9086 research outputs found
Sort by
A Hybrid Machine Learning and Multi-Objective Optimization Framework for Enhancing the Grinding Expert System in Smart Manufacturing
No full textThis paper presents a hybrid Machine Learning (ML) and multi-objective optimization framework for a data-driven grinding expert system, targeting optimal surface roughness, grinding force, and material removal rate. It integrates Gaussian Process Regression (GPR) for accurate ML-based predictive modeling, Knowledge-based Adaptive Design of Experiments (KADoE) to reduce experimental trials, NSGA-II for generating trade-off solutions, feedback-driven self-learning to refine model accuracy, and Multi-Criteria Decision-Making (MCDM) methods (e.g., TOPSIS, and AHP) for final selection. Experimental validation showed prediction accuracies above 80%, highlighting the framework’s robustness and effectiveness in enhancing grinding process efficiency and quality in smart manufacturing
Modern Dressing and Grinding Technologies, Volume 1 — Principles, Tools, and Process Fluids
No full textSimulation and Optimization of Magnetic Field Actuated MEMS-Based Metasurface for Terahertz Applications
No full textDistributed zero trust architecture based on policy negotiation secured by DPP in blockchain
No full textDistributed policy negotiation is a key issue in modern manufacturing landscapes. However, existing solutions struggle with ensuring efficient policy negotiation and enforcement across entities while maintaining device authentication and trust in dynamic, cross-organizational environments. Specifically, the reliance on distributed architectures introduces challenges related to scalability, data fairness, and implementation complexity, limiting its applicability in large-scale industrial networks. This paper provides an overview of foundational concepts critical to understanding the proposed distributed policy consensus with blockchain integration: the proposed architecture leverages the decentralized and immutable nature of blockchain to achieve secure and transparent consensus among the distributed policy engines (PE), addressing challenges in policy negotiation within a distributed zero trust architecture (ZTA). The proposed methodology is applied in the context of a multinational supply chain network, involving various stakeholders such as manufacturers, suppliers, logistics providers, distributors, and retailers. The methodology provides a scalable foundation for advancing secure cross-company interactions in increasingly interconnected systems
Virtuelle Ausstellung als Digitaler Zwilling : Ein Rahmenwerk zur Entscheidungsfindung bei virtuellen Repräsentationen
No full text
Dataset on the analysis of β-galactosidase immobilization efficiency on aminomethyl polystyrene (AMP) resin in syringe and column reactors
No full textMechanical Properties and Plastic Deformation Mechanism of Additively Manufactured Novel Metastable Β Ti–42nb and Near Β Ti–20nb–6ta Alloy
No full textOperationalizing the R4VR-Framework : Safe Human-in-the-Loop Machine Learning for Image Recognition
No full textVisual inspection is a crucial quality assurance process across many manufacturing industries. While many companies now employ machine learning-based systems, they face a significant challenge, particularly in safety-critical domains. The outcomes of these systems are often complex and difficult to comprehend, making them less reliable and trustworthy. To address this challenge, we build on our previously proposed R4VR-framework and provide practical, step-by-step guidelines that enable the safe and efficient implementation of machine learning in visual inspection tasks, even when starting from scratch. The framework leverages three complementary safety mechanisms—uncertainty detection, explainability, and model diversity—to enhance both accuracy and system safety while minimizing manual effort. Using the example of steel surface inspection, we demonstrate how a self-accelerating process of data collection where model performance improves while manual effort decreases progressively can arise. Based on that, we create a system with various safety mechanisms where less than 0.1% of images are classified wrongly and remain undetected. We provide concrete recommendations and an open-source code base to facilitate reproducibility and adaptation to diverse industrial contexts