In this competition, participants must predict the **percentage of Silica** in the final product of a flotation plant used to concentrate iron ore. The goal is to develop a model that forecasts iron concentration based on operational data, enabling real-time decision-making. Accurate predictions help engineers optimize the process, reducing the wait for lab results that take over an hour. Participants will use features like iron feed percentage, reagent flows, and pulp characteristics to create their models. This challenge focuses on improving operational efficiency in mining plants through predictive modeling.