Nicola Puca

The management of NOTAMs on the flight deck remains a challenge for flight crews, who must process large volumes of dense textual information to identify the notices relevant to their mission. Despite ongoing modernisation efforts, the NOTAM system still relies on a legacy format with limited support for automated filtering and prioritisation, which increases pilot workload and motivates the use of Natural Language Processing (NLP) techniques for information management. This paper presents therefore an NLP framework based on a shared pre-trained distilled BERT (Bidirectional Encoder Representations from Transformers) encoder with two task-specific prediction heads, designed to support NOTAM analysis at different stages of the flight. The first task classifies each NOTAM into one of 20 operational categories to support pre-flight briefing. The second task estimates the operational severity of a NOTAM under three emergency diversion scenarios to support in-flight decision-making within cockpit decision-support systems. To address the lack of annotated data and the imbalance across categories, a dataset of 3000 NOTAMs was built through a human-supervised Large Language Model (LLM) generation workflow seeded with real messages. All samples are annotated for both tasks according to written annotation rules. The proposed framework is based on a DistilBERT encoder which receives fine-tuned through Low-Rank Adapters (LoRA), achieves a macro F1-score of 0.967 on classification and R²; values between 0.82 and 0.87 across the three severity scenarios.

The management of NOTAMs on the flight deck remains a challenge for flight crews, who must process large volumes of dense textual information to identify the notices relevant to their mission. Despite ongoing modernisation efforts, the NOTAM system still relies on a legacy format with limited support for automated filtering and prioritisation, which increases pilot workload and motivates the use of Natural Language Processing techniques for information management. This paper presents an NLP framework based on a shared pre-trained BERT encoder with two task-specific prediction heads, designed to support NOTAM analysis at different stages of the flight. The first task classifies each NOTAM into one of 20 operational categories to support pre-flight briefing. The second task estimates the operational severity of a NOTAM under three emergency diversion scenarios to support in-flight decision-making within cockpit decision-support systems. To address the lack of annotated data and the imbalance across categories, a dataset of 3,000 NOTAMs was built by combining expert annotation with synthetic samples generated under human supervision. The proposed framework, based on a DistilBERT encoder fine-tuned with LoRA, achieves a macro F1-score of 0.97 on classification and R² values at or above 0.80 across the three severity scenarios.

Advanced avionics and automation technologies have significantly transformed cockpit operations, resulting in a gradual reduction of the crew members on-board. Single-pilot operations (SPO) concept is gaining significant attention in the aviation industry due to its potential for cost savings and to cope with the anticipated pilot shortage and the increasing air traffic demand. This paper conducts a scoping literature review on SPOs, serving as an initial step to map the scientific peer-reviewed content on the subject. The survey focuses on three thematic domains, which are, respectively, operations, automation, and the emerging field of digital and cognitive flight assistants. The methodology involved the use of Google Scholar and IEEE Xplore databases. Sources were selected adapting the search criteria to the proposed sub-topics and prioritizing either the most cited and recent contributions. The analysis of the literature reveals a growing body of work in the recent years. This review also highlights interest in the human-centered design for automation solutions which are responsive to cognitive and behavioral states of the pilot. While acknowledging the potential safety and operational challenges associated with SPOs and the pilot-automation cooperation, this work suggests that great research efforts should be made on the human factor and regulatory subjects to pave the way for a feasible and safe implementation of the single-pilot paradigm in commercial aviation.

Space electric propulsion represents a class of power-limited systems that utilize the interaction of electromagnetic fields with ionized inert gas propellants to generate thrust. This technology has emerged as a highly fuel-efficient and sustainable alternative to chemical propulsion systems, particularly for satellite constellations. However, the miniaturization potential of EP systems is impeded by certain limitations, necessitating the exploration of novel architectures. The high-efficiency multistage plasma thruster (HEMP-T) stands as a promising contender for stand-alone missions due to its employment of a cusped magnetic-field topology, which effectively mitigates plasma-wall interactions and enhances overall efficiency even at low thrust levels. Despite the growing interest in HEMP-Ts, there is a dearth of comprehensive and streamlined preliminary design procedures for these systems. Prior research has predominantly focused on extensive numerical analyses, neglecting the development of efficient and accessible design tools. To bridge this gap, this paper presents a novel preliminary design tool derived from integrating established analytical models available in the literature. The proposed design tool also incorporates an iterative procedure that refines geometric properties using a 2D magnetostatic solver. Through the application of this tool, a 4 mN HEMP thruster was analyzed. This finally exhibited a specific impulse of approximately 2000s and a good efficiency level of 23%. Also, the results obtained for a 10 mN application align closely with those achieved by other types of EP thrusters.