An ensemble approach of machine learning in evaluation of mechanical property of the rapid prototyping fabricated prototype

Rapid prototyping (RP) is a promising product development technology due to its unique characteristic of fabricating functional products timely and efficiently. Fused deposition modelling (FDM) process based on RP technology is used in industries for prototype fabrication and its properties testing. The properties of the RP fabricated prototypes such as wear strength, tensile strength, dimensional accuracy, etc. depends on the parameter settings of the RP machines. For selecting the appropriate parameter settings, various mathematical models developed based on physics and data can be formulated. In the present work, we introduced an ensemble method of genetic programming (GP) and artificial neural network for formulating a model for predicting the wear strength of the FDM fabricated prototype. The results indicate that ensemble model have performed better than that of the standardised GP, which may be then used by experts for optimising the performance of the FDM process.