V.O. Novitsky1, A.S. Maksimov2, М.А. Zinovieva3

1–3 ROSBIOTECH (Moscow, Russia)
1nvo60@mgupp.ru,2maksimov@mgupp.ru,3zinovevama2003@gmail.com

Automated information system for bakery enterprise using the subsystem of optimisation of bakery products baking process control represents an opportunity to improve product quality, reduce costs and increase the competitiveness of the enterprise. The approach is based on a thorough analysis, problem statement of development and identification of the mathematical model of baking process control optimisation, experimental validation, use of modern optimisation methods and appropriate software libraries, evaluation of model efficiency.

In the course of the study the following conclusions are drawn. Firstly, the solution of the optimisation problem with the help of the obtained mathematical model excludes personnel errors, because the change of values of initial approximations in the procedure of gradient descent showed the mean square deviation of results by less than 5%. Secondly, it is found that the algorithm is sensitive to varying the importance of each quality characteristic. For example, varying the importance factor for porosity resulted in a 77% change in the results, and 57% for upek. Third, the user adapts the algorithm to the requirements of the task at hand, given its constraints. For example, in order to obtain a favourable percentage of rye flour in the mixture, it is necessary to specify a narrow range (For example, from 50% to 60% or from 75% to 80%), which is proved in experiment #3. And also the economic profitability of the algorithm is confirmed. If a factory producing bread from 100% wheat flour starts adding 40% rye flour to the mix – the benefit is 7 rubles per 100 kg. For a factory producing 50 tonnes of bread per day – profit of 1.278 million rubles per year. Thus, the set goal – to improve the quality of bread, to ensure control and reduce production costs – has been achieved.

Further stages in the development of work can be the expansion of system functionality, improvement of algorithms, integration with other systems, creation of user interface, expansion of the volume of input data, as well as the study of the impact of the obtained optimal parameters on the economic efficiency of production of bakery products of a wide range.

Novitsky V.O., Maksimov A.S., Zinovieva М.А. Automated system for optimisation of bakery products baking process control. Dynamics of complex systems. 2024. V. 18. № 4. P. 58−71. DOI: 10.18127/j19997493-202404-06 (in Russian).

1. Kiryuhina A.N., Grigor'eva R.Z., Kozhevnikova A.Yu. Sovremennoe sostoyanie i perspektivy razvitiya proizvodstva hleba i hlebobulochnyh izdelij v Rossii. Tekhnika i tekhnologiya pishchevyh proizvodstv. 2019. T. 49. № 2. S. 330–337 (in Russian).
2. Erofeeva A.V. Tendencii razvitiya hlebopekarnogo proizvodstva v Rossii. Molodoj uchenyj. 2020. № 27 (317). S. 181–182 (in Russian).
3. Bryazun V.A. Teplotekhnicheskie aspekty effektivnoj vypechki pshenichnyh hlebobulochnyh izdelij: Monografiya. M.: Pishchepromizdat, 2004. 142 s. (in Russian).
4. Bryazun V.A., Adnodvorcev M.F. Racional'nyj rezhim uvlazhneniya testozagotovok pri vypechke hlebnyh izdelij. Hleboprodukty. 2017. № 9. S. 51–53 (in Russian).
5. Anoshina O.M., Kovaleva I.E. Vliyanie prodolzhitel'nosti vypechki na kachestvo rzhano-pshenichnogo hleba. Hleboprodukty. 2009. № 5. S. 52–53 (in Russian).
6. Anoshina O., Kovaleva I., Puchkova L. Izmenenie mikrostruktury rzhano-pshenichnogo hleba v processe vypechki. Hleboprodukty. 2010. № 1. S. 46–47 (in Russian).
7. Anoshina O.M., Kovaleva I.E. Vliyanie sootnosheniya rzhanoj i pshenichnoj muki na kachestvo hleba. Innovacionnye tekhnologii v pishchevoj i legkoj promyshlennosti: Materialy Mezhdunar. nauch.-prakt. konf. Almaty: ATU, 2009. S. 5–6 (in Russian).
8. Chernyh V.Ya., Maksimov A.S., Bryazun V.A., Artamonov A.V. Informacionno-izmeritel'nye sistemy kontrolya fiziko-himicheskih harakteristik pri zamese pshenichnogo testa. Hleboprodukty. 2018. № 2. S. 34–37 (in Russian).
9. Isabekova L.S., Utepbergenov I.T., Tojbaeva Sh.D. Sistema upravleniya temperaturoj i vlazhnost'yu hlebopekarnoj pechi. Intellectual Technologies on Transport. 2023. № S1. Special Issue MMIS. 2023. S. 77–85 (in Russian).
10. Kulachenko I.A. Primenenie avtomatizirovannoj sistemy upravleniya tekhnologicheskim processom v deyatel'nosti hlebopekaren. Perspektivy razvitiya nauchnyh issledovanij: Sb. nauch. trudov Mezhdunar. nauch.-prakt. konf.. g.-k. Tyumen', 12 maya 2022 g. Tyumen': NIC TI, 2022. S. 140–144 (in Russian).
11. Gurianov D.A., Myshenkov K.S., Terekhov V.I. Software Development Methodologies: Analysis and Classification. 2023 5th International Youth Conference on Radio Electronics, Electrical and Power Engineering (REEPE). Russian Federation, Moscow, 16–18 March 2023. M.: IEEE, 2023. Vol. 5. P. 1–8. DOI 10.1109/REEPE57272.2023.10086852.
12. Goryachkin B.S., Myshenkov K.S., Harlashkin A.I. Analiz metodov konceptual'nogo proektirovaniya avtomatizirovannyh informacionnyh sistem. Dinamika slozhnyh sistem – XXI vek. 2020. T. 14. № 3. S. 23–34. DOI 10.18127/j19997493-202003-02 (in Russian).
13. Myshenkov K.S., Simonov M.F., Guzilov A.V. Obosnovanie vybora programmnyh sredstv postroeniya funkcional'nyh modelej informacionnyh sistem. Dinamika slozhnyh sistem – XXI vek. 2020. T. 14. № 4. S. 38–45. DOI 10.18127/j19997493-202004-04 (in Russian). 
14. Kakalyev B.A., Ovezova G.Ch. Ispol'zovanie metodov matematicheskogo analiza dlya resheniya zadach optimizacii. Vsemirnyj uchenyj. 2024. T.1. № 20. S. 24–27 (in Russian).
15. Shevcova T.G., Kotlyarov R.V., Romanova V.V., Krol', A.N. Onlajn-kal'kulyator «Raschet receptur hlebobulochnyh izdelij» dlya ARM operatora-tekhnologa. Vestnik VGUIT. 2019. T. 81. № 1. S. 201–206 (in Russian).
16. Vardomackaya E.Yu. Primenenie metodov ekspertnyh ocenok pri vybore upravlencheskih reshenij. Innovacionnaya nauka. 2021. № 6. S. 83–85 (in Russian).
17. Dzhalilov Sh.A. Metod rascheta parametrov mnozhestvennoj linejnoj regressii. Dostizheniya nauki i obrazovaniya. 2020. № 3 (57). S. 24–28 (in Russian).
18. Zhulanov V.N. Primenenie gradientnogo spuska dlya obucheniya modeli mashinnogo obucheniya. Innovacionnye tekhnologii: teoriya, instrumenty, praktika. 2021. T. 1. S. 16–21 (in Russian).
19. Novickij V.O., Bol'shakov E.Yu. Razrabotka informacionnoj sistemy upravleniya na osnove sovremennyh Web-tekhnologij na primere CRM-sistemy. Dinamika slozhnyh sistem – XXI vek. 2023. № 4. S. 34–44 (in Russian).
20. Vvedenie v Apache Commons Math. URL: https://www.codeflow.site/ru/article/apache-commons-math (data obrashcheniya 27.05.24 (in Russian)).
21. DL4J (Deeplearning for Java) – Getting Started. URL: https://depiesml.wordpress.com/2015/08/26/dl4j-gettingstarted/ (data obrashcheniya 27.05.24) (in Russian).
22. TensorFlow dlya Java. URL: https://www.tensorflow.org/jvm?hl=ru (data obrashcheniya 27.05.24) (in Russian).
23. Encog Machine Learning Framework. URL: https://www.heatonresearch.com/encog/ (data obrashcheniya 27.05.24).