Optimization of operating costs in a multifunctional digitalized system based on the results of predictive analytics using the example of an aircraft manufacturing enterprise.
Аuthors
Voronezh State Technical University, VSTU, 14, Moskovsky prospect, Voronezh, 394026, Russia
e-mail: pgusev@cchgeu.ru
Abstract
The paper discusses the use of predictive analytics as a tool for correcting management decisions in multifunctional digitalized systems. Factors influencing the size of operating costs are considered. Among the most significant factors are: overestimation of system performance targets, unplanned system components, use of excessive amounts of resources for the operation of system components, unplanned maintenance and repair of system components. A formal description of three predictive analytics tasks that provide decision support for optimizing operating costs is proposed. conducting a separate study.
Solving problems of predictive analytics makes it possible to assess the optimality of carrying out an intelligent procedure for disaggregating resources and volumes of activity by type of activity. Combining the conditions for compliance of planned and forecast values forms an optimization model for reducing operating costs based on the results of predictive analytics. At the same time, failure to meet the model conditions associated with the results of predictive analytics leads to the need for a new iteration of the intelligent disaggregation procedure. Depending on the failure to meet a specific condition of the proposed optimization model, the management decisions described in the work can be made.
Predictive analytics problems can be classified as machine learning regression problems. It should be noted that in certain cases a classification task may also occur if the system performance indicator is binary, takes the values “completed” or “failed”, or can be classified by a finite set of values.
The production system of an aircraft manufacturing enterprise was selected as an experiment on the use of predictive analytics models. A manufacturing enterprise is an illustrative example of a multifunctional digitalized system, because management in such systems is carried out using information technologies based on digital platforms and services. For the system under consideration, a predictive analysis was carried out in accordance with the formalized tasks of predictive analytics.
As a result of the predictive analytics procedure at an aircraft manufacturing enterprise, the possibilities of fulfilling the production plan, as well as ways to reduce operating costs, were determined. Factors influencing the increase in operating costs and the ability to fulfill the production plan have been identified. Such factors in real systems are usually the human factor or the lack of external resources - for example, the supply of raw materials.
Keywords:
digitalized system, predictive analytics, operating costs, aircraft manufacturing enterpriseReferences
1. Vail P., Vorner S. Tsifrovaya transformatsiya biznesa: Izmenenie biznes-modeli dlya organizatsii novogo pokoleniya (Digital business transformation: Changing the business model for a new generation organization), Moscow, Al'pina Pablisher, 2019, 257 p.
2. Khasanov A.R. Vliyanie prediktivnoi analitiki na deyatel'nost' kompanii, Strategicheskie resheniya i risk-menedzhment, 2018, no. 3 (108), URL: https://doi.org/10.17747/2078-8886-2018-3-108-113
3. Zaichenko I.M., Yakovleva M.A. Nauchnyi vestnik Yuzhnogo instituta menedzhmenta, 2019, no. 2, pp. 18-23.
4. Gusev P.Yu. L'vovich Ya.E. Modelirovanie, optimizatsiya i informatsionnye tekhnologii, 2023, no. 11 (4). URL: https://moitvivt.ru/ru/journal/pdf?id=1441. DOI: 10.26102/2310-6018/2023.43.4.004
5. Dorozhko I.V., Gorokhov G.M., Kirillov I.A. Trudy MAI, 2022, no. 125. URL: https://trudymai.ru/eng/published.php?ID=168195. DOI: 10.34759/trd-2022-125-23
6. Rysbai N. Vestnik nauki, 2023, no. 5 (62). URL: www.vestnik-nauki.rf/volume/journal-5-62-2
7. Gavrilina E.A., Zakharov M.A., Karpenko A.P. Nauka i obrazovanie, 2015, no. 4. DOI: 10.7463/0415.0764221
8. Gusev P.Yu. Sistemy upravleniya i informatsionnye tekhnologii, 2023, no. 4 (94), pp. 4-7.
9. Minakov E.P., Privalov A.E., Bugaichenko P.Yu. Trudy MAI, 2023, no. 131. URL: https://trudymai.ru/eng/published.php?ID=175925. DOI: 10.34759/trd-2023-131-19
10. Kasatikov N.N., Brekhov O.M., Nikolaeva E.O. Trudy MAI, 2023, no. 131. URL: https://trudymai.ru/eng/published.php?ID=175929. DOI: 10.34759/trd-2023-131-23
11. Antropova E.M., Konkin N.A. Vserossiiskie otkrytye Armandovskie chteniya «Sovremennye problemy distantsionnogo zondirovaniya, radiolokatsii, rasprostraneniya i difraktsii voln»: sbornik trudov. Murom, Vladimirskii gosudarstvennyi universitet im. Aleksandra Grigor'evicha i Nikolaya Grigor'evicha Stoletovykh, 2023, pp. 275-283.
12. Lutai V.N., Khusainov N.Sh. Izvestiya YuFU. Tekhnicheskie nauki, 2022, no. 5 (229). URL: https://izv-tn.tti.sfedu.ru/index.php/izv_tn/article/view/707
13. Bazilevskii M.P. Problemy upravleniya, 2022, no. 4, pp. 3-14. DOI: 10.25728/pu.2022.4.1
14. Kobko L.I. Trudy MAI, 2001, no. 3. URL: https://trudymai.ru/published.php?ID=34687
15. Ksenofontov V.V. Problemy nauki, 2020, no. 11 (59). URL: https://cyberleninka.ru/article/n/neyronnye-seti-1
16. Gusev P.Yu. Gusev K.Yu. Trudy MAI, 2020, no. 110. URL: https://trudymai.ru/eng/published.php?ID=112933. DOI: 10.34759/trd-2020-110-20
17. Gunawardana A., Shani G. A survey of accuracy evaluation metrics of recommendation tasks, Journal of Machine Learning Research, 2009, vol. 10, no. 12. DOI: 10.1145/1577069.1755883
18. Chizhov M.I., Skripchenko Yu.S., Gusev P.Yu. Vestnik Voronezhskogo gosudarstvennogo tekhnicheskogo universiteta, 2012, vol. 8, no. 12-2, pp. 73-75.
19. Chizhov M.I., Skripchenko Yu.S., Gusev P.Yu. Komp'yuternye issledovaniya i modelirovanie, 2014, vol. 6, no. 2, pp. 245-252.
20. Stepanenko A.S. Nauchnyi vestnik MGTU GA, 2009, no. 150, pp. 135-138.
Download