Authors:
Kübra Selvi
1
;
Murat Tasyürek
1
and
Celal Öztürk
2
Affiliations:
1
Vocational School of Information Technologies, Kayseri University, Kayseri, Turkey
;
2
Faculty of Engineering, Erciyes University, Kayseri, Turkey
Keyword(s):
K-Means Clustering, Hierarchical Clustering, Spatial Data.
Abstract:
The location of water cashier offices is crucial in terms of both operational efficiency and citizens' easy access to payment points. This study aims to reduce the average distance covering the widest service area with the minimum number of cashier offices by using the K-Means and Hierarchical Clustering methods based on the geographical coordinates of independent sections in Altınordu district of Ordu province. Spatial analyses are playing an increasingly important role in urban planning, urban transformation and disaster management, and identifying regions with similar characteristics is of great value. The dataset contains the latitude and longitude information of the independent sections. The actual number of cashier offices in the Altınordu district of Ordu province is 3. The optimal number of clusters determined by the Elbow method was 5, while the optimal number of clusters found using dendrogram analysis was 8. In this context, clustering scenarios of 3, 5, and 8 were examine
d, and the performance of each algorithm was compared based on the average distance criterion. The analyses revealed that the K-Means algorithm provided the best average distance. The results demonstrate that the independent sections in Altınordu can be geographically clustered and that this clustering, taking into account settlement density and the current cashier distribution, can serve as a guide for cashier planning and resource allocation. This approach can guide the more effective placement of water cashier offices, thereby increasing service efficiency and accessibility for citizens.
(More)