Abstract
Nowadays numerical simulations are widely adopted in the industrial world in order to design and optimize the production process. In the design stage numerical simulations allow the optimal design of the dies and the determination of a suitable combination of nominal process parameters for the production. This approach has two significant limits: 1) the numerical model is a simplified representation of reality and some parameters of the model should be adjusted in order to fit prediction with the manufacturing results and, 2) the FE methodology is intrinsic deterministic and therefore it doesn’t take into account fluctuations of process conditions. The present paper proposes a new approach able to perform a robust optimization of a process, based on the integration of numerical modelling, response surface methodology and stochastic simulations in a sort of integrated environment, the Virtual Prototyping Environment (VPE). The kernel of this environment is a realistic numerical model of the process. The proposed tuning procedure limits the experimental activity at the injection moulding machine to a short utilization without any die (discharging tests). To clarify the proposed methodology, the paper details its application to the injection moulding process of a cover of an electric pump/motor (PA66 with glass fibres). The VPE has been applied to the dimensions of the moulded part which present some critical tolerances due to sealing and assembly needs.