Prediction of production plan implementation of aviation enterprise employing a fuzzy neural network

DOI: 10.34759/trd-2020-110-20

Аuthors

Gusev P. Y.^*, Gusev K. Y.

Voronezh State Technical University, VSTU, 14, Moskovsky prospect, Voronezh, 394026, Russia

*e-mail: GusevPvl@gmail.com

Abstract

Prior to a new product production startup or changing production program of aviation enterprise, it is necessary to unambiguously determine whether there are enough production resources to fulfill the new production program. The answer to this question is possible only in the case the presence of sufficient amount of reliable information on the production work.

In cases where information is insufficient, the authors propose to use simulation model as an information source. The article considers a simulation model of a workshop for parts production fr om polymer composite materials. The developed model was verified by the key points of production. In the simulation model, a series of experiments was carried out where production plan was the variable parameter, and the fact of the monthly production plan fulfillment was the criterion.

Based on the experiments, a data set was generated. This data set was analyzed, and certain conclusions were drawn on the production work.

To automate the predicting process, a fuzzy system model with an external knowledge base is employed, wh ere the Gauss function is used as a form for representing fuzzy sets. The first stage of this model functioning is a knowledge base compilation from the training sample in the form of “input vector”-“predicted value” pairs. The data in the knowledge base is stored in the form of membership functions for the corresponding Gaussian curves rather than in absolute values. At the second stage, using the generated knowledge base, a forecast is performed by calculating the degree of belonging of the existing situation and the reference.

The fuzzy-neural network functioning with fuzzy sets is required to perform mathematical operations performed by the following blocks: a block for bringing state variables to fuzzy sets, a block for generating a solution, a block for mapping output fuzzy sets to the forecast value.

Based on the above-described algorithm, a software module for the production plan analysis and its fulfillment forecasting was realized. The developed module allows employing various data entry formats, as well as conducting experiments.

A test sample was analyzed, from which a knowledge base was being formed, and control checks on the test sample were carried out. The results of check test gave 93% of forecast matches in the test sample.

Keywords:

neural network, fuzzy system, simulation, prediction, mathematical model, programming tool

References

1. Basova L.A. et al. Kazanskii meditsinskii zhurnal, 2014, vol. 95, no. 1, pp. 91 - 94.

2. Ivanov N.N. Otkrytoe obrazovanie, 2011, no. 2-2, pp. 52 – 55.

3. Strel'nikov D.D. Vestnik Astrakhanskogo gosudarstvennogo tekhnicheskogo universiteta. Seriya: Morskaya tekhnika i tekhnologiya, 2018, no. 3, pp. 123 - 128.

4. Sreekumar S., Bhakar R. Solar Power Prediction Models: Classification Based on Time Horizon, Input, Output and Application, 2018 International Conference on Inventive Research in Computing Applications (ICIRCA), IEEE, 2018, pp. 67 - 71. DOI: 10.1109/ICIRCA.2018.8597288

5. Mercy K.G., Rao S. K.S. A Framework for Rail Surface Defect Prediction Using Machine Learning Algorithms, 2018 International Conference on Inventive Research in Computing Applications (ICIRCA), IEEE, 2018, pp. 972 - 977. DOI: 10.1109/ICIRCA.2018.8597394

6. Paliwal S., Khatri S.K., Sharma M. Sentiment Analysis and Prediction Using Neural Networks, International Conference on Advanced Informatics for Computing Research, Springer, Singapore, 2018, pp. 458 - 470.

7. Yadykin E.A. Izvestiya Tul'skogo gosudarstvennogo universiteta. Tekhnicheskie nauki, 2012, no. 1, pp. 450 – 456.

8. Abramov A.A., Shershakov I.S. Vestnik Tambovskogo universiteta. Seriya: Gumanitarnye nauki, 2012, vol. 110, no. 6, pp. 72 – 77.

9. Zernov V.I., Paramonov A.P. Trudy MAI, 2008, no. 31, available at: http://trudymai.ru/eng/published.php?ID=7495

10. Semakov S.L., Semakov I.S. Trudy MAI, 2018, no. 100, available at: http://trudymai.ru/eng/published.php?ID=93446

11. Straková J. et al. Situational analysis and its role in the process of strategic business management, Polish Journal of Management Studies, 2018, vol. 18, no. 1, pp. 353 – 364. DOI: 10.17512/pjms.2018.18.1.26

12. Prasad R.B., Arif M. Workspace and Singularity Analysis of Five Bar Planar Parallel Manipulator, 2018 5th IEEE Uttar Pradesh Section International Conference on Electrical, Electronics and Computer Engineering (UPCON), IEEE, 2018, DOI: 10.1109/UPCON.2018.8596991

13. Wielki J., Kozioł P. The analysis of opportunities to use the lean IT concept in modern enterprise, Polish Journal of Management Studies, 2018, vol. 18, pp. 388 – 401. DOI:10.17512/pjms.2018.18.2.31

14. Rutkovskaya D., Rutkovskii L., Pilin'skii M. Neironnye seti, geneticheskie algoritmy i nechetkie sistemy (Neural networks, genetic algorithms, and fuzzy systems), Moscow, Goryachaya liniya-Telekom, 2003, 384 p.

15. Yarushkina N.G. Osnovy teorii nechetkikh i gibridnykh system (Fundamentals of fuzzy and hybrid systems theory), Moscow, Finansy i statistika, 2004, 320 p.

16. Tran D.T. Temporal attention-augmented bilinear network for financial time-series data analysis, IEEE transactions on neural networks and learning systems, 2018, vol. 30, no. 5, pp. 1407 - 1418.

17. Danilov A. Intellectual decision-making system in the context of potentially dangerous nuclear power facilities, MATEC Web of Conferences, EDP Sciences, 2018. DOI:10.1051/matecconf/201816102009

18. Privalov A.E., Dorozhko I.V., Zakharova E.A., Kopeika A.L. Trudy MAI, 2018, no. 103, available at: http://trudymai.ru/eng/published.php?ID=101526

19. Chizhov M.I., Skripchenko Yu.S., Gusev P.Yu. Komp'yuternye issledovaniya i modelirovanie, 2014, vol. 6, no. 2, pp. 245 - 252.

20. Chizhov M.I., Skripchenko Yu.S., Gusev P.Yu. Vestnik Voronezhskogo gosudarstvennogo tekhnicheskogo universiteta, 2012, vol. 8, no. 12-2, pp. 73 - 75.

Download