Researchers at the U.S. Bureau of Mines have developed an approach to tuning empirical computer models that is substantially more efficient than some traditional approaches. The mineral processing industry relies heavily on empirical computer models for both design and process control applications. Unfortunately, the empirical constants associated with the computer models have traditionally been selected using either statistical methods which are quite limited in some domains or trial-and-error procedures which are quite time consuming. A new and more efficient approach to selecting empirical constants for computer models has been developed and implemented. This approach employs a genetic algorithm for selecting the empirical constants for computer models of mineral processing equipment. Genetic algorithms are search algorithms based on the mechanics of natural genetics. They rapidly locate near optimum solutions in difficult search spaces and are shown to be effective in the search for the constants associated with empirical computer models. Bureau researchers have used these innovative search algorithms to both dramatically reduce the time needed to select empirical constants for computer models and to substantially improve the accuracy of the resulting models. The effectiveness of the approach is demonstrated with several examples from the field of mineral processing.