Additive manufacturing (AM) has revolutionized material fabrication by enabling the production of complex structures with enhanced design flexibility and material efficiency. However, the development of AM-specific materials remains a critical challenge due to the unique process characteristics of AM. Recent advancements in artificial intelligence (AI), for example, machine learning and deep learning, have emerged as powerful tools in accelerating material discovery, optimizing process parameters, and improving material performance for AM. This review provides a comprehensive overview of AI-driven material development for AM, focusing on metals, polymers, and bioinks/biomaterial inks. The discussion encompasses AI techniques applied to material development, including predictive modeling, generative algorithms, and intelligent optimization methods. Data collection and pre-processing methodologies for AI applications in AM are discussed. In addition, the applications of AI in material development in AM are also reviewed. Finally, the review highlights emerging trends, such as AI-driven high-throughput material screening, integration of AI with multiscale high-fidelity simulations, the use of digital twins for real-time process control, and active learning strategies for optimizing material compositions. By summarizing recent advancements and outlining future directions, this review provides insights into the evolving intersection of AI and AM, paving the way for more intelligent and efficient material development in the next generation of manufacturing.

Additive manufacturing (AM) has revolutionized modern fabrication by enabling complex geometries, material efficiency, and customized production. However, process variability, material inconsistencies, and defect formation remain critical challenges, limiting scalability and industrial adoption. Machine learning (ML) has emerged as a powerful tool to address these limitations by enabling data-driven optimization, defect detection, material property prediction, and real-time process control. This review provides a comprehensive analysis of ML applications in AM, spanning polymers, metals, ceramics, and carbon-based materials, with a focus on process optimization, quality assurance, and predictive modeling. Specifically, this review examines real-time defect detection through vision-based ML techniques, printing parameter optimization using supervised and reinforcement learning, and predictive modeling of material properties-laying the groundwork for deeper exploration of key methodologies such as deep learning and physics-informed models. Key ML methodologies, including deep learning, reinforcement learning, and hybrid physics-informed models, are explored in the context of enhancing print precision, mechanical performance, and functional properties. Despite significant advancements, challenges such as data scarcity, model generalization, and real-time integration into AM workflows persist. Emerging trends, including multimodal sensor fusion, in situ monitoring, and cloud-based predictive analytics, are discussed as potential pathways toward fully autonomous and intelligent manufacturing. By consolidating recent developments and outlining future directions, this review provides essential insights for researchers, engineers, and industry professionals looking to harness ML in AM, facilitating advancements in process efficiency, quality control, and overall manufacturing reliability.

Recent advancements in industrial artificial intelligence (AI) are reshaping the industry by driving smarter manufacturing, predictive maintenance, and intelligent decision-making. However, existing approaches often focus primarily on algorithms and models while overlooking the importance of systematically integrating domain knowledge, data, and models to develop more comprehensive and effective AI solutions. Therefore, the effective development and deployment of industrial AI require a more comprehensive and systematic approach. To address this gap, this paper reviews previous research, rethinks the role of industrial AI, and proposes a unified industrial AI foundation framework comprising three core modules: the knowledge module, data module, and model module. These modules help to extend and enhance the industrial AI methodology platform, supporting various industrial applications. In addition, a case study on rotating machinery diagnosis is presented to demonstrate the effectiveness of the proposed framework, and several future directions are highlighted for the development of the industrial AI foundation framework.

Laser powder bed fusion (LPBF) is one of the additive manufacturing (AM) techniques and the most studied laser-based AM process for metals and alloys. The optimization of the laser process parameters of LPBF and the prediction of defects, for example, keyholes, cracks, and lack of fusion (LOF), are important for improving the quality of products made with LPBF. Deep learning (DL) is powerful in analyzing complex processes and predicting anomalies; however, much data is generally required for training a DL model. Experimental studies on AM (e.g., LPBF) habitually employ the design of experiments to decrease the number of experiments and save time and costs. Hence, the experimental data are not prepared for DL model creation in most situations. This paper studies the creation of a DL model on a small experimental dataset with unbalanced data and the prediction of the LOF defect of LPBF utilizing the created DL model. Data analytics is mainly conducted based on four DL methods, including Elman neural networks, Jordan neural networks, deep neural networks (DNN) with weights initialized by the deep belief network, and the regular DNN based on four algorithms: “rprop+”, “rprop−”, “sag,” and “slr.” It is shown that the regular DNN after the z-score standardization of the small dataset helps create a more accurate DL model and achieve better analytics and prediction results than the three other DL methods in this paper. The three other DL methods do not work well in the prediction of LOF based on the small dataset (with unbalanced data).

Traditional fruit grading methods are mostly time-consuming and subjective, thereby limiting efficiency in the agricultural sector. To address these problems, this paper presents the design and implementation of an automated fruit sorting system for classifying certain fruits, namely oranges, tomatoes, and mangoes, using image processing and support vector machine (SVM) techniques. An ESP32 camera was used to capture images of the fruits, which were later passed through algorithms in Python. Extracted features were then fed into a SVM model for the classification process of fruits. The model demonstrated excellent performance, achieving an accuracy of 100%, a precision of 96%, a recall of 92%, and an F1 score of 89%. The results indicated that incorporating multiple features significantly increases the accuracy of the classification. Moreover, the performance was optimized by selecting an appropriate regularization parameter during the training of the model and the use of polynomial kernel functions. Finally, the whole automated system was assembled to physically sort the classified fruits into different containers. This research highlights the potential of integrating image processing and machine learning technologies to revolutionize fruit classification processes, thereby improving both efficiency and quality control in agriculture.