Neural network-based justification of management decisions on city digitalization

Osipov V.Y.
 pdf (2186K)

The modern digitalization of cities occurs according to laws that are not yet clearly defined. The difficulty in understanding these laws is due to the high uncertainty of urban events and the difficulties in developing adequate analytical models for them. To substantiate management decisions on city digitalization in these conditions, effective methods for processing multidimensional time series are necessary. The objective is to develop a promising method. The proposed method is based on a new model of a neural network control system and algorithms for implementing its main functions. Distinctive features of the model include both structural aspects and new rules for neural network-based substantiation of management decisions on city digitalization, using unsupervised training of the applied recurrent neural network. This model provides for adaptive neural network management of city digitalization according to a sliding program. New rules are proposed for the forward and backward transformation of multidimensional time series, modeling the operation of a streaming recurrent neural network, and identifying feasible programs for city digitalization. By adhering to the proposed rules, it is possible not only to eliminate information loss in such a network but also to ensure its stable operation. The features of this neural network’s operation in substantiating feasible urban digitalization programs are described. The requirements for the software implementation of the proposed method are substantiated. The results of experiments using data for St. Petersburg are presented. These data included monthly volumes of shipped computers, electronic and optical products, as well as monthly volumes of shipped products for all manufacturing industries minus the first-time series values ??for the period from January 2021 to December 2025. The effectiveness of management decisions on the city’s digitalization was predicted based on the total volumes of shipped products to the horizon from January to December 2026. It is shown that the application of the proposed method allows for expanded capabilities for adaptively substantiating feasible management decisions on the city’s digitalization while reducing computer modeling costs.

Keywords: city, digitalization, neural network control, modeling algorithms, forecasting
Citation in English: Osipov V.Y. Neural network-based justification of management decisions on city digitalization // Computer Research and Modeling, 2026, vol. 18, no. 4, pp. 1053-1070
Citation in English: Osipov V.Y. Neural network-based justification of management decisions on city digitalization // Computer Research and Modeling, 2026, vol. 18, no. 4, pp. 1053-1070
DOI: 10.20537/2076-7633-2026-18-4-1053-1070
URL: http://crm-en.ics.org.ru/journal/article/3759/
Creative Commons License This work is licensed under a Creative Commons Attribution-NoDerivs 3.0 Unported License.

Copyright © 2026 Osipov V.Y.

Indexed in Scopus

Full-text version of the journal is also available on the web site of the scientific electronic library eLIBRARY.RU

The journal is included in the Russian Science Citation Index

The journal is included in the RSCI

International Interdisciplinary Conference "Mathematics. Computing. Education"

Website Copyright © 2009–2026 Institute of Computer Science