Achieving zero-defect manufacturing (ZDM) in semiconductor assembly requires root-cause analysis (RCA) methods that do more than flag excursions. They must produce explanations and corrective actions that are traceable to process evidence, interpretable to quality engineers, and maintainable within governed manufacturing workflows. This paper presents a deterministic, provenance-aware RCA framework for surface-mount technology (SMT) assembly that transforms heterogeneous line data, including specifications, measurements, inspection outcomes, and troubleshooting knowledge, into explicit causal-chain representations connecting parameter deviations to mechanisms, defects, and corrective-action specifications. The framework operationalizes three recurring reasoning patterns: Specification–Observation–Conformance Assessment, Disposition–Realization–Failure Cause, and Cause–Effect–Corrective Action. These patterns are implemented over a governed knowledge graph using SPARQL rule templates that materialize conformance assessments, failure causes, directed causal links, and provenance traces. We demonstrate the approach on solder-bridging and open-circuit scenarios using synthetic data designed to reflect industrial SMT data schemas and industrially plausible value regimes. System inferences are compared against expert-reviewed reference graphs using chain precision and recall, provenance completeness, and cycle rate. Across 30 simulated shifts of 5000 printed circuit boards (PCBs) each, the framework achieves approximately 90

Root-cause analysis (RCA) in surface-mount assembly must support rapid containment and verification while remaining auditable under imperfect evidence. We present an uncertainty-aware neurosymbolic pipeline in which a neural evidence layer converts per-board observations into probabilities for defect class, stage-wise mechanism, and parameter risk, and a deterministic semantic layer assembles and ranks defect–mechanism–parameter-violation hypotheses over an explicit causal mapping. The output is an RCA packet: a small ranked hypothesis set with evidence and provenance pointers (observation value, specification bounds, timestamp, and model version). We evaluate on a synthetic multi-shift dataset with correlated variation and graded ground truth, reporting set-level agreement (hit rate, Jaccard similarity, exact match), triage behavior (mean reciprocal rank), and edge-level fidelity with a cycle sanity check.

Modern manufacturing environments require dynamic, data-driven facility layouts that can adapt to changing demands for efficiency, resilience, sustainability, and flexibility. Advancements in Industry 4.0 technologies have enabled transformative methods for optimizing layout design. Unlike previous reviews that emphasize traditional algorithms or single technologies, this paper presents a structured literature review that synthesizes recent applications of machine learning, digital twin technologies, and augmented and virtual reality in facility layout planning and broader manufacturing layout planning contexts. Application domains, methodological approaches, dataset and performance characteristics, and integration challenges are the four dimensions examined in this review. The results show that digital twin frameworks facilitate real-time simulation, monitoring, and continuous layout evaluation; machine learning enables adaptive and sequential optimization through predictive modeling, reinforcement learning, and surrogate-assisted techniques; augmented reality supports in-situ spatial validation within physical environments while virtual reality enables immersive design exploration in synthetic ones, together improving ergonomics evaluation and collaborative decision-making. Although each technology provides significant benefits, fully integrated machine learning, digital twin, and augmented or virtual reality systems remain limited and face challenges related to heterogeneous data sources, scalability, and interoperability. This review contributes a structured map of current approaches and presents a forward-looking roadmap for intelligent, interoperable, and human-centered manufacturing facility layout planning in the Industry 4.0 era by synthesizing trends, research gaps, and emerging technological synergies.

This paper presents an innovative framework for defect diagnosis and quality control in semiconductor manufacturing using neurosymbolic AI. By integrating knowledge graphs, ontologies, and deep neural networks, the approach harmonizes human expertise and machine learning to predict, diagnose, and mitigate defects in real time. Focusing on Failure Mode and Effects Analysis (FMEA) and other expert sources, the research transforms this knowledge into structured physics-aware ontologies and knowledge graphs to enable seamless integration with neural networks. Additionally, novel ensemble neural network architectures are explored to leverage the full potential of physics-aware neurosymbolic techniques. The proposed framework enhances defect prediction and root cause analysis and also paves the way for more interpretable and adaptive AI-driven quality control systems in semiconductor manufacturing.

In modern manufacturing, managing certificates efficiently is crucial for compliance and quality assurance. However, existing systems often suffer from data fragmentation and interoperability issues. To address these challenges, we developed CertOn, a Certificate Ontology aligned with the Industrial Ontologies Foundry (IOF) and Basic Formal Ontology (BFO). CertOn provides a structured framework for handling certificates across various industrial domains, enhancing data interoperability, and supporting automated reasoning for compliance tracking and product lifecycle assessment. This paper presents CertOn’s class structure and offers natural language and semi-formal definitions for them. The ontology illustrates different agents and processes that are involved in the attestation of the certified entity as well as the generation of the certificate. We demonstrate CertOn’s capabilities through competency questions and represent them via SPARQL Protocol and RDF Query Language (SPARQL) queries. These queries showcase CertOn’s ability to manage complex certificate-related data and fundamental relationships between the entities. They also prove CertOn’s ability to support critical use cases in manufacturing and supply chain management. The queries are designed to handle a manually generative dataset that describes a small knife manufacturer to show the complexity of certificates for such simple systems. The paper concludes that by integrating CertOn into existing industrial frameworks, industries can enhance operational efficiency and regulatory compliance.

Underwater robotics produces diverse and complex streams of sensor, image, video, and navigational data under challenging environmental conditions, creating obstacles for seamless integration and interpretation. This paper introduces ROVON (Remotely Operated Vehicle Ontology), a semantic framework designed to enhance interoperability and reasoning in underwater operations. While ROVON is conceptually scalable to large, heterogeneous datasets, its validation in this study focuses on controlled underwater inspection data collected for pipeline applications. ROVON enables the representation and analysis of multimodal underwater data by semantically annotating raw sensor feeds, enforcing data integrity, and leveraging knowledge graphs to convert disparate inputs into actionable insights. The ontology demonstrates how a structured semantic approach facilitates advanced analysis that improves decision-making, supports proactive maintenance strategies, and enhances operational safety. The proposed framework was validated through a controlled pipeline inspection scenario.

A cognitive digital twin (CDT) is a digital replica of a physical system that leverages cognitive computing capabilities to learn, adapt, and optimize its performance based on data collected from the system. One key requirement for realizing the vision of a CDT is enabling it to comprehend the semantics of data. This paper presents a semantic framework designed to support cognitive digital twins (CDTs). The proposed framework uses an ontology based on Basic Formal Ontology (BFO) and Industrial Ontology Foundry (IOF) to systematically organize and semantically annotate the diverse data collected from Digital Twins (DTs). A simulated machine cell, created in Unity 3D, provides the test case for the proposed framework. The simulation also serves as a testbed for real-world deployment for scalable development of CDTs. This model generates a wide range of operational data such as PLC signals, sensor readings, robot movements recordings, and drive dynamics. By aligning the generated data with the ontology, a Knowledge Graph (KG) was created to enrich data for downstream analytics, including real-time system monitoring and fault detection. The KG was validated through formulating and executing a series of queries for state monitoring and realistic fault detection scenarios. This work demonstrates the potential of semantic-driven methods for advancing diagnostic accuracy, interoperability, and decisionmaking capabilities within Digital Twin (DT) applications.

The adaptability of robotic systems is expanding the horizons of manufacturing flexibility. However, fully leveraging the potential of these systems poses considerable challenges. A key requirement is the ability to understand and model their diverse capabilities through a standardized and semantically well-defined framework. In this paper, we introduce the Robotic Capability Ontology (RCO), developed through a systematic investigation of various types of robotic capabilities, including those related to function, quality, and process performance. We define two types of capabilities: Advertised capabilities, as specified by manufacturers, and Operational capabilities, which reflect real-world performance. The RCO framework provides an ontology-based approach to representing these capabilities in a structured and interpretable manner. Within the manufacturing context, RCO serves as a reference ontology that bridges manufacturer specifications and empirical performance data to support more accurate, explainable, and interoperable representations of robotic capabilities.

Contract manufacturers rely on company websites to showcase their technological capabilities and expertise. However, this information is often presented in an unstructured and informal format, making it difficult to search and retrieve using automated methods. Obtaining a comprehensive view of available capabilities is challenging due to the fragmented, disconnected, and heterogeneous nature of the data. This research introduces a web information extraction pipeline that processes raw textual data and transforms it into a semantic knowledge graph aligned with formal ontologies. The proposed pipeline leverages Large Language Models (LLMs) to extract, structure, and enrich the data. A formal ontology is used for providing a unifying semantic framework for data alignment. The results demonstrate that a semantic knowledge graph significantly enhances the efficiency of searching and mapping manufacturing capabilities, ultimately streamlining the supplier discovery process. The capability knowledge graph provides the stakeholders in the manufacturing value chain with open access to shared and verifiable data resources and trustable knowledge models that can support data-driven and AI-powered construction and analysis of supply chains. This project aims to leverage the power of open knowledge networks to enable the search and discovery of small manufacturing firms and better connect them with prospective supply chain partners.

Manufacturers often struggle to determine the appropriate processes, machines, and procedures when developing a new product, especially without extensive technical expertise. While knowledge graphs (KGs) help structure and access domain-specific information, existing solutions lack commonsense reasoning to support intuitive and explainable decision-making. This demo presents MACS-KG, a KG designed to assist manufacturers in identifying the right manufacturing processes for a product, selecting suitable machines, and following standardized procedures. By integrating Manufacturing Commonsense Knowledge (MCSK) rules, MACS-KG enables reasoning and explainable decision-making, ensuring that manufacturers access relevant information and understand the rationale behind each recommendation derived using MCSK.

As supply chain complexity and dynamism challenge traditional management approaches, integrating large language models (LLMs) and knowledge graphs (KGs) emerges as a promising method for advancing supply chain analytics. This article presents a methodology crafted to harness the synergies between LLMs and KGs, with a particular focus on enhancing supplier discovery practices. The primary goal is to transform and integrate a vast body of unstructured supplier capability data into a harmonized KG, thus improving the supplier discovery process and enhancing the accessibility and findability of manufacturing suppliers. Through an ontology-driven graph construction process, the presented methodology integrates KGs and retrieval-augmented generation with advanced LLM-based natural language processing techniques. With the aid of a detailed case study, we showcase how this integrated approach not only enhances the quality of answers and increases visibility for small- and medium-sized manufacturers but also amplifies agility and provides strategic insights into supply chain management.

This paper underscores the critical role of integrating common knowledge across the manufacturing industry by introducing and formalizing a new concept called Manufacturing Commonsense Knowledge (MCSK). Although commonsense knowledge is crucial for enhancing AI-driven operational intelligence and decision-making, its structured application within the broader manufacturing sector has been insufficient. To bridge this gap, we present a structured methodology for translating MCSK into first-order logic (FOL), employing standard ontological frameworks such as the Basic Formal Ontology (BFO), Industrial Ontologies Foundry (IOF), Relations Ontology (RO), and Machine Services Description Language (MSDL). This translation process is pivotal, whether the underlying AI systems employ symbolic, sub-symbolic, or hybrid approaches, as it transforms intuitive MCSK into organized semantic rules. Our findings demonstrate that structured MCSK patterns enhance knowledge representation's clarity and utility and significantly improve the explanatory capabilities of AI decision-making processes across the industry. The broader impacts of our research extend to enhancing machine interoperability, predictive analytics, and advanced manufacturing practices, thus paving the way for a more informed and efficient industrial future.

This research conducts a comparative analysis and scoping review of 105 studies in the field of Fracture Liaison Service (FLS). The resulting two-dimensional framework represents a significant step toward FLS implementation. PURPOSE:The primary goal is to review interventions in real world settings in order to provide the FLS framework that specifies the essential elements of its implementation and offers different perspectives on that. METHOD:This study encompasses two phases: a comparative analysis of existing FLS models, including "Capture the Fracture," "5IQ," and "Ganda," and a scoping review from 2012 to 2022 in PubMed, Web of Science, Scopus, ProQuest, and IEEE databases limited to publications in English. RESULTS:The resulting model of comparative analysis identifies patient identification, investigation, intervention and integration or continuity of care as the four main stages of FLS. Additionally, the elements of quality and information span across all stages. Following comparative analysis, the framework is designed to be used for content analysis of the included studies in the scoping review. The intersection of columns (Who, Where, When, What, How, Quality) with rows (Identification, Investigation, Intervention, and continuity of care) yields a set of questions, answered in tabular form based on the scoping review. CONCLUSION:The framework offers potential benefits in facilitating the adoption of effective approaches for FLS implementation. It is recommended to undertake an in-depth review of each of these components in order to uncover novel and innovative approaches for improving their implementation.

Access to accurate manufacturing capability information is necessary for efficient supplier discovery and agile supply chain formation. However, manufacturing capability data, particularly for small and medium-sized manufacturers, is often unavailable or, if accessible, lacks essential qualities such as correctness, completeness, interoperability, and openness. The objective of the research presented in this paper is to develop an open Manufacturing Capability Network (MCN) that represents various manufacturers' capabilities as an interconnected and formal knowledge graph. This capability graph is part of a larger graph referred to as the Supply and Demand Open Knowledge Network (SUDOKN). The ontologies that provide the semantics of the knowledge graph comply with the Basic Formal Ontology (BFO). A proof-of-concept knowledge graph, based on 1700 manufacturers, is presented in this work. The graph's validity was assessed by submitting queries related to supplier discovery use cases. SUDOKN, once fully deployed, serves as a shared, canonical, and consensus-driven knowledge backbone, that supports supply chain analytics solutions with AI-ready data.

Traceability of food products to their sources is critical for quick responses to food emergencies. However, having the complete and consistent information needed to quickly investigate sources and identify affected material has proven difficult. Food trace-ability is challenging for a variety of reasons including diversity and heterogenicity of participants, complexity of the supply chain and its processes and lack of a common understanding of steps in a supply chain, and incompleteness of data, and unwillingness of actors to expose information of their internal operations. The objective of this work is to address the traceability challenge by developing a formal ontology that can provide a shared and common understanding of the traceability model across all stakeholders in a food supply chain. In previous research, an ontological approach was adopted to address the traceability problem in the context of a use case related to harvest to on-farm storage activities. This work extends the Supply Chain Traceability ontology by introducing additional critical tracking events including transform event, sampling event, observation event, custody change event, and ownership change event in the context of a scenario involving shipments of commodity grain from a primary grain elevator to a processor of grain such as a feed manufacturer. A knowledge graph is generated based on a simulated dataset and the ontology is validated through query, reasoning, and visualization conducted in the RDFox environment.

The global transition from traditional manufacturing systems to Industry 4.0 compatible systems has already begun. Therefore, the digitization of the manufacturing systems across the globe is increasing with exponential growth which implies a significant increase in the volume and variety of the generated data. Industry 4.0 technologies are mostly data driven and therefore, manufacturers need to be equipped with the appropriate tools and skill sets to extract useful knowledge and insights from the plethora of data continually collected form shop floors. Furthermore, quality assurance is a key domain in manufacturing that uses almost all the industry 4.0 technologies and has great impact on the sustainability of a manufacturing systems. The latest approach to higher quality and manufacturing sustainability is named Zero Defect Manufacturing (ZDM). ZDM interest has spiked the last three years illustrating the need for an alternative quality assurance approach from the traditional such as Six Sigma and Lean manufacturing. Therefore, the goal of this paper is to create a ZDM ontology that can semantically align multiple software systems that interact in a ZDM ecosystem. The development of the proposed ZDM ontology was performed using the principles introduced by Industrial Ontology Foundry (IOF) and with the use of Basic formal ontology (BFO) as an upper level ontology. The proposed ontology was utilized in the Prediction Optimization Designer tool developed, to assist developers to create new projects reusing existing resources, or to respond to a specific challenge. The use case validation results show that the combination of Natural Language Processing (NLP) using Sentence-BERT and ontology-based search methods rooted in the ZDM ontology is a promising strategy to implement effective search engines for applications in the ZDM domain.

The unstructured data available on the websites of manufacturing suppliers and contractors can provide valuable insights into their technological and organizational capabilities. However, since the capability data are often represented in an unstructured and informal fashion using natural language text, they do not lend themselves well to computational analysis. The objective of this work is to propose framework to enable automated classification and ranking of manufacturing suppliers based on their online capability descriptions in the context of a supplier search and discovery use case. The proposed text analytics framework is supported by a formal thesaurus that uses Simple Knowledge Organization System (SKOS) that provides lexical and structural semantics. Normalized Google Distance (NGD) is used as the metric for measuring the relatedness of terms when ranking suppliers based on their similarities with the queried capabilities. The proposed framework is validated experimentally using a hypothetical supplier search scenario. The results indicate that the generated ranked list is highly correlated with human judgment, especially when the search space is partitioned into multiple classes of suppliers with distinct capabilities. However, the correlation decreases when multiple overlapping classes of suppliers are merged together to form a heterogenous search space. The proposed framework can support supplier screening and discovery solutions by improving the precision, reliability, and intelligence of their underlying search engines.