Arkadeva O.G., Petrova M.A.
Optimizing the placement of a commercial bank’s branch network
Keywords: geomarketing, location of the bank's offices, development of the branch network, geoinformation systems, socio-economic indicators of the region
The banking sector has a significant impact on the development of regional socio-economic systems, providing them with necessary financial resources. At this, there is a high concentration of banking resources in the central regions of Russia and limited access to banking resources in certain regions. The purpose of the study is to analyze unevenness of the territorial distribution of bank offices and to find ways to eliminate such imbalances. Materials and methods. The analytical model for optimizing the location of the branch network was built using data from the Bank of Russia on loans owed by individuals, customer funds, loans from legal entities, as well as Rosstat data on the average monthly salary of employees, the number of active enterprises, population and the index of output of goods and services. The geomarketing model is implemented in Python using Pandas, NumPy, Matplotlib, and Scikit-learn libraries (RobustScaler, LinearRegression, RandomForestRegressor, and KMeans). The constructed ensemble model combines 3 areas of the machine learning model (market potential, efficiency (ROI), gap analysis (under-coverage)). The K-Means algorithm was applied to cluster the regions. Results. To identify the most promising regions for opening bank offices, models of market potential, efficiency per office (ROI), a gap model (under-coverage) and an ensemble model were built. The market potential model identifies the most promising regions based on the dynamics of socio-economic development indicators. The Office Efficiency Model (Office ROI) determines the regions in terms of the existing banking network's effectiveness. The gap (under-coverage) model is based on constructing a linear regression model to determine the projected number of offices, as well as to find deviations from the actual number of existing offices. Based on these models, an ensemble model was built that provides an overall assessment of the region's prospects for the development of a branch network. In addition, a cluster analysis of the regions was conducted, according to the results of which the regions were divided into five clusters: highly developed regions with a high economic activity (cluster 0), regions with a low level of development (cluster 1), regions with an average level of development and growth potential (cluster 2), regions with uneven activity (cluster 3), highly specialized regions with high activity of the corporate segment (cluster 4). The constructed geomarketing model makes it possible to identify promising regions for the development of a branch network and takes into account the socio-economic situation of the region, the effectiveness of banking activities in the region, as well as the region's need for new banking offices. Conclusions. In the context of banking digitalization, geomarketing technologies are becoming particularly important, enabling to determine the optimal location of credit institutions' branches in order to increase the efficiency of the bank's activities and availability of banking services to the public. The constructed geomarketing model combines an ensemble model that evaluates the development potential of the region itself, and a cluster analysis that takes into account the socio-economic and geographical features characteristic of the cluster. This approach serves as the basis for developing a strategy for the development of the banking network, including determining the main directions and optimal formats of banking presence in the region.
References
- Dmitriyeva S.I., Zakharova M.V. Integratsiya bol’shikh dannykh i prostranstvennogo analiza v geomarketingovykh issledovaniyakh: metodologicheskiy podkhod k otsenke potrebitel’skogo povedeniya v megapolisakh [Integration of big data and spatial analysis in geomarketing research: a methodological approach to assessing consumer behavior in megacities]. Ekonomika i predprinimatel’stvo, 2025, no. 4(177), pp. 1409–1417. DOI: 10.34925/EIP.2025.177.4.249.
- Dolzhenko R.A. Metodika otsenki moshchnosti filial’noy seti banka dlya tseley geomarketingovogo modelirovaniya [Methodology for assessing the capacity of a bank’s branch network for the purposes of geomarketing modeling]. Vestnik Omskogo universiteta. Seriya: Ekonomika, 2017, no. 3, pp. 107–117. DOI: 10.25513/1812-3988.2017.3.107-117.
- Dolzhenko R.A. Podkhody k geomarketingu raspolozheniya ofisov obsluzhivaniya fizicheskikh i yuridicheskikh lits kommercheskogo banka [Approaches to geomarketing for the location of service offices for individuals and legal entities of a commercial bank]. Biznes-informatika, 2017, no. 3(41), pp. 41–55. DOI: 10.17323/1998-0663.2017.3.41.55.
- Yepifanova S.O. Ispol’zovaniye geomarketinga v bankovskom sektore [The use of geomarketing in the banking sector]. In: Sovremennyye tendentsii razvitiya teorii i praktiki upravleniya v Rossii i za rubezhom: materialy VIII Mezhdunarodnoy nauchno-prakticheskoy konferentsii, Tambov, 23–24 noyabrya 2017 goda [ of the 8th Int. Sci. and Pract. Conf. “Current Trends in the Development of Management Theory and Practice in Russia and Abroad”]. Tambov, 2017, pp. 87–97.
- Naumov I.V. Issledovaniye i modelirovaniye prostranstvennoy lokalizatsii i peremeshcheniya bankovskogo kapitala [Research and modeling of spatial localization and movement of banking capital]. Nalogi. Pravo, 2021, vol. 14, no. 6, pp. 41–51. DOI: 10.26794/1999-849X-2021-14-6-41-51.
- Osipovskaya A.V. Bankovskaya sistema Rossii: regional’nyy aspekt [Banking system of Russia: regional aspect]. Vestnik RGGU. Seriya «Ekonomika. Upravleniye. Pravo», 2021, no. 4, pp. 111– DOI: 10.28995/2073-6304-2021-4-111-122.
- Pervitskaya L.A. Vybor lokatsii novoy torgovoy tochki s ispol’zovaniyem metodov prostranstvennogo analiza dannykh [Selecting the location of a new retail outlet using methods of spatial data analysis]. Shag v nauku, 2023, no. 4, pp. 118–122.
- Shaytura S.V., Galkin D.A. Geomarketingovyy analiz bol’shikh dannykh [Geomarketing analysis of big data]. Informatsionnyye tekhnologii, 2021, vol. 27, no. 4, pp. 180–187. DOI: 10.17587/it.27.180-187.
- Indeks vypuska tovarov i uslug po bazovym vidam ekonomicheskoy deyatel’nosti po sub”ektam Rossiyskoy Federatsii i federal’nym okrugam [Index of output of goods and services by basic types of economic activity by constituent entities of the Russian Federation and federal districts]. Available at: https://rosstat.gov.ru/storage/mediabank/IVBO_sub_RF_08-2025.xlsx.
- Kolichestvo aktivnykh predpriyatiy [Number of active enterprises]. Available at: https://rosstat.gov.ru/storage/mediabank/demb-2024.xlsx.
- Kredity, predostavlennyye fizicheskim litsam – rezidentam (regional’nyy razrez) [Loans provided to resident individuals (regional breakdown)]. Available at: https://cbr.ru/vfs/statistics/BankSector/Mortgage/02_05_Debt_ind.xlsx.
- Kredity, predostavlennyye yuridicheskim litsam – rezidentam i individual’nym predprinimatelyam (regional’nyy razrez) [Loans provided to resident legal entities and individual entrepreneurs (regional breakdown)]. Available at: https://cbr.ru/vfs/statistics/BankSector/Loans_to_corporations/01_03_Loans_corp_20250101.xlsx.
- Srednemesyachnaya nominal’naya nachislennaya zarabotnaya plata rabotnikov v tselom po ekonomike po sub”yektam Rossiyskoy Federatsii za 2000–2024 gg. [Average monthly nominal accrued wages of workers in the economy as a whole by constituent entities of the Russian Federation for 2000–2024.]. Available at: https://rosstat.gov.ru/storage/mediabank/tab4-zpl_2024.xlsx.
- Statistika territorial’nogo prisutstviya deystvuyushchikh kreditnykh organizatsiy i ikh podrazdeleniy [Statistics of the territorial presence of existing credit institutions and their divisions]. Available at: https://cbr.ru/vfs/statistics/bank_system/DivisionsPerRegion_010125.xlsx.
- Chislennost’ naseleniya sub”yekta Rossiyskoy Federatsii v srednem za 2024 god [Average population of a constituent entity of the Russian Federation for 2024]. Available at: https://rosstat.gov.ru/storage/mediabank/pok_monitor_2024_68.rar.
- Ardan D. Optimization of the BANK’s Branch Network Using Machine Learning Methods. In: Arai, K. (eds) Intelligent Systems and Applications. IntelliSys 2022. Lecture Notes in Networks and Systems. Springer, Cham, 2023, vol 542. DOI: 10.1007/978-3-031-16072-1_38.
- Hernik J., Linke H.J., Krol K. et al. Artificial Intelligence and Geodata for Local Community Sensibilization to Sustainable Spatial Development. In: Soliman K.S., eds. Artificial intelligence and Machine Learning. IBIMA-AI 2023. Communications in Computer and Information Science, 2024, vol. 2101. DOI: 10.1007/978-3-031-62843-6_4.
- Kerzina E., Antineskul E., Patrusheva L. Geomarketing in the Placement of Retail in the Development of a Million-Plus City (on the Example of a Regional Retail in Russia, Perm). In: Isaeva E., Rocha Á., eds. Science and Global Challenges of the 21st Century – Innovations and Technologies in Interdisciplinary Applications. Perm Forum 2022. Lecture Notes in Networks and Systems. Springer, Cham, 2023, vol 622. DOI: 10.1007/978-3-031-28086-3_68.
About authors
- Arkadeva Olga G.
- Candidate of Economics Sciences, Associate Professor, Department of Finance, Credit and Economic Security, Chuvash State University, Russia, Cheboksary (knedlix@yandex.ru; ORCID: https://orcid.org/0000-0003-4868-2365)
- Petrova Mariya A.
- Senior Risk Analyst, Underwriting Center in Cheboksary, VTB Bank (JSPC), Russia, Cheboksary (mashapetrova2002@gmail.com; )
Article link
Arkadeva O.G., Petrova M.A. Optimizing the placement of a commercial bank’s branch network [Electronic resource] // Oeconomia et Jus. – 2026. – №3. P. 1-28. – URL: https://oecomia-et-jus.ru/en/single/2026/3/1/. DOI: 10.47026/2499-9636-2026-3-1-28.
Copyright holder: authors. License: Attribution 4.0 International (CC BY 4.0)
