Training model for air traffic controllers using a multi-agent approach

DOI: 10.31673/2412-4338.2020.016878

Abstract

Air traffic control dispatchers are a key element of the air traffic control system, officials on whose competence and education depends on the safety of civil aviation flights. The article shows that training air traffic controllers is a complex task. The use of intelligent teaching systems is a modern trend towards improving educational activities. One of the main problems of traditional research in the field of intelligent learning systems is the problem of modeling the knowledge and understanding of people who are learning. Provided that the formation of an effective training system for air traffic controllers is an important task, the development of training models in an intelligent training system to improve the effectiveness of the training process is relevant. A powerful tool for solving this problem is the use of an agent-based approach. In practice, when preparing a training plan for air traffic control dispatchers without taking into account the model that describes the formation and loss of the training level, there may be situations when the level of training falls below the minimum level, which means the inability of the air traffic control dispatcher to ensure the safety of civil aviation flights. It is proved that when drawing up a preparation plan, the level for each event must be calculated taking into account the previous activities that create an information base for it and directly affect the effectiveness of its implementation. This makes it possible to increase the level of training without increasing the costs of each training event and the training plan as a whole.
The article has developed a training model for air traffic control dispatchers based on an agent-based approach and can improve training efficiency by controlling the level of knowledge, skills and abilities. The use of this model is advisable in intelligent learning systems - training planning modules. The directions of further research are the synthesis of the method of scheduling classes for air traffic controllers taking into account this model, as well as the procedures for constructing ontologies of academic disciplines.

Keywords: agent-based approach, air traffic control dispatcher, training model, intelligent learning system.

References
1. Vereshchahin, I.I. (2007), Automated synthesis and models of flexible computer professional simulators for general purposes: Author’s thesis, NAN Ukrainy, Kiev, 20 p.
2. Zhdanov, I.S., and Shabalina, O.A. (2005),"Intellektual'noe upravlenie protsessom razrabotki uchebnykh proektov"[Intelligent management of the development of educational projects], Higher professional education in modern Russia: prospects, problems, solutions: mater, int. N.-met. Conf .: within the framework of the International scientific sympos., ded. 140 years old. MSTU MAMI, Section 5, pp. 40–41.
3. Lytvyn, V.V. (2009), "Multyahentni systemy pidtrymky pryiniattia rishen, shcho bazuiutsia na pretsedentakh ta vykorystovuiut adaptyvni ontolohii" [Case-based multi-agent decision support systems using adaptive ontologies],Artificial Intelligence, No. 2, pp. 24–33.
4. Lomakina, M.Ie., Surkova, K.V., and Surkov, K.Iu. (2018), "Zasoby korektsii profesiinoi pidhotovky maibutnikh aviadyspetcheriv" [Correction tools for the training of future air traffic controllers], Scientific Bulletin of the Flight Academy. Series: Pedagogical sciences: collection of scientific papers, Vol. 4, pp. 138–144.
5. Musatova, E.G., Lazarev, A.A., Ponomarev, K.V., Yadrentsev, D.A., Bronnikov S.V. and Khusnullin, N.F. (2016), "A Mathematical Model for the Astronaut Training Scheduling Problem", IFAC-PapersOnLine, Vol. 49, No. 12, pp. 221–225.
6. Gafarov, E.R. (2016), "Graficheskii metod resheniya zadach kombinatornoi optimizatsii" [Graphic method for solving combinatorial optimization problems], Automation and Telemechanics, No. 12, pp. 26–36.
7. Jaein Choi,Matthew J. Realff and Jay H. Lee (2004) Dynamic programming in a heuristically confined state space: a stochastic resource-constrained project scheduling applicationComputers and Chemical Engineering, Vol. 28 (6), pp. 1039–1058.
8. M. Palpant, C. Artigues, and P. Michelon. (2003), "Lssper: Solving the resource–constrained project scheduling problem with large neighbourhood search". Technical Report LIA report 255, Laboratoire d’Informatique d’Avignon, Avignon.
9. Kovalev, M.M. (2004) "Models and methods of scheduling", Lectures, pp. 234–253.
10. Khachaturov, V.R., Veselovskii, V.E., Zlotov A.V. and dr. (2000),"Kombinirovannye metody i algoritmy resheniya zadach diskretnoi optimizatsii bol'shoi razmernosti" [Combined methods and algorithms for solving discrete optimization problems of large dimension], Nauka, Moscow, 360 p.
11. Glover, Fred W. and Kochenberger, Gary A. (2003), "Handbook of metaheuristics", Springer Science & Business Media, 557 p.
12. Pape C. Le (1994),"Implementation of resource constraints in ILOG SCHEDULE: a library for the development of constraint-based scheduling systems", Intelligent Systems Engineering, Vol. 3, pp. 55–66.