CC BY
17MILITARY SCIENCES
ALGORITHM OF DETERMINING THE READINESS LEVEL OF THE FLIGHT CREW BASED ON
FUZZY LOGIC APPROACHES
Blyskun O.
National Defence University of Ukraine named after Ivan Cherniakhovskyi,
PhD Student, Ukraine ORCID: 0000-0002-7751-8313 Herasymenko V.
National Defence University of Ukraine named after Ivan Cherniakhovskyi,
Doctoral Student, Ukraine ORCID: 0000-0003-2014-7408 Kolomiiets Y.
National Defence University of Ukraine named after Ivan Cherniakhovskyi,
PhD Student, Ukraine ORCID: 0000-0002-9767-0750 Honcharenko Y.
National Defence University of Ukraine named after Ivan Cherniakhovskyi,
Senior Lector, Ukraine ORCID: 0000-0001-7654-6083 Yaroshenko Y.
National Defence University of Ukraine named after Ivan Cherniakhovskyi,
PhD Student, Ukraine ORCID: 0000-0002-8651-4920
ABSTRACT
In this article, the authors analyzed the existing scientific and methodological apparatus for determining the level of readiness of the flight crews. Established that the available methods can determine the level of readiness, but their accuracy is not enough to ensure an adequate level of flight safety. Also, these approaches do not preclude a subjective approach when selecting a flight crew to perform a particular task. To solve these inconsistencies, the authors in this work propose an algorithm for determining the flight crew readiness level based on fuzzy logic approaches. Also, the description of the algorithm according to its main steps is made.
Keywords: flight crew readiness level, aviation, flight crew, mission, fuzzy logic, aviation application, flight safety.
Introduction
Taking into account the influence of the flight crew readiness degree on the success of the flight task is carried out by entering the appropriate coefficient Cprep. In the works [1, 2, 3, 4] for the flight crew the level of preparation considers only as the level of class qualification. And according to [5] the level of class qualification takes into account only the total flight hours and exercises that have been completed in the relevant training course. That is, in works [1, 2, 3, 4], it is said that the pilot of the first class, when performing missions uses all capabilities inherent in the aircraft without exception. Therefore, the coefficient of flight crew readiness of these class qualification levels is proposed to be equal to one (Cprep) [6]. For a flight crew with a level of qualification lower than the first class, the values of this coefficient are assigned depending on the passage of the relevant training course and the total flight hours acquired by them (Cprep < 1). This does not take into account weighty indicators that effect on the readiness level of crews to perform specific combat missions [7].
The readiness level does not have clear boundaries. In conditions when there are no clear boundaries of readiness level, the raised problem can be solved
quite successfully with the use of fuzzy logic which is successfully implemented in MATLAB software which authors use for building a fuzzy logic system. Fuzzy set theory is one of the mathematical theories designed to formalize indefinite information for solving analytical problems. Therefore, the purpose of this work is to determine the scientific and methodological apparatus using fuzzy logic approaches to determine the readiness level of the flight crew with taking into account weighty indicators.
Algorithm of determining the readiness level of the flight crew
At the stage of decision-making for flights and missions, the Senior Aviation Chief assesses the situation, hears and analyzes the proposals of his deputies, heads of services, heads of aviation departments, representatives of support units, etc. [8]. Commander relies on his own experience and intuition. The decision is made in conditions of some uncertainty.
Obviously, the higher the readiness level of the flight crew to perform the mission, the higher probability of its successful completion. In times of shortage time or non-possession of information about own flight crews, the executive must be able to identify the crew to perform a specific mission.
To determine the effectiveness of the fighter application it is necessary to investigate all the factors that affect to the readiness level of the flight crew and identify the main ones. But these factors have no clear boundaries. In conditions when there are no clear boundaries, the problem can be solved quite successfully using fuzzy logic. The fuzzy logic theory is one of the most suitable mathematical theories designed to
Start
formalize indefinite information for solving these kinds of issues.
To perform calculations by the fuzzy logic apparatus, it is necessary to create an algorithm of determining the readiness level, which will allow assigning flight crews on different types of missions based on the results of determining their readiness level. This algorithm is shown in Figure 1.
Flight crew database
m
Input
I
Determining the indicators that affect the readiness level and determining the value of the weight factor of these indicators
Formalization of defined indicators (linguistic variables) <Sj, T, K, G>
Construction membership functions of linguistic variables G = {MX)|X}
^ Creating a database of rules for linguistic variables that form the ' linguistic variable "level of readiness" (if-then)
Construction of a fuzzy logic system
Calculation the readiness level of the flight crew
Checking the results for resonable
Adjusting database of rules
No
-Yes
Assigning flight crews by missions
End
Figure 1: Algorithm of determining the readiness level of the flight crew
Description of the algorithm steps
Step 1. Selection of indicators to determine the readiness and formation of input data. That is, to decide the flight crew appointment to perform the particular mission will be used to quantify the readiness of the crew. The group of experts should determine by voting a certain number of indicators that have the greatest impact on the level of readiness of the flight crew.
Step 2. Formalization of the assessment of input readiness indicators as a tuple is carried out, <£j, T, K, G> where £j - name, T - terms, K - boundaries, G={^£(X)|X} - membership functions [7]. Definition of terms for linguistic variables that characterize the level of readiness of the flight crew and the linguistic variable "level of readiness".
Since quantitative values of variables are required to formalize the decision-making algorithm, use fuzzy logic methods to assess the qualitative indicator of the
level of readiness [10], namely, place on the scale of the value of the linguistic variable "the level of readiness": dangerously low; low; medium; sufficient; high.
To describe the membership function of the linguistic variable £j = "the level of readiness" the terms were named T = {dangerously low; low; medium; sufficient; high} and their boundaries in the range K = [0, 1] were determined. The maximum value of each term takes as 1.
Step 3. Construction of membership functions of linguistic variables. At this step, the limits of the terms selected to determine the level of readiness and for the linguistic variable "level of readiness" are also set. The construction of membership functions is carried out based on regulatory requirements and expert assessments.
Step 4. Determining the relationship between input and output data in the form of linguistic rules "if -then".
Fuzzy rules "if-then" are the core of a fuzzy logic system because they combine all the other components and determine the output of the system. When assessing the level of readiness, input data are often assigned as indicators and results as readiness. Then fuzzy rules "if - then" are established for the ratio of readiness and set of indicators with a certain level of linguistic tolerance [14]. For example, the following is a fuzzy rule "if -then", consisting of five inputs and one output:
IF indicator 1 is low, AND indicator 2 is high, AND indicator 3 is medium AND indicator 4 is high AND indicator 5 is sufficient THEN readiness is medium.
The rules are built systematically, looking at all possible combinations of fuzzy sets of each input from the smallest to the largest. The consequences are adjusted so that the smallest sum of fuzzy sets is equal to the minimum, and the largest sum is equal to the maximum value of readiness. Subtotals are interpolated between these two values. The number of rules is the product of the number of fuzzy sets of each input. For example, for FIS "flight crew level of readiness" the number of logic inputs - 5, the number of terms of the output function - 5, the number of rules is 55 = 3125.
Step 5. Building a fuzzy logic system using the graphical toolkit Fuzzy Logic Designer, from the MATLAB software package. In this application, there is a choice of either Sugeno or Mamdani system [9]. The functions of membership should be determined through statistics and consultation with aviation experts. In this research, the authors propose to use the Mamdani fuzzy inference algorithm. This is the most common inference in fuzzy systems. It uses a minimax composition of fuzzy sets. The centroid of area method of Defuzzification uses.
Step 6. The initial readiness level of the flight crew is calculated and its values are checked for reasonable. The operation of each fuzzy logic block is checked so that it gives the expected initial values and, therefore, confirms that the developed method of analysis is acceptable.
After that, need to run several launches with different input values, and compare the results with each other. The aim is to determine whether the results are reasonable for the model to give realistic and consistent results. After confirming this, the result should be checked for acceptable limits set for the type of operation. If necessary, appropriate adjustments are made.
The assessment of the initial level of readiness is being formalized. Also, values are determined, as well as the choice of the required fuzzy inference algorithm.
Step 7. The obtained values of the readiness level are compared with quantitative indicators that correspond to the values of the linguistic variable "level of readiness" and then appoint flight crew to a mission with an applicable level of complexity.
Conclusions
Thus, analyzing the existing scientific and methodological apparatus for determining the level of read-
iness of the flight crews it becomes obvious that existing methods do not take into account the required number of indicators to determine the flight crew readiness level. And the approaches to assigning tasks to the flight crews are purely subjective.
To replace outdated approaches in determining the level of readiness of flight crews, the authors in this article have developed an algorithm that could take into account any number of indicators to determine the flight crew readiness level. And this will cause to ensure the appropriate level of flight safety when performing tasks by aviation.
Also, in this article, a description of the algorithm according to its main steps is made. The algorithm could be implemented by the software package MATLAB: Simulink and Fuzzy Logic Designer.
The proposed methodology will allow quantifying the readiness level of the flight crew, to take timely measures to organize effective training of crews for possible tasks. By reducing the time in the decision-making process in assigning flight crews on missions, taking into account the level of readiness of crews, and exclusion of a subjective approach in solving this task, the methodology could increase the efficiency of aviation.
References
1. A.S. Bonin, Osnovnye polozheniya metodicheskih podhodov k ocenke boevyh potencialov i boevyh vozmozhnostej aviacionnyh formirovanij, 1nd. ed., Voennaya mysl', Voen. Izd, Moscow, 2008, pp. 43-47.
2. N.M. Skomorohov (Ed.), Bor'ba za gos-podstvo v vozduhe, Voenizdat, Moscow, 1990.
3. B.I. Semon, Suchasnyi metod boiovykh po-tentsialiv v prykladnykh zadachakh planuvannia rozvytku ta zastosuvannia taktychnoi aviatsii, NAOU, Kyiv, 2009.
4. V.N. Shubin, Modelirovanie boevyh dejstvij aviacionnyh chastej i soedinenij pri unichtozhenii vozdushnogo protivnika, Monino, VVA im. YU.A. Ga-garina, 1989.
5. Doc № 79/26524, Instrukciya pro klas-ifikaciyu aviacijnogo personalu derzhavnoi aviacii Ukrainy, Ofic. vid., MOU, Kyiv, 2015.
6. S.S. Drozdov, Metodychnyi pidhid do kil'kisnogo ociniuvannia vplyvu rivnia pidgotovlenosti ekipazhiv na bojovu mogutnist' bojovogo skladu tak-tichnoi aviacii, Nauka i tekhnika Povitrianih Syl Zbro-jnyh Syl Ukrainy 3 (24) (2016), 49-53.
7. Y. Goncharenko, O. Blyskun, O. Martyniuk et al., Flight safety fuzzy risk assessment for combat aviation system, in: Proceedings of the 2nd. IEEE International Conference on Advanced Trent in Information Theory, Kyiv, 2020, pp. 132-137.
8. Doc 82/26527, Pravyla vykonannya pol'otiv v derzhavnij aviacii Ukrainy, Ofic. vid., MOU, Kyiv, 2015.
9. V.I. Gostev (Ed.), Nechetkie regulyatory v sistemah avtomaticheskogo upravleniya, Radioamator, Kyiv, 2008.
10. A.N. Borisov, Prinyatie reshenij na osnove nechetkih modelej: primery ispol'zovaniya, Zinatne, Riga, 1990.
11. V. Harchenko, T. ShmeFova, Yu. Sikirda, Pryjnyattya rishen v sociotexnichnyh systemah: monograph, NAU, Kiyv, 2016, ISBN 978-966-932-010-0.
12. O.M. Kernic'kij, Metodyka formuvannya psi-hologichnoi gotovnosti kursantiv-l'otchikiv do l'otnoi diyal'nosti, Ph.D. thesis, Kharkiv University of Air Force, Kharkiv, 2005.
13. Doc, Metodychni rekomendacii z psyhologichnoi pidgotovky l'otnogo skladu pid chas or-ganizacii zahodiv kolektyvnoi pidgotovky osobovogo skladu Povitryanih Syl Zbrojnyh Syl Ukrainy, Ofic. vid., PS ZSU, Vinnytsia, 2015.
14. A. Leonenkov, Fuzzy modeling in Matlab and Fuzzy Tech, St. PTB, BHV, 2003, ISBN 5-94157-0872.
