can also create images and 3D models. In the aerospace field, this is particularly useful for generating digital Li Sheng.
Nowadays, aircraft manufacturers and MRO suppliers rely on complex virtual models known as digital twins to simulate aircraft and various subsystems in a digital environment. Engineers can use digital twins to reduce the need for physical construction and testing, thereby minimizing costs and shortening product development time. MRO technicians use digital Li Sheng for predictive maintenance and detect abnormal situations by comparing real sensor data with data generated by data generation. Although digital Li Sheng can help save time and resources, creating them is expensive and time-consuming. However, generative artificial intelligence will soon make the process of building digital twins much faster.
Concerns about the future
The future of artificial intelligence is not whether you use it, but how you use it. However, there are also concerns about the future development of artificial intelligence. One potential concern is that artificial intelligence is developing so rapidly that it may become uncontrollable, and machines may eventually surpass humans and make their own decisions.
In May of this year, the European Aviation Safety Agency released a new Artificial Intelligence Roadmap 2.0, aimed at promoting a people-centered approach and integrating artificial intelligence into the aviation industry. The updated document incorporates the progress made in this field since the release of the first roadmap by the European Aviation Safety Agency in February 2020. According to the European Aviation Safety Agency, the new roadmap “outlines a comprehensive plan for the safe and reliable integration of artificial intelligence into the aviation industry, with a focus on safety, security, human factors, and ethical considerations”. This is part of the organization’s efforts to promote the integration of emerging technologies and aviation industry safety.
It is worth mentioning that two researchers specifically mentioned in a paper prepared for the World Economic Forum that in terms of artificial intelligence regulation, the technology industry should seek inspiration from the aviation industry. They wrote that due to strict regulation and critical safety procedures, the aviation industry has provided valuable experience for artificial intelligence regulation. If the technology industry implements similar policies and creates a learning culture, future disasters related to artificial intelligence can be avoided. Now is the time for us to consider the artificial intelligence industry as a safety critical industry. In the aviation industry, which is a high-risk industry, what changes can be made to make artificial intelligence flourish while protecting humans from possible harm? Legislators and artificial intelligence developers can look at the sky and learn from the aviation industry.
They pointed out that for decades, the aviation industry has improved communication and teamwork in high-pressure situations through the development of regulations, standard operating procedures, and training programs. Instead of waiting for a new AI disaster to strike, we now have the opportunity to learn from past disasters by implementing three reforms: comprehensive training programs, regulation and standardization, and a culture of sharing secure information.
