Abstract:Advances in the trusted autonomy of air-traffic management (ATM) systems are currently being pursued to cope with the predicted growth in air-traffic densities in all classes of airspace. Highly automated ATM systems relying on artificial intelligence (AI) algorithms for anomaly detection, pattern identification, accurate inference, and optimal conflict resolution are technically feasible and demonstrably able to take on a wide variety of tasks currently accomplished by humans. However, the opaqueness and inexplicability of most intelligent algorithms restrict the usability of such technology. Consequently, AI-based ATM decision-support systems (DSS) are foreseen to integrate eXplainable AI (XAI) in order to increase interpretability and transparency of the system reasoning and, consequently, build the human operators’ trust in these systems. This research presents a viable solution to implement XAI in ATM DSS, providing explanations that can be appraised and analysed by the human air-traffic control operator (ATCO). The maturity of XAI approaches and their application in ATM operational risk prediction is investigated in this paper, which can support both existing ATM advisory services in uncontrolled airspace (Classes E and F) and also drive the inflation of avoidance volumes in emerging performance-driven autonomy concepts. In particular, aviation occurrences and meteorological databases are exploited to train a machine learning (ML)-based risk-prediction tool capable of real-time situation analysis and operational risk monitoring. The proposed approach is based on the XGBoost library, which is a gradient-boost decision tree algorithm for which post-hoc explanations are produced by SHapley Additive exPlanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME). Results are presented and discussed, and considerations are made on the most promising strategies for evolving the human–machine interactions (HMI) to strengthen the mutual trust between ATCO and systems. The presented approach is not limited only to conventional applications but also suitable for UAS-traffic management (UTM) and other emerging applications.

Real-time decision support for human-machine interaction in digital railway control rooms

Machine Learning Techniques to Predict Reactionary Delays and Other Associated Key Performance Indicators on British Railway Network

Selecting Decision Trees for Power System Security Assessment

Theoretical basics of the self-learning system of intelligent locomotive decision support systems

Analyzing Operator States and the Impact of AI-Enhanced Decision Support in Control Rooms: A Human-in-the-Loop Specialized Reinforcement Learning Framework for Intervention Strategies

Enhancing Control Room Operator Decision Making

Data-driven decision support for rail traffic control: A predictive approach

A novel ML-MCDM-based decision support system for evaluating autonomous vehicle integration scenarios in Geneva's public transportation

Explanation of Machine-Learning Solutions in Air-Traffic Management

Data-driven hierarchical learning and real-time decision-making of equipment scheduling and location assignment in automatic high-density storage systems

Traded Control of Human–Machine Systems for Sequential Decision-Making Based on Reinforcement Learning

Robotic Railway Multi-Sensing and Profiling Unit Based on Artificial Intelligence and Data Fusion

A Human-Centric Model for Sustainable Asset Management in Railway: A Case Study

Research on intelligent fault diagnosis for railway point machines using deep reinforcement learning

Intelligent Decision Support System for Updating Control Plans

What is Interpretable? Using Machine Learning to Design Interpretable Decision-Support Systems

Human–computer interaction-based Decision Support System with Applications in Data Mining

Predicting transitions across macroscopic states for railway systems

Algorithm for MCDM in Intelligent Braking Diagnostics System of Railway Transport

Human Response to an AI-Based Decision Support System: A User Study on the Effects of Accuracy and Bias

Predictive Maintenance for Switch Machine Based on Digital Twins