AI-Powered Public Administration: Enhancing Policy Implementation and Service Delivery through Intelligent Decision Support Systems

Abstract

States derive economic development, service delivery, and public trust from effective governance systems, processes, and practices, but many countries persistently struggle with institutional and governance failures. This paper presents an AI-enabled decision-making approach for evaluating and enhancing public administration quality based on country policy and institutional assessment (CPIA) indicators produced by the World Bank. With the CPIA Quality of Public Administration (PADM) score being the dependent variable, the paper uses an econometric approach combined with machine learning methodologies to reveal main governance determinants and model-based predictive trends. This analysis employs unbalanced panel of 1,152 country–years data for low-and lower-middle income countries. Methodology-wise, the analysis combines OLS, Ridge, and Lasso regression with supervised machine learning techniques, among which are Logistic Regression (with SMOTE), Random Forest, and Gradient Boosting, as well as Support Vector Machines and unsupervised clustering. Model performance metrics include R², RMSE, accuracy, precision, recall F1-score and ROC–AUC. Results are robust to specify the model and show that public sector management and institutional quality are the decisive determinants of PADM across models. Transparency, property rights, fiscal policy, and tax revenue all have positive contributions to support the main one. That is, Logistic Regression with SMOTE and Gradient Boosting has the best classification performance (AUC up to 0.94), while clustering can help identify unique governance profiles to make differential policy strategies. Intel's AI-enriched analytics can enrich traditional governance assessments by advancing prediction, prioritization, and benchmarking. In general, the research demonstrates the potential of AI-based decision-support to inform more focused data-driven public-sector reforms and better policies implementations and service delivery.