With the rapid development of semiconductor manufacturing technology, methods to effectively control the production process, reduce variation in the manufacturing process, and improve the yield rate represent important competitive factors for wafer factories. Wafer bin maps, a method for characterizing wafer defect patterns, provide valuable information for engineers to quickly identify potential root causes through accurate pattern recognition. Vision-based deep learning approaches rely on visual patterns to achieve robust performance. However, they rarely exploit the rich semantic information embedded in defect descriptions, limiting interpretability and generalization. To address this gap, we propose YOLO-LA, a lightweight prototype-based vision–language alignment framework that integrates a pretrained frozen YOLO backbone with a frozen text encoder to enhance wafer defect recognition. A learnable projection head is introduced to map visual features into a shared embedding space, enabling classification through cosine similarity Experimental results on the WM-811K dataset demonstrate that YOLO-LA consistently improves classification accuracy across different backbones while introducing minimal additional parameters. In particular, YOLOv12 achieves the fastest speed while maintaining competitive accuracy, whereas YOLOv10 benefits most from semantic prototype alignment. The proposed framework is lightweight and suitable for real-time industrial wafer inspection systems.
Sapphire crystals, owing to their outstanding mechanical and optical properties, which are widely used in advanced optics, microelectronic devices, and medical instruments. The manufacturing precision of sapphire optical components critically affects the performance of advanced optical systems. However, the extremely high hardness and low fracture toughness of sapphire make it a typical hard-to-machine material, prone to brittle surface fractures and subsurface damage during material removal. Improving the machinability of sapphire remains a pressing challenge in advanced manufacturing. In this study, surface modification and enhanced ductility of C-plane sapphire were achieved via ion implantation, and the machinability of the modified sapphire was further improved through laser-assisted diamond machining (LADM). Monte Carlo simulations were employed to investigate the interaction mechanisms between incident ions and the target material. Based on the simulation results, phosphorus ion implantation experiments were conducted, and transmission electron microscopy observation was used to characterize the microstructural evolution of the modified layer, while the optical properties of the samples before and after modification were analyzed. Finally, groove cutting experiments verified the enhancement in ductile machinability of the modified sapphire under LADM. At a laser power of 16 W, the ductile–brittle transition depth of the modified sapphire increased to 450.67 nm, representing a 51.57% improvement over conventional cutting. The findings of this study provide valuable insights for improving the ductile machining performance of hard and brittle materials.
The multi-jet polishing process (MJP) demonstrates high shape accuracy and surface quality in the machining of nonlinear and complex surfaces, and it achieves precise and adjustable material removal rates through computer control. However, there are still challenges in terms of machining efficiency, system complexity, and stability. In particular, maintaining the polishing quality presents a greater challenge when working conditions change. To overcome these issues, this paper conceptually proposes a digital twin (DT)-driven, human-centric design framework that integrates key factors of MJP, such as jet kinetic energy, nozzle structure, abrasive type, and machining path. Within this framework, a feature-encoded transfer learning-based model is introduced to enhance surface roughness prediction accuracy and robustness under varying working conditions. The effectiveness of the proposed model was verified by conducting experiments on 3D printed workpieces under two different MJP working conditions. The results show that our proposed method yields better predictive performance and cross-condition adaptability. Overall, this work provides a predictive modeling component that supports DT-driven process design, offering a practical and extensible perspective for optimizing complex ultra-precision manufacturing processes under data-scarce and uncertainty-dominated conditions.
Fluid jet polishing process (FJP) demonstrates high shape accuracy and surface quality in the machining of nonlinear and complex surfaces, and it achieves precise and adjustable material removal rates through computer control. However, there are still challenges in terms of machining efficiency, system complexity, and stability. Particularly, there is uncertainty in process optimization, especially with higher challenges in optimizing process parameters after changes in working conditions. This study utilizes digital twin technology to propose a new framework for optimizing the FJP process. By reviewing the application of DT in the machining field, this paper identifies the limitations of existing methods and proposes a human-centric design approach that integrates key factors of DT-driven FJP, such as jet kinetic energy, nozzle structure, abrasive type, and machining path. This method encompasses multiple aspects from removal function models to machining path algorithms. By introducing a core method based on transfer learning, this research aims to improve the predictive accuracy, machining efficiency, and stability of the FJP process, realizing efficient and precise polishing operations. Ultimately, this paper validates the proposed method through a case study on 3D printed workpieces, discusses the key enabling technologies, and main challenges. This study not only advances the application potential of FJP process but also provides a new perspective and strategy for optimizing complex machining processes using DT technology.
It is difficult to achieve ultra-precision machining (UPM) on semiconductor materials like single-crystal silicon because of their hardness and brittleness. To solve this issue, numerous field-assisted machining systems and their combinations have been suggested and developed. However, the difficulty in directly observing the physical variables limits our comprehension of the in-depth machining mechanisms of field-assisted machining. In this work, we investigated the machining mechanism of single-crystal silicon under the combination of laser heating and tool vibration using molecular dynamics (MD) simulations. The effect of tool vibration trajectory determined by different tool edge radii is discussed under the condition of raising temperature. The simulation results indicate that the surface morphology is closely related to vibration and heating parameters. Raising the cutting temperature causes a reversed relation between tool edge radius and surface roughness. While the subsurface damage and internal stress are also determined by the tool edge radius and cutting temperature. The findings in this simulation could help to improve the understanding of machining mechanics in multi-field-assisted machining.
21 August 2026 in International Journal of Precision Engineering and Manufacturing