Statistical Modeling and Optimization of Dimensional Accuracy for Solidified Polylactic Acid Parts Fabricated by Fused Filament Modeling Process

Abstract

3D Printing is a layered manufacturing process that builds prototypes by depositing material in layered form using heaters. Prototypes made by 3D Printing are widely used in product development as they can be used for product testing. 3D Printing prototypes should have a very good dimensional accuracy for functional performance as well as aesthetics. The dimensional accuracy in 3D printing depends upon different process parameters, namely Layer Thickness, Nozzle Diameter, Part Bed Temperature, Speed of Deposition, Raster Angle of Deposition, Raster Width and Length of the parts. In this present work, an attempt has been made to improve the dimensional accuracy of prototypes of Solidified PolyLactic acid parts fabricated using Fused Filament Modelling process of 3D Printing. Experiments have been performed according to Central Composite Rotatable Design (CCRD) considering four parameters namely layer thickness, nozzle diameter, part bed temperature, and length of parts at three levels. Two different material laying, x direction and y direction have been used in the study. Empirical statistical models have been developed for predicting the dimensional accuracy of the parts in both the directions of laying. Analysis of variance (ANOVA) has been used to test the significance of process variables on dimensional accuracy. In case of x direction laying, for shrinkage along the length, length of part, layer thickness and nozzle diameter significantly affects the shrinkage. Further, for shrinkage along the width, nozzle diameter, layer thickness and part bed temperature are significant parameters and for shrinkage along the height, nozzle diameter and layer thickness are most significant parameters. Similarly, in y direction laying, for shrinkage along the length, length of part, part bed temperature and layer thickness significantly affects the shrinkage. Further, for shrinkage along the width, layer thickness and nozzle diameter are significant parameters and for shrinkage along the height, layer thickness and part bed temperature are most significant parameters. Optimization of the dimensional accuracy for x direction laying and y direction laying have been done using trust region based MATLAB technique and confirmation of statistical model have been done by performing experiments at different parameters other than experiments in DOE and the results were found to be satisfactory.