Material synthesis through simulations guided by machine learning: a position paper

Abstract

In this position paper, we propose an approach for sustainable data collection in the field of optimal mix design for marble sludge reuse. Marble sludge, a calcium-rich residual from stone-cutting processes, can be repurposed by mixing it with various ingredients. However, determining the optimal mix design is challenging due to the variability in sludge composition and the costly, time-consuming nature of experimental data collection. Also, we investigate the possibility of using machine learning models using meta-learning as an optimization tool to estimate the correct quantity of stone-cutting sludge to be used in aggregates to obtain a mix design with specific mechanical properties that can be used successfully in the building industry. Our approach offers two key advantages: (i) through simulations, a large dataset can be generated, saving time and money during the data collection phase, and (ii) Utilizing machine learning models, with performance enhancement through hyper-parameter optimization via meta-learning, to estimate optimal mix designs reducing the need for extensive manual experimentation, lowering costs, minimizing environmental impact, and accelerating the processing of quarry sludge. Our idea promises to streamline the marble sludge reuse process by leveraging collective data and advanced machine learning, promoting sustainability and efficiency in the stonecutting sector.

Figures (4)

• Figure 1: The stone-cutting sludge problem overview. The stone-cutting processes of marble and similar materials produce some byproducts, such as stone-cutting sludge. These byproducts come with a high cost for the environment and industries. They can be disposed of as waste, leading only to a cost for the industry, or be re-used to create mix-designs. Traditionally, the reuse is subject to trial & error approaches, involving a high cost and requiring a lot of time. Instead, we argue that it is possible to use machine learning-driven simulations to predict optimal mix-designs. In this way, it is possible to obtain products from byproducts that are considered waste, thus saving time, increasing sustainability, and reducing the costs for the industry.
• Figure 2: The end-to-end workflow of a simulator based on the machine learning model development and deployment process. It is divided into four main phases: Preprocessing, Training, Testing, and Simulation, with the ultimate goal of producing simulations based on material characteristics. The circle represents the data, while the rectangle represents the process.
• Figure 3: The plot shows the performance of different models on training and testing sets, respectively. The x-axis represents the true values, while the y-axis represents the predictions made by the model without meta-learning. The blue points are individual data points, and the red dashed line represents the ideal line where the predictions perfectly match the true values.
• Figure 4: The plot displays the best model's performance, with meta-learning through Bayesian Optimization.