AI-driven optimization in 3D printing: Reducing time and material consumption through machine learning models
Phuong Dong Nguyen, Manh Dao, Thanh Q. Nguyen
Abstract
This paper investigates the impact of various 3D printing parameters on two critical resources: printing time and plastic material consumption. Through a series of experiments, the paper aimed to optimize these resources by adjusting the relevant parameters. First, by incorporating multiple algorithmic models, our stacking artificial intelligence model achieved superior prediction accuracy and a more comprehensive evaluation of the parameters. This novel approach amalgamates machine learning models to identify the optimal configuration or leverage the strengths of various models. Second, the proposed method paves the way for developing automated tools that simplify the configuration of the printer parameters, making 3D printing more accessible. This innovation allows users with limited expertise to focus on essential aspects of 3D printing, reducing the process to a few elementary steps. Third, our findings enable 3D printing entities to offer advanced estimates of printing time and filament usage, particularly for large-scale projects. With a prediction accuracy of 96.19% based on the R 2 score, the proposed method enables users to plan uninterrupted printing sessions more effectively and optimize both print time and resource consumption according to their specific preferences. Lastly, mastering the printing parameters that affect time and material consumption is crucial to producing high-quality 3D models efficiently. This optimization reduces plastic waste and environmental pollution while increasing profits and operational efficiency. In general, through these contributions, our study advances the field of 3D printing by providing practical tools and insights that enhance both the efficiency and accessibility of the technology.