Data-Driven Decision Support for Sustainable Animal Environment and Welfare Management

Abstract

Sustainable animal environment management requires continuous assessment of environmental conditions, animal welfare, and resource utilization to support timely and effective management decisions. This study proposes a Data-Driven Decision Support Framework for Sustainable Animal Environment and Welfare Management that integrates environmental monitoring, animal-centered indicators, machine learning-based prediction, sustainability assessment, and actionable management recommendations. The framework considers environmental variables including temperature, relative humidity, air quality, ventilation, light intensity, feed and water availability, together with behavioral and physiological indicators representing animal welfare. Heterogeneous monitoring data are processed through data cleaning, normalization, temporal synchronization, and feature engineering to construct an integrated representation of animal environmental conditions. Five machine learning algorithms, namely Logistic Regression, Decision Tree, Support Vector Machine, Random Forest, and Gradient Boosting, are comparatively evaluated using accuracy, precision, recall, F1-score, and ROC-AUC. The experimental evaluation demonstrates that Gradient Boosting provides the strongest overall predictive performance, achieving 94.3% accuracy, 93.8% precision, 93.5% recall, 93.6% F1-score, and a ROC-AUC of 0.96. The predicted welfare conditions are further combined with environmental suitability and resource-efficiency indicators to categorize operating conditions into normal, moderate-risk, and high-risk levels. Based on the identified risk level, the decision-support component generates management recommendations related to ventilation, cooling, water availability, feeding conditions, environmental control, and animal monitoring. The proposed framework consequently transforms heterogeneous monitoring data into predictive welfare information and actionable decisions, providing a systematic approach for proactive animal welfare management while supporting environmental sustainability and efficient resource utilization.