Public policy implementation often struggles with uneven budget allocation across sectors and regions, leading to inefficiencies in resource use. This study presents a data-driven framework for smart budget allocation through predictive analytics and optimization methods. The proposed model helps policymakers ensure fair and efficient distribution of public funds by integrating socioeconomic indicators, sector-specific requirements, and past expenditure outcomes. Using linear regression forecasting combined with constrained linear programming, the framework determines sector-wise budgets. The analysis focuses on five crucial public sectors—Agriculture, Health, Education, Rural Development, and Road Transport—using data from 2020 to 2025, along with macroeconomic indicators such as GDP growth, unemployment rates, and simulated public sentiment. A custom interactive dashboard enables real-time visualization and engagement with predicted and optimized budgets. Evaluation results highlight the potential of blending machine learning with operations research for evidence-based governance. The study introduces a scalable and reproducible model that aligns with national missions like India@2047 and Swarna Andhra Pradesh. By embedding data science into fiscal decision-making, this work contributes to advancing digital governance, improving transparency, and fostering citizen-centric planning.
Kaki et al. (Mon,) studied this question.