Rapid and reliable qualification of advanced materials remains a bottleneck in industrial manufacturing, particularly for heterogeneous structures produced via non-conventional additive manufacturing processes. This study introduces a novel framework that links microstructure informatics with a range of expert characterization knowledge using customized and hybrid vision-language representations (VLRs). By integrating deep semantic segmentation with pre-trained multi-modal models (CLIP and FLAVA), we encode both visual microstructural data and textual expert assessments into shared representations. To overcome limitations in general-purpose embeddings, we developed a customized similarity-based representation that incorporates both positive and negative references from expert-annotated images and their associated textual descriptions. This allowed zero-shot classification of previously unseen microstructures through a net similarity scoring approach. Validation on an additively manufactured metal matrix composite (MMC) dataset demonstrated the framework’s ability to distinguish between acceptable and defective samples across a range of characterization criteria with up to 80% top-5 retrieval accuracy. Comparative analysis revealed that FLAVA model offers higher visual sensitivity and penalized weak similarities with score differences as large as 0.17 relative to CLIP. However, FLAVA’s text encoder exhibited sharp drops in similarity for paraphrased expert descriptions (falling below 0.20), whereas CLIP maintained more stable alignment with textual criteria (0.29–0.36). Z-score normalization adjusted raw unimodal and cross-modal similarity scores based on their local dataset-driven distributions, enabling more effective alignment and classification in the hybrid vision-language framework. The standardized scores provided strong binary classification results across three categories (82% for distribution, 90% for dilution, and 82% for reinforcement). The proposed method enhanced traceability and interpretability in qualification pipelines via human-in-the-loop decision-making without task-specific model retraining. By advancing semantic interoperability between raw data and expert knowledge, this work contributes toward scalable and domain-adaptable qualification strategies in engineering informatics.

Additive manufacturing (AM) is increasingly adopted across industries for its ability to support design flexibility, rapid prototyping, and mass customization. Machine learning (ML)-based cyber-physical systems (CPSs) have been extensively developed to improve the print quality of AM. However, the reproducibility of these systems has not been thoroughly investigated due to a lack of formal evaluation methods. Reproducibility, a critical component of trustworthy artificial intelligence, is achieved when an independent team can replicate the findings or artifacts of a study using a different experimental setup and achieve comparable performance. In many publications, critical information necessary for reproduction is often missing due to a lack of comprehensive AM and ML domain knowledge, resulting in systems that fail to replicate the reported performance. Integrating AM and ML domain knowledge, this paper proposes a reproducibility investigation pipeline and a reproducibility checklist for ML-based AM process monitoring and quality prediction systems. Based on the CRoss Industry Standard Process (CRISP) methodology, the pipeline guides researchers through the key steps required to reproduce a study, while the checklist systematically extracts information relevant to reproducibility from the publication. We validated the proposed approach through two case studies: reproducing a fused filament fabrication warping detection system and a laser powder bed fusion melt pool area prediction model. Both case studies confirmed that the pipeline and checklist successfully identified missing information, improved reproducibility, and enhanced the performance of reproduced systems. Based on the proposed checklist and leveraging large language models, a reproducibility survey was conducted to assess the current reproducibility status within this research domain.

Lattice structures with different desired physical properties are promising for a broad spectrum of applications. The availability of additive manufacturing (AM) technology has relaxed the fabricating limitation of lattice structures. However, manufacturing constraints still exist for AM-fabricated lattice structures, which have a significant influence on the printing quality and mechanical properties of lattice struts. In this paper, a design and optimization strategy is proposed for lattice structures with the consideration of manufacturability to ensure desired printing quality. The concept of manufacturable element is used to link the design and manufacturing process. A meta-model is constructed by experiments and the artificial neural network to obtain the manufacturing constraints. Sizes of struts are optimized by a bidirectional evolutionary structural optimization-based algorithm with these manufacturing constraints. An arm of quadcopter is redesigned and optimized to validate the proposed method. Its result shows that optimized heterogeneous lattice structures can improve the stiffness of the model compared to the homogeneous lattice structure and the original design. Both the Von-Mises stress and the maximum displacement are reduced without increasing the weight of designed part. And by considering the manufacturability constraints, the optimized design has been successfully fabricated by the selected additive manufacturing process. Note to Practitioners-Lattice structures might fail to be fabricated by the additive manufacturing technique if the designed model exceeds the processability of the machine. Our approach has the capability of considering the manufacturing constraints in the design and optimization process. We conducted experiments to investigate the manufacturability and proposed a method that can give the domain of the design variables for a selected manufacturing process. And we also designed an algorithm that can optimize the lattice structure inside the domain of design variables. It ensures that the lattice model can be successfully fabricated by the selected process and the performance is dramatically increased compared to the original design. Engineers can use our approach to optimize the lattice structure automatically without knowing the knowledge of optimization and manufacturability.