Hiperparametrize edilmiş makine öğrenme algoritmaları ile iş yükü tahmini: Bankacılık sektöründe uygulaması.

Saved in:

Bibliographic Details
Title:	Hiperparametrize edilmiş makine öğrenme algoritmaları ile iş yükü tahmini: Bankacılık sektöründe uygulaması.
Alternate Title:	Workload prediction with hyperparametrized machine learning algorithms: Application in the banking sector.
Authors:	Balcı, Yunus Emre¹ yunus.balc6@ogr.sakarya.edu.tr, Demir, Alparslan Serhat² alparslanserhat@sakarya.edu.tr
Source:	Journal of the Faculty of Engineering & Architecture of Gazi University / Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi,. Mar2026, Vol. 41 Issue 1, p339-346. 9p.
Subjects:	Banking industry, Workforce planning, Forecasting, Cost functions, Digital transformation, Ensemble learning, Machine learning, Long short-term memory
Abstract (English):	The digitalizing world has led to fundamental transformations, particularly in the financial sector, where the increasing number of digital banks has emerged as both a threat and a catalyst for digitalization for branch- based banks. In this context, it is crucial for branch banks to manage major cost components--especially personnel expenses-effectively in order to gain a competitive advantage against digital banks. To maintain their market share and achieve sustainable growth, these banks must embrace digital transformation and perform accurate staffing planning. Optimizing staff levels is a strategic factor directly influencing employee satisfaction and customer experience. While understaffing increases workload and reduces satisfaction, overstaffing raises costs and may negatively affect employees' social lives due to possible relocations. This study examines the advantages of applying artificial intelligence techniques to staffing and resource management processes in a bank branch operating in the financial sector. Past performance data, customer demand, and other relevant factors for a specific job title group were analyzed, and workload forecasting was conducted using XGBoost and LSTM methods. The obtained results-0.99 MSE and correlation values between 0.99 and 0.91-show that XGBoost performs more effectively compared to LSTM. These findings contribute to cost optimization while enhancing both employee and customer satisfaction. [ABSTRACT FROM AUTHOR]
Abstract (Turkish):	Dijitalleşen dünya, özellikle finans sektöründe köklü değişimlere yol açmakta; artan dijital banka sayısı, şubeli bankalar için hem tehdit hem de dijital dönüşüm açısından önemli bir uyarıcı niteliği taşımaktadır. Bu dönemde şubeli bankaların, personel giderleri gibi büyük maliyet kalemlerini etkin yöneterek dijital bankalarla rekabette avantaj sağlaması kritik öneme sahiptir. Sektörde paylarını koruyabilmeleri ve sürdürülebilir büyüme elde edebilmeleri için dijitalleşmeyi benimsemeleri ve kadro planlamasını doğru şekilde yapmaları gerekmektedir. Kadro sayısının optimize edilmesi, çalışan memnuniyeti ile müşteri deneyimini doğrudan etkileyen stratejik bir faktördür. Eksik kadro iş yükünü artırarak memnuniyeti azaltabilirken, fazla kadro ise maliyetleri yükseltmekte ve yer değişiklikleri nedeniyle çalışanların sosyal yaşamını olumsuz etkileyebilmektedir. Bu çalışma, finans sektöründe faaliyet gösteren bir banka şubesinin kadro planlaması ve kaynak yönetimi süreçlerinde yapay zekâ tekniklerinin avantajlarını incelemektedir. Bir unvan grubuna ait geçmiş performans verileri, müşteri talepleri ve diğer etkenler analiz edilerek XGBoost ve LSTM metotlarıyla iş yükü tahmini yapılmıştır. Elde edilen 0.99 MSE ve 0.99-0.91 korelasyon değerleri, XGBoost'un LSTM'ye kıyasla daha etkili sonuçlar sunduğunu göstermiştir. Bu bulgular, maliyet optimizasyonu ile çalışan ve müşteri memnuniyetinin artırılmasına katkı sağlamaktadır. [ABSTRACT FROM AUTHOR]
	Copyright of Journal of the Faculty of Engineering & Architecture of Gazi University / Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi, is the property of Gazi University, Faculty of Engineering & Architecture and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	The digitalizing world has led to fundamental transformations, particularly in the financial sector, where the increasing number of digital banks has emerged as both a threat and a catalyst for digitalization for branch- based banks. In this context, it is crucial for branch banks to manage major cost components--especially personnel expenses-effectively in order to gain a competitive advantage against digital banks. To maintain their market share and achieve sustainable growth, these banks must embrace digital transformation and perform accurate staffing planning. Optimizing staff levels is a strategic factor directly influencing employee satisfaction and customer experience. While understaffing increases workload and reduces satisfaction, overstaffing raises costs and may negatively affect employees' social lives due to possible relocations. This study examines the advantages of applying artificial intelligence techniques to staffing and resource management processes in a bank branch operating in the financial sector. Past performance data, customer demand, and other relevant factors for a specific job title group were analyzed, and workload forecasting was conducted using XGBoost and LSTM methods. The obtained results-0.99 MSE and correlation values between 0.99 and 0.91-show that XGBoost performs more effectively compared to LSTM. These findings contribute to cost optimization while enhancing both employee and customer satisfaction. [ABSTRACT FROM AUTHOR]
ISSN:	13001884
DOI:	10.17341/gazimmfd.1496603