Development of the Integrated Rural Access Index (iRAI) using Remote Sensing and GIS

Abstract. This study proposes an improvement to the measurement of Sustainable Development Goal (SDG) Indicator 9.1.1 which quantifies the proportion of the rural population living within 2 kilometers of an all-season road through the development of the Integrated Rural Access Index (iRAI). While the conventional Rural Access Index (RAI) measures solely road proximity, it fails to incorporate access to essential services. To address this limitation, the iRAI integrates the proximity to three critical facilities – healthcare, education, and service – using data from the Philippine Statistics Authority’s Community-Based Monitoring System. Gridded population datasets from WorldPop and LandScan were evaluated to determine suitability for computing RAI and iRAI. WorldPop demonstrated superior performance, with a lower mean absolute error (MAE = 18.60), root mean square error (RMSE = 36.41), and a higher correlation coefficient (r = 0.94), and was thus used for calculating both RAI and iRAI. In the provinces of Cavite and Batanes, Sentinel-2 imagery and Maximum Likelihood Classification (MLC) were utilized to delineate rural boundaries. Error metrics (MAE = 0.52, RMSE = 0.65) show strong agreement and practical interchangeability between conventional and satellite-based RAI/iRAI values. Regression analyses revealed that the iRAI exhibited stronger explanatory power (R² = 0.4878) and statistical significance (p = 0.0396) compared to RAI. A scenario simulation suggested that improvements in poverty, education, and healthcare access could result in a perfect iRAI score. Additionally, a network-based methodology was developed and applied in three areas to overcome the limitations of planar distance, demonstrating the potential for more accurate, household-level accessibility assessments.