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.

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.

Failure Analysis (FA) uses combinations of a multitude of methods to identify and localize failures in semiconductor devices. The performance or efficiency of a FA lab is often measured by throughput or the total time needed to complete an analysis. These KPIs (Key Performance Indicators) can be optimized, but sequences of executed methods might be long and, therefore, hard to understand and optimize. Processes executed in an ad hoc manner might block each other causing long queues for specific tools and overloading specialists working with them. These and other factors can lead to a significant decrease in lab performance. In this paper, we propose an approach to analyzing FA processes with a focus on Internal Physical Inspection (IPI) jobs. Specifically, we use machine learning and statistical methods to estimate (a) the workflow that engineers follow while completing IPI jobs and (b) the duration of each operation executed in the workflow. Our approach starts with data extraction and preprocessing, aiming at extracting features characterizing the workflow, like the package or technology of a sample, as well as providing information about the complexity of each task, thus, allowing us to predict their duration. The resulting tool allows lab management and team leads to analyze the execution of IPI jobs and optimize them. Moreover, the information provided by the tool can be used in automated scheduling methods providing recommendations to FA engineers about sequences of jobs improving utilization of the lab’s resources.

Failure Analysis (FA) is a complex activity that requires careful and complete documentation of all findings and conclusions to preserve knowledge acquired by engineers in this process. Modern FA systems store this data in text or image formats and organize it in databases, file shares, wikis, or other human-readable forms. Given a large volume of generated FA data, navigating it or searching for particular information is hard since machines cannot process the stored knowledge automatically and require much interaction with experts. In this paper, we investigate applications of modern Natural Language Processing (NLP) approaches to the classification of FA texts with respect to electrical and/or physical failures they describe. In particular, we study the efficiency of pretrained Language Models (LM) in the semiconductors domain for text classification with deep neural networks. Evaluation results of LMs show that their vocabulary is not suitable for FA applications, and the best classification accuracy of appr. 60% and 70% for physical and electrical failures, respectively, can only be reached with fine-tuning techniques.

The development of intelligent assistants helping Failure Analysis (FA) engineers in their daily work is essential to any digitalization strategy. In particular, these systems must solve various computer vision or natural language processing problems to select the most critical information from heterogeneous data, like images or texts, and present it to the users. Modern artificial intelligence (AI) techniques approach these tasks with machine learning (ML) methods. The latter, however, require large volumes of training data to create models to solve the required problems. In most cases, enterprise clouds store vast volumes of data captured while applying various FA methods. Nevertheless, this data is useless for ML training algorithms since it is stored in forms that can only be interpreted by highly-trained specialists. In this paper, we present an approach to embedding an annotation process in the everyday routines of FA engineers. Its services can easily be embedded in existing software solutions to (i) capture and store the semantics of each data piece in machine-readable form, as well as (ii) provide predictions of ML models trained on previously annotated data to simplify the annotation task. Preliminary experiments of the built prototype show that the extension of an image editor used by FA engineers with the services provided by the infrastructure can significantly simplify and speed up the annotation process.

In their daily work, engineers in semiconductor Failure Analysis (FA) laboratories generate numerous documents, recording the tasks, findings, and conclusions related to every device they handle. This data stores valuable knowledge for the laboratory that other experts can consult, but being in the form of a collection of documents pertaining to particular devices and their processing history makes it difficult if not practically impossible to find answers to specific questions. This paper therefore proposes a Natural Language Processing (NLP) solution to make the gathering of FA knowledge from numerous documents more efficient. It explains how the authors generated a dataset of FA reports along with corresponding electrical signatures and physical failures in order to train different machine-learning algorithms and compare their performance. Three of the most common classification algorithms were used in the study: K-Nearest Neighbors (kNN), Support Vector Machines (SVM), and Deep Neural Networks (DNN). All of the classification models produced were able to capture patterns associated with different types of failures and predict the causes. The outcomes were best with the SVM classifier and all classifiers did slightly better in regard to physical faults. The reasons are discussed in the paper, which also provides suggestions for future work.

Fault analysis is a complex task that requires engineers to perform various analyses to detect and localize physical defects in semiconductor devices. The process is knowledge intensive and must be precisely documented. In order to ensure unambiguous documentation, engineers must agree on a clearly defined terminology specifying methods, tools, physical faults and their electrical signatures among other things, and it must be stored in a way that is usable for both engineers and software. One possible solution to this challenge is to formalize domain knowledge as an ontology, a knowledge base designed to store terminological definitions. This paper discusses the development of an ontology for electronic device failure analysis that uses a logic-based representation. The latter ensures that terms are interpreted the same way by engineers and software systems, facilitating the automation of tasks such as text classification, information retrieval, and workflow verification.