AI-driven forecasting of rural development in India: a deep learning case study on the gram panchayat development plan

This research explores applying deep learning models to forecast the progress of the Gram Panchayat Development Plan (GPDP), a vital initiative aimed at the holistic development of rural areas in India. The study focuses on two key performance indicators: the number of facilitators registered and the number of Sabha scheduled, using data collected from the GPDP website spanning from September 1st, 2020, to June 2nd, 2024. The dataset covers five states: Uttar Pradesh, Maharashtra, Madhya Pradesh, Gujarat and Andhra Pradesh. Comprehensive preprocessing steps, including hourly interpolation, differencing, time series decomposition, and scaling, were employed to ensure high-quality data for model training. We applied a range of deep learning models, such as Recurrent Neural Networks (RNN), Long Short-Term Memory networks (LSTM), Bidirectional LSTM (BiLSTM), Gated Recurrent Units, and several hybrid models. The models were evaluated based on Root Mean Squared Error, Mean Absolute Error, Mean Squared Error, and the coefficient of determination (R2). The study identified the RNN model as the best performer, achieving the highest average Friedman ranking across all metrics and states. Furthermore, we extended our forecast to May 28th, 2025, using the RNN model, providing valuable projections for the GPDP's future progress. The findings of this research offer critical insights for policymakers and administrators, enabling data-driven decision-making to enhance the planning and implementation of rural development initiatives. This study not only highlights the efficacy of deep learning models in time series forecasting but also contributes to the sustained success and improvement of the GPDP.