| Abstract Scope |
Lithium is a critical mineral required for batteries, renewable energy technologies, and electric vehicles. Optimizing lithium recovery from hydrometallurgical processes often requires extensive experimental studies involving multiple operating parameters. This work investigates the use of explainable machine learning models to predict lithium recovery efficiency using process variables reported in the literature, including leaching temperature, time, reagent concentration, and solid-to-liquid ratio. Several machine learning algorithms are evaluated and compared for recovery prediction performance. To improve model transparency and support process optimization, explainable artificial intelligence techniques are employed to identify the most influential process parameters affecting lithium recovery. The proposed approach aims to reduce experimental effort, accelerate process development, and provide data-driven insights for hydrometallurgical operations. Results demonstrate the potential of explainable machine learning as a decision-support tool for the efficient extraction and processing of critical minerals. |