Forecasting the volume of raw materials procurement on the territory of communities using artificial neural networks
DOI:
https://doi.org/10.31734/agroengineering2020.24.143Keywords:
forecasting, artificial neural networks, raw materials, procurement, communityAbstract
The analysis of scientific works and subject area is performed. It testifies the expediency of the process of forecasting the volumes of raw materials procurement on the territory of communities. The unsolved scientific and applied problem of developing tools for forecasting the volume of raw material procurement in the communities using artificial neural networks has been identified.
The architecture of the artificial neural network for forecasting the volumes of raw material procurement on the territory of communities is substantiated. It suggests the use of a three-layer perceptron. The block diagram of training of an artificial neural network for forecasting of volumes of daily preparation of raw materials in the territory of communities is developed. It involves implementation of fifteen stages, which are based on the study and preparation of initial data for forecasting, as well as calculations and verification of the conditions of their accuracy, which ensures the proper quality of training of artificial neural networks.
The proposed approach to classification of the data for forecasting suggests separation of individual days and months of the year, which, under some climatic conditions (average atmospheric temperature and precipitation per day), determine the method of keeping the dairy herd (tethered and grazed), significantly effects the volume of milk production and underlies their forecasting.
Based on the prepared initial data, an artificial neural network was trained. It provided the creation of a model that can predict the daily volume of milk production in the community based on the input data in the Neural Network Wizard. The studies show that for the number of learning epochs over 20,000, the error of the predicted values does not exceed 3.6%. On the basis of the performed research of the adjusted artificial neural network, the tendencies of the change of daily volumes of milk procurement on the territory of the community (real and projected) for the conditions of Ponykovychi community in Brody district of Lviv region are established.
References
Bazhenov, R. I., & Stupnikov, A. V. (2015). Prognozirovanie tseny legkovykh avtomobilei s pomoshchiu neironnykh setei v srede Neural Network Wizard. Sovremennaia tekhnika i tekhnologii: elektron. izd., 7. Retrieved from: http://technology.snauka.ru/2015/07/7462.
Vychuzhanin, V. V., & Rudnichenko, N. D. (2016). Razrabotka modeli neiroseti dlia prognozirovaniia riska otkazov komponentov slozhnykh tekhnicheskikh sistem. Informatics and Mathematical Methods in Simulation, 6 (4), 333-338.
Kostina, L. N., & Gareeva G. A. (2015). Neironnye seti v zadachakh prognozirovaniia vremennykh riadov. Innovacionnaia nauka, 6-2, 70-73.
Liubimova, T. V., & Gorelova, A. V. (2015). Reshenie zadachi prognozirovaniia s pomoshchiu neironnyh setei. Innovacionnaia nauka, 4, 39-42.
Orlova, E. V., Dudnikova, E. V., Vodopianov, A. S., & Chernova, M. S. (2019). Primenenie neironnykh setei v prognozirovanii riska formirovaniia tserebralnogo paralicha u detei na osnove otsenki vliianiia perinatalnykh faktorov. Vestnik sovremennoi klinicheskoi mediciny, 12, 3, 40–43. doi:10.20969/ VSKM.2019.12(3).40-43.
Pokusaev, M. N., & Kasimov, N. N. (2012). Sistema diagnostiki sudovykh energeticheskikh ustanovok s primeneniem neirosetevykh modelei. Vestnik AGTU. Upravlenievychislitelnaia tekhnika i informatika, 2, 88-92.
Soldatova, O. P., & Semenov, V. V. (2006). Primenenie neironnyh setei dlia resheniia zadach prognozirovaniia. Issledovano v Rossii: elektron. nauch. zhurn. Retrieved from http://zhurnal.gpi.ru/ articles/2006/136.pdf.
Tryhuba, A. M. (2011). Klasyfikatsiia ta osoblyvosti realizatsii intehrovanykh proektiv ahrarnoho vyrobnytstva. Upravlinnia proektamy, systemnyi analiz i lohistyka, 8, 197-201. Retrieved from http://nbuv.gov.ua/UJRN/Upsal_2011_8_53.
Tryhuba, A. M., Sheleha, O. V., Pukas, V. L., & Mykhaliuk, V. M. (2015). Uzghodzhennia konfihuratsii intehrovanykh proektiv ahrarnoho vyrobnytstva. Visnyk NTU «KhPI». Seriia: Stratehichne upravlinnia, upravlinnia portfeliamy, prohramamy ta proektamy, 2 (1111), 135-140.
Tsaregorodtsev, V. G. (2012). Proizvodstvo poluempiricheskikh znanii iz tablits dannykh s pomoshchiu obuchaemykh iskustvennykh neironnykh setei. In Metody neiroinformatiki (pp. 89-101). Krasnojarsk: Izd-vo KHTU.
Awadalla, H. A., Ismaeil, I. I., & Sadek, M. A. (2011). Spiking neural network-based control chart pattern recognition. Journal of Engineering and Technology Research, 3 (1), 5-15.
Dede, G., & Sazli, M. Speech recognition with artificial neural networks. (2010). Digital Signal Processing, 20 (3), 763-768.
Tryhuba, A., Boyarchuk, V., Tryhuba, I., Boyarchuk, O., & Ftoma, O. (2019). Evaluation of risk value of investors of projects for the creation of crop protection of family dairy farms. Acta universitatis agriculturae et silviculturae mendelianae brunensis, 67, 5, 1357-1367. Retrieved from https://acta.mendelu.cz/67/5/1357/.
Тryhuba, А., Boyarchuk, V., Tryhuba, I., Ftoma, O., Padyuka, R., & Rudynets, M. (2020). Forecasting the Risk of the Resource Demand for Dairy Farms Basing on Machine Learning. Proceedings of the 2nd International Workshop on Modern Machine Learning Technologies and Data Science (MoMLeT+DS 2020), I, 327-340.
Neural Network Wizard. BaseGroup Labs. Retrieved from http://www.basegroup.ru/download/demoprg/nnw/.
Ratushny, R., Tryhuba, A., Bashynsky, O., & Ptashnyk, V. (2019). Development and usage of a computer model of evaluating the scenarios of projects for the creation of fire fighting systems of rural communities. XI-th International Scientific and Practical Conference on Electronics (ELIT-2019) (pp. 34-39). Retrieved from https://ieeexplore.ieee.org/document/8892320.
Tryhuba, A. (2014). Argumentation of the parameters of the system of purveyance of milk collected from the private farm-steads within a single administratinve district. Econtechmod: an international quarterly journal on economics in technology, new technologies and modelling processes, 4 (3), 23-27.
Тryhuba, А., & Bashynsky, O. (2019). Coordination of dairy workshops projects on the community territory and their project environment. 14th International Scientific and Technical Conference on Computer Sciences and Information Technologies (CSIT), 3 (рр. 51-54). Retrieved from https://ieeexplore.ieee.org/document/8929816.
Тryhuba, А., Ftoma, O., Tryhuba, I., & Boyarchuk, O. (2019). Method of quantitative evaluation of the risk of benefits for investors of fodder-producing cooperatives. 14th International Scientific and Technical Conference on Computer Sciences and Information Technologies (CSIT), 3 (pp. 55-58). Retrieved from https://ieeexplore.ieee.org/document/8929788.
