The automation of nanoprobing application relies on the accurate detection of probe tips in scanning electron microscope (SEM) images. This work explores the application of deep learning models to automate and improve this critical process. Different models such as Mask R-CNN, YOLOv8 and RTMDet, trained on a specialized dataset, are used to accurately detect, segment and localize probe tips in SEM images, even under challenging conditions. Results show that these models have the potential to improve the automation of nanoprobing workflows, particularly in automatic tip positioning and crash prevention. Future work will focus on production-level deployment and the integration of tracking algorithms.

Image segmentation is a valuable tool for visual image data inspection of semiconductor device structures. For the large amounts of data provided by recent advancements in automated scanning electron microscope (SEM) and focused ion beam-scanning electron microscope (FIB-SEM) data acquisition, automatic segmentation becomes indispensable to fully exploit the information contained in the data in automated characterization workflows. Using two exemplary FIB-SEM tomography datasets, we explored artificial intelligence based image segmentation using only a minimum amount of training images annotated by a human user.

In critical fields such as automotive, medicine, and defense, ensuring the reliability of microelectronics has been paramount given the extensive nature of their globalized supply chain. Automated visual inspection (AVI) of printed circuit boards (PCBs) offers a solution through computer vision and deep learning to automate defect detection, component verification, and quality assurance. In this paper, our research follows this precedent by introducing a novel dataset and annotations to train artificial intelligence (AI) models for extracting PCB connectivity components. Utilizing high-resolution images, and state-of-the-art instance segmentation models, this study aims to examine the difficulties in this implementation and lay the groundwork for more robust automated visual inspection.

Before failure analysis (FA) can start, a product must get from the customer to the correct location, which is not always trivial, especially in larger companies with many FA labs. Automating and optimizing this routing, therefore reducing manual labor, misrouting, and turnaround time, requires the development of problem-solving methods utilizing both explicit and implicit knowledge. The first type refers to known routing rules, e.g., based on lab equipment or certifications, whereas the second type must be induced from available data, e.g., by analyzing customer descriptions using machine learning (ML) methods. Therefore, to solve the routing problem, we suggest a neurosymbolic integration of natural language processing methods into the symbolic context of a logic-based solver. The conducted evaluation shows that the suggested method can reduce the reships by appr. 33% while ensuring the fulfillment of all shipment constraints.

The job of yield and failure analysis (YA and FA) engineers is to identify the root cause of low-yielding wafers. While physical FA is the most definitive method for determining root cause, resource limitations require YA engineers to search for root cause by identifying other wafers with similar yield signatures. The immense number of yield parameters, or features, collected in modern semiconductor processes makes this a difficult task. This paper presents a workflow employing multiple AI techniques to separate groups of wafers by their distinct yield signatures and determine the parameters most important to defining each group. This aids in the disposition of new low-yield wafers, maximizes the learning from previously collected FA wafers, and allows FA resources to be allocated more effectively, prioritizing them for the highest-impact, unknown fail modes.

Non-destructive inspection and analysis techniques are crucial for quality assessment and defect analysis in various industries. They enable for screening and monitoring of parts and products without alteration or impact, facilitating the exploration of material interactions and defect formation. With increasing complexity in microelectronic technologies, high reliability, robustness and thus, successful failure analysis is essential. Machine learning (ML) approaches have been developed and evaluated for the analysis of acoustic echo signals and time-resolved thermal responses for assessing their ability for defect detection. In the present paper different ML architectures were evaluated, including 1D and 2D convolutional neural networks (CNNs) after transforming time-domain data into the spectra-land wavelet domains. Results showed that 2D CNN with wavelet domain representation performed best, however at the expense of additional computational effort. Furthermore, ML-based analysis was explored for lock-in thermography to detect and locate defects in the axial dimension based on thermal emissions. While promising, further research is needed to fully realize its potential.

During the activity in the Failure Analysis (FA) laboratory, all corresponding findings and conclusions are included in a series of documents known as the FA reports. They shall, in the first place, inform the requestor about the analysis results. But additionally, they shall provide information to solve similar cases. Therefore, these documents play a key role in preserving the knowledge acquired by the engineers as they become available for consultation during future works. The different information systems in FA consist of databases, file shares, wikis, or other human-readable forms. However, the heterogeneity of these databases and the large number of independent documents make it inefficient for manual consultation. In this context, this paper proposes an application of Natural Language Processing (NLP) known as Named Entity Recognition (NER), consisting of an AI-based detection of key concepts in textual data in the form of annotations. These annotations can then be used to boost search systems or other AI models.

The goal of this work was to automate the crack detection of pads on a wafer piece. This process allows the engineer to check a huge number of pads for cracks to obtain a meaningful statistical result of the Pad-Over-Active-Areas (POAA) stability, which is a typical task in the failure analysis laboratories. It is possible that cracks in POAA appear during the electrical test or the bonding process. The current analysis process is very time consuming as thousands of pads have to be inspected for cracks by an engineer. The process starts with the chemical preparation of a wafer piece to make the crack below the pad visible. After that, the engineer examines each pad individually through the optical microscope for cracks. For the automation of this process a new workflow had to be developed and is described in this work. Moreover, it comprises the automation of a light microscope as well as an automated image evaluation based on a neural network.