Site Selection for Elderly Care Facilities in the Context of Big Data: A Case Study of Xi’an, China

Abstract

The accelerated aging of China’s population has made the optimization of elderly care facility locations a critical priority. The field of big data presents innovative approaches for determining the optimal site selection for such facilities. This study uses Xi’an City, in Shaanxi Province, China, as a case study to explore how big data and the ID3 decision tree model can enhance the optimization of elderly care service facility locations. The study begins with a comprehensive analysis of Xi’an’s aging demographics, focusing on the current aging trends and the distribution characteristics of existing elderly care facilities. Utilizing the Baidu Map API, the study collected Point of Interest (POI) data for Xi’an, which were spatially analyzed using ArcGIS 10.8 software to identify the distribution patterns of elderly care facilities and their relationships with other public amenities. The ID3 algorithm was then employed to construct a decision tree model to simulate and predict optimal sites for elderly care facilities in Xi’an. By classifying and filtering POI data and dividing Xi’an into 500 m × 500 m grid units, the model was trained and validated, achieving an accuracy of 85.8%. The findings suggest that suitable sites for elderly care facilities in Xi’an should prioritize proximity to government offices and medical institutions, which would better address the healthcare needs of the elderly population. The application of the ID3 algorithm in planning the locations of elderly care facilities helps mitigate human biases and provides valuable insights for the planning of other public amenities.