Information technology today constitutes a fundamental and often critical component of all activities and functions within the air transport ecosystem. From aircraft and air traffic control to the management of airlines and airports, technology has become an integral part of daily operations and strategic planning. Advanced information systems enable businesses to remain competitive by adopting innovative business models while optimizing resource utilization and enhancing overall efficiency and effectiveness. Simultaneously, technology plays a central role in ensuring the safety and reliability of travel, thereby enabling commercial air transport to remain the safest mode of passenger and cargo transportation worldwide.
Throughout the history of flight, many significant contemporary technologies widely used across all sectors were initially developed to meet aviation needs. Additionally, any significant innovation from other sectors quickly finds its way into air transport, providing a variety of benefits.
This course examines a wide range of technologies shaping the air transport industry, from communications and networks to big data and artificial intelligence.
Based on recent statistics from 2022 on the aviation industry, the global market size for Artificial Intelligence (AI) in aviation was approximately 728.05 million dollars and is expected to reach $23 billion by 2031. The current and future use of AI in the international aviation sector extends (but is not limited to) optimizing pricing strategies, predicting and preventing maintenance issues, enhancing flight operations, and optimizing air traffic management. AI may also significantly aid airports, improving their operations, boosting security, enhancing passenger services, providing traveller guidance, and offering personalized assistance for seamless travel. Big data analytics and AI modelling can enhance aviation safety. Such data can be collected from aircraft sensors, flight data recorders, weather inputs, airport traffic data, and passenger characteristics. When accurate and precise data is provided, AI can identify potential safety gaps/concerns to mitigate risks and improve safety standards. Advanced AI technologies can lead to predictive maintenance solutions for aircraft to forecast potential failures in a timely manner and avoid severe consequences. Aircraft components can be monitored in real-time and analysed by advanced AI technologies, addressing/avoiding unforeseen maintenance (an issue that caused 7% of flight delays in 2023).
The modelling of aviation processes represents a complex and regulated field where safety is of utmost importance. There are smart AI systems worth mentioning (e.g., “Predix” launched by General Electric). Predix enhances the support of the General Aviation Fleet by facilitating accurate management of large volumes of data concerning engines and strengthening diagnosis capabilities. Finally, AI can handle Sentiment Analysis, which is a crucial process for airlines as it provides a clear picture of customer-related issues. The Automated Neural Intelligence Mechanism (ANIE) is an emotion analysis tool based on AI, which is used to examine feedback channels (e.g., social media, customer review websites, surveys).
The explanation of AI should be broadly utilized within aviation systems. AI systems must not only explain how they arrive at their recommendations but also be monitored and evaluated for their performance.
Upon successful completion of the course, students will be able to:
- Understand the fundamental concepts and principles of Artificial Intelligence (AI).
- Understand the basic principles of Data Mining for the discovery of useful information.
- Understand the key steps in developing Machine Learning models.
- Understand the core principles, architecture, and theoretical-mathematical framework of Feedforward Artificial Neural Networks (ANNs).
- Develop classification and regression ANNs using Feedforward Artificial Neural Networks with applications in air transport management.
- Understand performance evaluation metrics for ANN models.
- Appreciate the necessity of Avoiding Overfitting and Memorization.
- Grasp the concept of Generalization and its importance in machine learning model evaluation.
- Understand the principles of other well-known Machine Learning algorithms.
- Solve problems arising from class imbalance.
- Understand Deep Learning Algorithms.
- Develop robust Machine Learning models for air transport management applications.