The geometallurgical models that predicting the throughput/comminution performance of the a processing plant often rely on rock hardness models, which are based on very sparse information, and are not typically are not designed to interact with mine production scheduling, and while ignoringignore the non-additive nature of hardness. This article presents a novel approach to integrate a geometallurgical throughput prediction model for the ball mill into short-term stochastic production scheduling in industrial mining complexes. The utilized datasets for this prediction model include penetration rates from blast-hole drilling, truck cycle data and measured throughput rates of the operating ball mill, offering an easily accessible and cost-effective method compared to other geometallurgical programs. Firstly, the comminution behavior of the mineral reserve is geostatistically simulated by building additive hardness proportions using blast-hole penetration rates. Then, a material tracking approach considers all material movements in the mining complex to inform the throughput prediction model about rock properties of blended materials sent to the ball mill. Subsequently, a multiple regression model is constructed, which predicts throughput rates as a function of blended rock properties. Finally, the prediction model is integrated into a stochastic integer programming model for short-term production scheduling in mining complexes. A case study at the Tropicana Gold Mining Complex in Western Australia shows that ball mill throughput can be predicted with an error of less than 30 t/h (RMSE) and a correlation coefficient between predicted and observed values of up to 0.8. By integrating the prediction model and newly proposed stochastic components into the optimization, the resulting production schedule can achieve weekly planned production reliably because scheduled materials can now be matched with the predicted performance of the ball mill. Comparisons to optimization using conventional mill tonnage constraints reveal that expected production shortfalls of up to 7% per period can be mitigated this way.