Artificial Neural Network for Pre-Simulation Training of Air Traffic Controller

In this chapter, the four layers neural network model for evaluating correctness and timeliness of decision making by the specialist of air traffic services during the pre-simulation training has been presented. The first layer (input) includes exercises that cadet/listener performs to solve a potential conflict situation; the second layer (hidden) depends physiological characteristics of cadet/listener; the third layer (hidden) takes into account the complexity of the exercise depending on the number of potential conflict situations; the fourth layer (output) is assessment of cadet/listener during performance of exercise. Neural network model also has additional inputs (bias) that including restrictions on calculating parameters. The program “Fusion” of visualization of the state of execution of an exercise by a cadet/listener has been developed. Three types of simulation training exercises for CTR (control zone), TMA (terminal control area), and CTA (control area) with different complexity have been analyzed.

Online Estimation of Terminal Airspace Sector Capacity from ATC Workload

Neural Partial Differentiation (NPD) approach is applied to estimate terminal airspace sector capacity in real-time from the ATC (Air Traffic Controller) dynamical neural model with permissible safe separation and affordable workload. A neural model of a multi-input-single-output (MISO) ATC dynamical system is primarily established and used to estimate parameters from the experimental data using NPD. Since the relative standard deviations of these estimated parameters are lesser, the predicted neural model response is well matched with the intervention of ATC workload. Moreover, the proposed neural network-based approach works well with the experimental data online as it does not require the initial values of model parameters that are unknown in practice.

CLARITY AND PRONUNCIATION OF AB-INITIO AIR TRAFFIC CONTROLLER

The increasing number of aircraft flying around the world has led to the requirement for air traffic controllers to improve their communication skills to face high demand traffic in the future. The paper examines the communication errors in the pilot-controller communication of six ab-initio air traffic controllers during simulation training. More than three hours of conversation were collected and analyzed qualitatively using conversational analysis. The transcribed data yielded a total of 62 instances of communication errors. The data revealed that clarity and pronunciation of ab-initio controllers contributed to problematic communication and reduced the efficiency of the air traffic controllers in communicating. In contrast, pronunciation errors rarely diminished comprehension amongst the controllers and pilots who share a similar first language and are familiar with the use of English in a lingua franca setting. The study also describes other instances of communication errors in pilot-controller communication. The results indicate that ab-initio air traffic controllers need to be proficient in three main areas in pilot controller communication to improve their performance: aviation phraseology, aviation English, and aviation knowledge. The findings suggest that pilots and air traffic controllers should achieve level 4 (operational) in aviation language proficiency test, before proceeding to aviation training that requires them to be proficient in their language skills.

Heading Toward Trusted ATCO-AI Systems: A Literature Review

The current review addresses emerging issues that arise from the creation of safe, beneficial, and trusted artificial intelligence–air traffic controller (AI-ATCO) systems for air traffic management (ATM). These issues concern trust between the human user and automated or AI tools of interest, resilience, safety, and transparency. To tackle these issues, we advocate the development of practical AI ATCO teaming frameworks by bringing together concepts and theories from neuroscience and explainable AI (XAI). By pooling together knowledge from both ATCO and AI perspectives, we seek to establish confidence in AI-enabled technologies for ATCOs. In this review, we present an overview of the extant studies that shed light on the research and development of trusted human-AI systems, and discuss the prospects of extending such works to building better trusted ATCO-AI systems. This paper contains three sections elucidating trust-related human performance, AI and explainable AI (XAI), and human-AI teaming.

Automatic Speech Recognition for Air Traffic Control Communications

A significant fraction of communications between air traffic controllers and pilots is through speech, via radio channels. Automatic transcription of air traffic control (ATC) communications has the potential to improve system safety, operational performance, and conformance monitoring, and to enhance air traffic controller training. We present an automatic speech recognition model tailored to the ATC domain that can transcribe ATC voice to text. The transcribed text is used to extract operational information such as call-sign and runway number. The models are based on recent improvements in machine learning techniques for speech recognition and natural language processing. We evaluate the performance of the model on diverse datasets.

Application of the ANP in the interview phase of air traffic controller candidate selection process

Purpose This paper investigates the interview examination applied during the admission process of air traffic controller (ATCO) candidates. This paper aims to select the most appropriate candidates by minimizing any shortcomings of the process. Design/methodology/approach ATCOs have a very important role in the air traffic system. They are responsible for ensuring the safe, regular and rapid flow of aircraft traffic. They carry out this challenging task by monitoring the aircraft under their responsibility and by giving instructions to the pilots when necessary. So the selection of ATCO candidates is of critical importance. This process is usually conducted through multistage examinations. It is a critical issue to use correct methods in the selection process to identify the most suitable candidates. Besides, the application of subjective examination in a standard way and standardization of criteria can assist in selecting the right candidates. Within this context, the analytic network process (ANP) and the rating method are used. The weights of the selection criteria are calculated by the ANP, and the candidates are evaluated by the rating method according to the defined criteria. Findings 39 candidates were ranked using the ANP, and the rankings obtained by the ANP and the current system were compared. The ANP rankings were also compared with the results of a previous study conducted by analytic hierarchy process (AHP). The results indicate that using the ANP yields consistent results with those obtained from the AHP in terms of the rankings. Research limitations/implications ATCOs are one of the most important operators in air transport system. They undertake critical tasks where even a small error may cause serious consequences. Therefore, selecting the appropriate candidate is an important first step in training qualified staff. The authors believe that this study will contribute to the training of more qualified ATCOs. On the other hand, the approach about the use of ANP method on the interview examination and the standardization of the criteria can provide insight into improving interview process in different area. Originality/value To the best of the authors’ knowledge, this is the first study that presents an ANP-based methodology for the selection process of ATCOs and the interview process.

HUBUNGAN SHIFT KERJA DAN LAMA KERJA DENGAN KELELAHAN PETUGAS AIR TRAFFIC CONTROLLER (ATC) DI TOWER ATC BANDARA SOEKARNO HATTA

<p>Di Bandara Soekarno Hatta dalam sehari frekuensi pesawat yang <em>take off–landing</em> bisa mencapai 1216 pesawat per hari atau sekitar 72 penerbangan per jam. Berdasarkan rekapitulasi data kondisi petugas ATC Di Bandara Soekarno Hatta pada bulan Februari hingga 11 April 2017, terdapat 50 petugas (2,1%) ATC yang mengalami kelelahan (Yoga, 2017). Meskipun persentase kelelahan rendah, akan tetapi risiko terjadinya kecelakaan pesawat tinggi. Adapun tujuan dari penelitian ini yaitu mengetahui hubungan shift kerja dan lama kerja dengan kelelahan petugas ATC di Bandara Soekarno Hatta tahun 2017.</p><p>Metode yang digunakan dalam penelitian ini adalah analitik observasional dengan pendekatan <em>cross sectional </em>yaitu pendekatan, observasi atau pengumpulan data sekaligus pada suatu saat. Sumber data yang digunakan berasal dari data primer dan data sekunder. Cara penetapan sampel pada penelitian ini dengan menggunakan rumus Slovin sebanyak 185 sampel. Teknik pengambilan sampel menggunakan <em>purposive sampling</em>.</p><p>Berdasarkan hasil uji statistik yang telah dilakukan, dapat disimpulkan Sebanyak 29 responden (49,5%) petugas ATC mengalami kelelahan rendah, Petugas ATC bekerja dalam 3 shift yaitu pagi, siang, dan malam, Responden yang bekerja &gt;120 menit sebanyak 53 orang (28,6%), tidak ada hubungan antara <em>shift</em> kerja dengan kelelahan dan Tidak ada hubungan antara lama kerja dengan kelelahan petugas ATC di Tower ATC Bandara Soekarno Hatta.</p>Saran yang diberikan kepada petugas ATC berkaitan dengan pentingnya pemanfaatan istirahat yang optimal seperti peregangan otot, berjalan-jalan di ruang kantor, berbincang-bincang, dan makan-makanan bergizi. Selain itu peningkatan kerja sama antara petugas ATC dan asisten petugas ATC, dan peningkatan fungsi koordinator ATC.