Poultry farming is a vital contributor to global food production. Optimal Feed Conversion Ratio (FCR) is essential to achieve profitability and sustainability in broiler chicken rearing. Leveraging machine learning (ML) algorithms, this study analyzed environmental factors, feed-related factors, and broiler growth patterns and developed predictive models for FCR optimization. In the first phase of the study, three ensemble models, such as decision tree regressor, XGBoost, and artificial neural networks (ANN), are used for predicting FCR. Performance measures such as R2, root mean square error, mean absolute error, and mean absolute percentage error were used to evaluate the models. It was found that ANN provided better accuracy. In the second phase, the dataset is divided into 5 sets based on the feed type given to birds in their life cycle, and ANN is applied to the five different datasets. The findings highlight the adverse effects of heat stress, especially temperature, on FCR and growth, and emphasize the role of feed composition and environment management in mitigating these impacts. By integrating real-time data on temperature, humidity, feed intake, and broiler physiology, the study demonstrates ML’s potential to provide actionable insights for adaptive interventions and nutrient optimization.