Digitization Of GIS-based Poverty Level Mapping and Prediction in Sukabumi Regency

Abstract

Poverty in Sukabumi Regency remains a significant development challenge, with the poverty rate recorded at 6.87% (approximately 175,930 people) in 2024 (BPS). Poverty data traditionally presented in tabular form make it difficult for local governments to identify priority areas and project future poverty growth. This study develops a digitalized poverty mapping and prediction system using Geographic Information Systems (GIS) across 47 sub-districts in Sukabumi Regency, using secondary population and poverty data processed through spatial interpolation (ordinary kriging) to generate poverty distribution maps, along with linear regression to examine the relationship between population size and the number of poor residents. Results show actual poverty percentages ranging from 14.11% to 18.09% (average 15.79%), with all sub-districts falling into the “Medium” poverty status category. The kriging model achieved high accuracy (RMSE = 0.1056%; MAPE = 0.148%), while linear regression revealed a very strong, significant relationship between population size and the number of poor residents (R² = 0.9945; p < 0.001). The next-year poverty prediction remains a baseline model based on fixed growth-rate assumptions; results should therefore be read as indicative directions of change, requiring further refinement using time-series data and additional socioeconomic variables. The resulting interactive GIS system integrates maps of actual poverty, poverty change, poverty status, and prediction into a single platform, serving as a decision-support tool for local governments in planning region-based poverty alleviation programs.