Supply Risk-Aware Alloy Discovery and Design
Materials design is a critical driver of innovation, yet overlooking the technological, economic, and environmental risks inherent in materials and their supply chains can lead to unsustainable and risk-prone solutions. To address this, we present a novel risk-aware design approach that integrates Supply-Chain Aware Design Strategies into the materials development process. This approach leverages existing language models and text analysis to develop a specialized model for predicting materials feedstock supply risk indices. To efficiently navigate the multi-objective, multi-constraint design space, we employ Batch Bayesian Optimization (BBO), enabling the identification of Pareto-optimal high entropy alloys (HEAs) that balance performance objectives with minimized supply risk. A case study using the MoNbTiVW system demonstrates the efficacy of our approach in four scenarios, highlighting the significant impact of incorporating supply risk into the design process. By optimizing for both performance and supply risk, we ensure that the developed alloys are not only high-performing but also sustainable and economically viable. This integrated approach represents a critical step towards a future where materials discovery and design seamlessly consider sustainability, supply chain dynamics, and comprehensive life cycle analysis.
Code (0)
등록된 구현이 없습니다.
Tasks
Bayesian OptimizationNavigateMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
MATAI: A Generalist Machine Learning Framework for Property Prediction and Inverse Design of Advanced Alloys
The discovery of advanced metallic alloys is hindered by vast composition spaces, competing property objectives, and real-world constraints on manufacturability. Here we introduce MATAI, a generalist machine learning fra…
Multi-Task LearningPhysics-Informed Gaussian Process Classification for Constraint-Aware Alloy Design
Alloy design can be framed as a constraint-satisfaction problem. Building on previous methodologies, we propose equipping Gaussian Process Classifiers (GPCs) with physics-informed prior mean functions to model the bounda…
Active LearningFrom Phase Prediction to Phase Design: A ReAct Agent Framework for High-Entropy Alloy Discovery
Discovering high-entropy alloy (HEA) compositions that reliably form a target crystal phase is a high-dimensional inverse design problem that conventional trial-and-error experimentation and forward-only machine learning…
Constructing material network representations for intelligent amorphous alloys design
Designing high-performance amorphous alloys is demanding for various applications. But this process intensively relies on empirical laws and unlimited attempts. The high-cost and low-efficiency nature of the traditional …
Autonomous Multi-objective Alloy Design through Simulation-guided Optimization
Alloy discovery is constrained by vast compositional spaces, competing objectives, and prohibitive experimental costs. Although simulations and machine learning have each accelerated parts of this process, unifying scien…