For environmental sustainability and energy security, renewable sources must be incorporated into sustainable energy solutions. Machine learning (ML) techniques are explored in this study to optimize the adoption of renewable energy sources in Bangladesh. Specifically, it proposes a three-phase methodology: (1) forecasting demand for nonrenewable energy, (2) predicting renewable energy availability and costs, and (3) analyzing potential savings and environmental benefits. Utilizing decision trees and random forests, this study presents a comparative analysis of energy demand and cost predictions, contributing to a data-driven framework for energy transition. The results indicate that strategic adoption of renewable energy can mitigate Bangladesh’s electricity shortages while reducing dependency on fossil fuels. Machine learning plays a crucial role in energy optimization by accurately forecasting energy demand and availability, allowing for better resource allocation. It helps identify patterns and trends in energy consumption, enabling more efficient integration of renewable sources. By using techniques like decision trees and random forests, machine learning models can optimize energy production and distribution, ultimately leading to more sustainable and cost-effective energy systems.The findings provide policymakers and energy planners with insights to enhance sustainability efforts.