Bøger af Janos Abonyi

This book presents a comprehensive framework for developing Industry 4.0 and 5.0 solutions through the use of ontology modeling and graph-based optimization techniques. With effective information management being critical to successful manufacturing processes, this book emphasizes the importance of adequate modeling and systematic analysis of interacting elements in the era of smart manufacturing.The book provides an extensive overview of semantic technologies and their potential to integrate with existing industrial standards, planning, and execution systems to provide efficient data processing and analysis. It also investigates the design of Industry 5.0 solutions and the need for problem-specific descriptions of production processes, operator skills and states, and sensor monitoring in intelligent spaces.The book proposes that ontology-based data can efficiently represent enterprise and manufacturing datasets.The book is divided into two parts: modelingand optimization. The semantic modeling part provides an overview of ontologies and knowledge graphs that can be used to create Industry 4.0 and 5.0 applications, with two detailed applications presented on a reproducible industrial case study. The optimization part of the book focuses on network science-based process optimization and presents various detailed applications, such as graph-based analytics, assembly line balancing, and community detection.The book is based on six key points: the need for horizontal and vertical integration in modern industry; the potential benefits of integrating semantic technologies into ERP and MES systems; the importance of optimization methods in Industry 4.0 and 5.0 concepts; the need to process large amounts of data while ensuring interoperability and re-usability factors; the potential for digital twin models to model smart factories, including big data access; and the need to integrate human factors in CPSs and provide adequate methods tofacilitate collaboration and support shop floor workers.

This work presents a data visualization technique that combines graph-based topology representation and dimensionality reduction methods to visualize the intrinsic data structure in a low-dimensional vector space. The application of graphs in clustering and visualization has several advantages. A graph of important edges (where edges characterize relations and weights represent similarities or distances) provides a compact representation of the entire complex data set. This text describes clustering and visualization methods that are able to utilize information hidden in these graphs, based on the synergistic combination of clustering, graph-theory, neural networks, data visualization, dimensionality reduction, fuzzy methods, and topology learning. The work contains numerous examples to aid in the understanding and implementation of the proposed algorithms, supported by a MATLAB toolbox available at an associated website.

Overview Since the early 1990s, fuzzy modeling and identification from process data have been and continue to be an evolving subject of interest. Although the application of fuzzy models proved to be effective for the approxima- tion of uncertain nonlinear processes, the data-driven identification offuzzy models alone sometimes yields complex and unrealistic models. Typically, this is due to the over-parameterization of the model and insufficient in- formation content of the identification data set. These difficulties stem from a lack of initial a priori knowledge or information about the system to be modeled. To solve the problem of limited knowledge, in the area of modeling and identification, there is a tendency to blend information of different natures to employ as much knowledge for model building as possible. Hence, the incorporation of different types of a priori knowledge into the data-driven fuzzy model generation is a challenging and important task. Motivated by our research into this topic, our book presents new ap- proaches to the construction of fuzzy models for model-based control. New model structures and identification algorithms are described for the effec- tive use of heterogenous information in the form of numerical data, qualita- tive knowledge and first-principle models. By exploiting the mathematical properties of the proposed model structures, such as invertibility and local linearity, new control algorithms will be presented.

The concept of industry 4.0 is spreading worldwide and readiness models exist to determine organizational or national maturity. This book identifies the regional aspect of industry 4.0 and provides a regional (NUTS 2 classified) industry 4.0 indicator system model that is based on open data sources.

The key idea of this book is that hinging hyperplanes, neural networks and support vector machines can be transformed into fuzzy models, and interpretability of the resulting rule-based systems can be ensured by special model reduction and visualization techniques.

The aim of this book is to illustrate that advanced fuzzy clustering algorithms can be used not only for partitioning of the data. It can also be used for visualization, regression, classification and time-series analysis, hence fuzzy cluster analysis is a good approach to solve complex data mining and system identification problems.