International Journal for Advanced Research in Science & Technology

Artificial Intelligence (AI) applications in Smart Agricultural Facilities (SAF) often face challenges related to explainability, limiting farmers' ability to fully utilize their capabilities.This study addresses this gap by proposing a model that integrates eXplainable Artificial Intelligence (XAI) with Predictive Maintenance (PdM).The model aims to offer predictive insights and explanations across four key dimensions: data, model, outcome, and end-user.This approach represents a paradigm shift in agricultural AI, transforming how these technologies are interpreted and applied.The proposed model outperforms existing approaches, demonstrating notable improvements in performance.Specifically, the Long-Short-Term Memory (LSTM) classifier shows a 5.81% improvement in accuracy, while the eXtreme Gradient Boosting (XGBoost) classifier achieves a 7.09% higher F1 score, a 10.66% boost in accuracy, and a 4.29% increase in Receiver Operating Characteristic-Area Under the Curve (ROC-AUC).These results suggest that the model can enhance maintenance predictions in real-world agricultural settings.Additionally, this study offers valuable insights into data purity, both global and local explanations, and counterfactual scenarios in the context of PdM for SAF.By emphasizing explainability alongside traditional performance metrics, the study advances AI applications in agriculture.It also encourages future research in areas like multi-modal data integration and the implementation of Human-in-the-Loop (HITL) systems, which can improve AI effectiveness while addressing ethical considerations such as Fairness, Accountability, and Transparency (FAT) in agricultural AI.