Spatial Accessibility Index and Health Facility Location Suitability Based on POI and Machine Learning in DIY Province

Abstract

The unequal distribution of health facilities is a major cause of inequality in accessibility, especially in hard-to-reach rural areas. Therefore, equitable development of health facilities is needed, targeting both access and location suitability. This study aims to measure spatial accessibility and identify suitable locations for health facility development to support efficient, equitable, and targeted planning. The method used includes calculating the accessibility index using Enhanced Two Step Floating Catchment Area (E2SFCA) and location suitability modeling with machine learning algorithms, as well as analysis of the relationship between the two using scatterplots. Random forest demonstrated the best performance with an accuracy of 85.71%. Identification of the relationship between accessibility measurements and suitability modeling successfully identified 85 villages with low access but high suitability, which are recommended as priority locations for health facility development. These findings are expected to form the basis for evidence-based planning to achieve equitable distribution of health services in the Special Region of Yogyakarta Province.