Evolutionary programming to adjust the process in high precision foundries

Foundry is one of the activities that has contributed to evolve the society; however, the manufacturing process is carried out in the same manner as it was many years ago. Therefore, several defects may appear in castings when the production process is already finished. One of the most difficult defect to detect is the microshrinkage: tiny porosities that appear inside the casting. Another important aspect that foundries have to control is the attributes that measure the faculty of the casting to withstand several loads and tensions, also called mechanical properties. Both cases need specialised staff and expensive machines to test the castings and, in the case of mechanical properties, destructive inspections that render the casting invalid are also required. In our previous research, we have modelled the foundry process to apply machine-learning techniques to foresee these defects before making the castings. Nevertheless, these approaches do not correct what is happening in the plant. In this paper, we extend our approach proposing a genetic algorithm that employs meta-classifiers and cosine similarity measure to find the optimal parameters that allow us to change the foundry production, avoiding the aforementioned defects on-line.