Development of location suitability prediction for health facilities using random forest machine learning in 2030 integrating remote sensing and GIS in West Java, Indonesia

• The improvement of health facilities is needed in line with population growth. • The prediction for priority location for health facilities in the future. • The utilization of machine learning and GIS for health facilities site suitability. Access to healthcare facilities is crucial in the present day. Healthcare facilities must be proportional to the population in a given area. Therefore, it is important to increase the number of healthcare facilities in regions where there is an imbalance. Proper planning and a sustainability review are necessary when determining locations for healthcare facilities. Environmental changes can affect the suitability of a location in the future. Thus, planning that can predict future suitability conditions is required to ensure that the built locations have high sustainability. This study predicts the suitability of healthcare facility locations in 2030 in West Java using remote sensing and Geographic Information System (GIS). Both static and dynamic data processed at 30×30 meter intervals across West Java will be used. Geospatial and remote sensing data are utilized in the study. Dynamic parameter extrapolation uses data from 2000 to 2018. The random forest machine learning method is employed to obtain the suitability values for healthcare facility locations using existing health facility training data in West Java. The results show changes in the suitability classes of healthcare facilities in each region from 2018 to 2030, with some areas experiencing an increase or decrease in class. This research highlights consistently suitable locations, ensuring their sustainability as healthcare facility sites.