Basics of Artificial Intelligence:
Science fiction narratives based on fear of a dystopian future have narrated various fictional contents, where Intelligent robots take over the human race and enslave all of humanity. Though the technology is not quite at the superhuman level of robots as portrayed in the science fiction stories, artificial intelligence tools have already had a significant impact on our day to day lives. May it be a simple google search where search results displayed as per RankBrain, walking into your office where your identity is recognized by image / facial recognition software, interaction with a chatbot where Natural Language Processing (NLP) is at work or an email spam filter. Artificial Intelligence (AI) is already assisting the humans whether known or unknowingly.
Artificial intelligence is a branch of computer science that endeavours to replicate human intelligence into machines so that machines can perform specific tasks that require human intelligence or AI can also be defined as the ability of a computer to mimic human intelligence. Artificial intelligence is not a new invention, this subject has been an area of research and education in academics for the last 70 years, and remained in academic laboratories. Now, due to the advancement in the processing power, statistical techniques and tools have supported this field of science to innovate new applications for real-world problem/scenarios and also commercially viable. This article is intended to give a high-level view of artificial intelligence in a simple language for the understanding of the aviation community.
As we have already seen that, AI technologies endeavours to replicate the human intelligence in machines and based on this characteristic, AI can be classified into two types, Type 1 is based on capability (to the extent by which AI is capable to replicate Human Intelligence) & Type 2 is based on functionality (performing simple basic operations to advanced level conscious decisions). Tuning test is the methodology used to determine the capability of artificial intelligence.
Artificial Narrow Intelligence (ANI), Artificial General Intelligence (AGI) and Artificial Super Intelligence (ASI) are the types of AI-based on capability. Reactive machines, limited memory machines, the theory of mind, and self-aware AI are classifications based on functionality. The table below shows the types of AI, its attributes and real-world examples or applications.
The moment a search for any content related to AI is made on the web, it will be flooded with various technical jargons such as machine learning, deep learning, data science, computer vision, Robot Process Automation (RPA) and so on, let’s understand in a structured way.
Artificial Intelligence is like a brain (manmade thinking power) for perceiving, learning, language understanding, reasoning, and solving problems. Data is like the energy that supports the activity of the brain through data science, machine learning techniques and algorithms to make decisions. Robotic process automation is like hands, i.e., the tools to execute the decisions of AI. The tools can be soft tools (e.g. software, chatbots) or hard tools (e.g. robots, intelligent machines). Today's AI is probabilistic and based on machine learning, which means it is no longer based on conventional “if - then - else if” rules. Since, digitalization and AI rely on data, both the quantity and the quality of data are of critical importance to support the successful implementation of these technologies. Computer vision to help computers “see” and understand the content of digital images such as photographs and videos.
How does the AI Impact Aviation Industry?
Artificial intelligence and digitalization is impacting the commercial aviation industry in a huge way and has the potential to disrupt the aviation industry and its entire value chain. AI capabilities can build new services, new business capabilities and models or can enhance the existing airline business capabilities such as customer touchpoints, operational, support, management processes. The business establishment believes that investing in AI strategically can position them better in the marketplace. A few use cases where AI got its place are revenue management, targeted advertising, ticketing kiosks, passenger identification, assisting customers, baggage screening, maintenance prediction, predictive analytics, pattern recognition, auto-scheduling, customer feedback etc. This use cases will not only provide better customer experience but also at the same time will reduce the operational and labour cost for airlines.
It can be very well concluded that AI and digitalization are game-changers in aviation, as in every other sector. The use of AI and digitalization technologies allows for more safety, adaptability, optimization, efficiency, capacity and support to all aviation stakeholders.
AI-led growth can impact the economy in three ways, creating new virtual workforce, enhancing the existing skills and tasks (as a complement system) and driving innovations. The AI in the aviation market is likely to reach USD 2,222.5 million by 2025, at a CAGR of 46.65%. However, the limited number of experts in AI is restraining the growth of AI in the aviation market.
The aviation industry is a safety-sensitive / critical industry, as such application of AI in the industry should have due consideration on safety-sensitive and non-safety sensitive aspects. There is considerable work ongoing in industry and academia in the area of “Provable AI” and “Trustable AI” which is required to bring AI into the safety-critical operational domains. These new technologies will contribute to the future of aviation and will redefine the core competencies of aviation professionals.
No comments:
Post a Comment