The purpose of this work is to develop an experiment applied to milling process cutting parameters optimization based on data collection from the factory shop floor. The cutting parameters to be optimized were cutting speed, feed rate and depth of cut. Hardened steel dies currently used for forging process were milled to generate data to be used during the experiments. The optimization focus was to minimize the milling process cost or to increase it’s the productivity. Two stages had been used: one based in Design of Experiment (DOE) technique and another based on a deterministic method. Both were used to optimize cutting condition parameters to be applied instead of that being used before the experiments. It was possible to conclude, following DOE method results that were recommended to use the smaller cutting speed, the smaller feed rate and the greatest depth of cut, considering the cutting parameters tested. These results were considered by the authors as initial values to apply a deterministic method. The greatest depth of cut and the smaller feed rate were compatible with the machining conditions of the part, considering the initial blank and surface finish required respectively. So, deterministic method was used only to optimize cutting speed. As final result, the authors conclude that minimum cost cutting speed, greatest around 10% to that one found by DOE method was much more convenient to be applied in shop floor, because the smaller cost involved. Finally, this work allowed also concluding that both DOE and deterministic methods can be used to optimize cutting process parameters taking a small number of results collected from factory shop floor during the process evolution.
Study on high-speed cutting parameters optimization of AlMn1Cu based on neural network and genetic algorithm
