Enhancing Experimental Efficiency in Materials Design: A Comparative Study of Taguchi and Machine Learning Methods

2025-06-04

AI Summary

Multi-parameter optimization in additive manufacturing processes faces challenges including prohibitively high costs of full-factorial experiments and the inability of conventional orthogonal designs (e.g., Taguchi methods) to capture nonlinear process–property relationships. To address weld-bead geometry prediction (penetration, width, height) in Wire Arc Additive Manufacturing (WAAM), this work introduces, for the first time, an uncertainty-driven active learning framework based on Gaussian Process Regression (GPR). The approach integrates Latin Hypercube Sampling for initial design and iteratively refines the experimental dataset using predictive uncertainty–guided sampling. Compared against a Taguchi L25 orthogonal array, the GPR-based strategy achieves a 42% reduction in prediction error and requires 36% fewer experiments across 15 test cases. By relaxing the linearity assumption inherent in orthogonal designs, the proposed method significantly enhances both modeling accuracy and sampling efficiency in multi-parameter process optimization.

Technology Category

• Machine Learning: Active Learning
• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Materials design problems often require optimizing multiple variables, rendering full factorial exploration impractical. Design of experiment (DOE) methods, such as Taguchi technique, are commonly used to efficiently sample the design space but they inherently lack the ability to capture non-linear dependency of process variables. In this work, we demonstrate how machine learning (ML) methods can be used to overcome these limitations. We compare the performance of Taguchi method against an active learning based Gaussian process regression (GPR) model in a wire arc additive manufacturing (WAAM) process to accurately predict aspects of bead geometry, including penetration depth, bead width, and height. While Taguchi method utilized a three-factor, five-level L25 orthogonal array to suggest weld parameters, the GPR model used an uncertainty-based exploration acquisition function coupled with latin hypercube sampling for initial training data. Accuracy and efficiency of both models was evaluated on 15 test cases, with GPR outperforming Taguchi in both metrics. This work applies to broader materials processing domain requiring efficient exploration of complex parameters.

Problem

• Optimizing multiple variables in materials design efficiently
• Overcoming limitations of Taguchi method with machine learning
• Comparing Taguchi and GPR models for bead geometry prediction

Innovation

• Machine learning overcomes Taguchi's non-linear limitations
• Gaussian process regression with active learning
• Latin hypercube sampling for initial training data