检索结果-内蒙古大学图书馆

An Approach to Integrate Domain Knowledge into Feature Engineering to Enhance Data-Driven Surrogate Models of Simulations

引用

Procedia CIRP 2024年 130卷 1276-1281页

作者： Julian Ziegler Peter Reimann Alexander Kilian Christoph Schulz Florian Keller Bernhard Mitschang Graduate School of advanced Manufacturing Engineering (GSaME) University of Stuttgart Germany MANN+HUMMEL GmbH Ludwigsburg Germany Institute for Parallel and Distributed Systems (IPVS) University of Stuttgart Germany

Surrogate models are approximations of time-consuming computer simulations. These approximations may be derived by applying machine learning algorithms on a training set of simulation data. Due to the long runtime of these simulations, the amount of available simulation data to train surrogate models is however limited. This leads to a subpar performance of the resulting surrogate models. Here, feature engineering is important to improve the prediction performance of these models. However, automated feature engineering often results in surrogate models that no longer adhere to laws of physics. Therefore, we present an approach to integrate domain knowledge into feature engineering for the data-driven creation of surrogate models. This reduces the prediction error of surrogate models and ensures they adhere to laws of physics and are thus more reliable. We validate our approach with a prototypical implementation and a use case for simulations in virtual product development, and we show that our approach significantly improves the prediction quality of the resulting surrogate models.

关键词： surrogate modeling feature engineering domain experts domain knowledge machine learning

来源：评论

学校读者我要写书评

暂无评论

A Kanban-based Approach to Manage Machine Learning Projects in Manufacturing

引用

Procedia CIRP 2025年 134卷 109-114页

作者： Ulf Schreier Peter Reimann Bernhard Mitschang Business Information Systems - Furtwangen University of Applied Science Robert-Gerwig-Platz 1 D-78120 Furtwangen Germany Graduate School of advanced Manufacturing Engineering (GSaME) University of Stuttgart Nobelstr. 12 D-70569 Stuttgart Germany Institute for Program Structures and Data Organization (IPD) Karlsruhe Institute of Technology (KIT) Am Fasanengarten 5 D-76131 Karlsruhe Germany Institute of Parallel and Distributed Systems (IPVS) University of Stuttgart Universitätsstr. 38 D-70569 Stuttgart Germany

A growing number of machine learning (ML) projects in manufacturing require the collaboration of various experts. In addition to data scientists, stakeholders with production engineering knowledge have to specify and prioritize individual project tasks. Data engineers prepare input data, while machine learning operations (MLOps) engineers ensure that trained models are deployed and monitored within IT landscapes. Existing project management approaches, e.g., Scrum, have problems for ML projects, as they do not consider various expert roles or ML project stages. We propose a project management approach defining a Kanban workflow by readjusting stages of ML development lifecycles, e.g., CRISP DM. This makes it possible to map expert roles to stages of the Kanban workflow. An adapted Kanban board allows visualizing and reviewing the status of all project tasks. We validate our approach with specific use cases, showing that it facilitates ML project management in manufacturing.

关键词： Machine learning (ML) ML project management machine learning operations (MLOps) Kanban Scrum

来源：评论

学校读者我要写书评

暂无评论

A Flexible Digital Twin Framework for ATMP Production – Towards an efficient CAR T Cell Manufacturing

引用

Procedia CIRP 2024年 125卷 124-129页

作者： Arber Shoshi Yuchen Xia Andrea Fieschi Thomas Ackermann Peter Reimann Michael Weyrich Bernhard Mitschang Thomas Bauernhansl Robert Miehe Graduate School of Excellence advanced Manufacturing Engineering (GSaME) University of Stuttgart Nobelstraße 12 70569 Stuttgart Germany Fraunhofer Institute for Manufacturing Engineering and Automation (IPA) Nobelstraße 12 70569 Stuttgart Germany Institute of Industrial Automation and Software Engineering (IAS) University of Stuttgart Pfaffenwaldring 47 70550 Stuttgart Germany Institute for Parallel and Distributed Systems (IPVS) University of Stuttgart Universitätsstraße 38 70569 Stuttgart Germany Institute of Industrial Manufacturing and Management University of Stuttgart Allmandring 35 70569 Stuttgart Germany

This paper presents a flexible digital twin framework tailored for the biomanufacturing of advanced therapeutic medicinal products (ATMPs), particularly focusing on chimeric antigen receptor T cell (CAR T cell) therapy. CAR T cell therapies face significant challenges in the management of their personalized and complex biomanufacturing processes. To tackle these issues, we propose a novel software framework for a digital twin system aimed at digitalizing, monitoring, and managing the physical production processes. The framework has a hierarchical architecture that methodically organizes production into distinct abstraction levels: processes, tasks and skills, and microservices. This layered architecture simplifies the modeling of complex biomanufacturing production and ensures that each component can be individually updated without disrupting the overall system. Furthermore, the digital twin framework integrates both manual and automated operations within a unified system, accommodating varied requirements of diverse tasks in production. The framework’s flexibility enables easier adaptation to dynamic technological advancements and allows for swift modifications in production processes, making it a sustainable and resilient digitalization infrastructure for ATMP manufacturing.

关键词： ditgital twin ATMP CAR T cell production flexible software framework skill modeling

来源：评论

学校读者我要写书评

暂无评论

Formalizing Cost Fairness for Two-Party Exchange Protocols using Game Theory and Applications to Blockchain

Formalizing Cost Fairness for Two-Party Exchange Protocols u...

引用

IEEE International Conference on Blockchain and Cryptocurrency (ICBC)

作者： Matthias Lohr Kenneth Skiba Marco Konersmann Jan Jü rjens Steffen Staab Institute for Software Technology University of Koblenz-Landau Koblenz Germany Fernuniversit&#x00E4 t in Hagen Artificial Intelligence Group Hagen Germany Fraunhofer ISST Dortmund Germany Institute for Parallel and Distributed Systems (IPVS) University of Stuttgart Stuttgart Germany University of Southampton Southampton United Kingdom

Existing fair exchange protocols usually neglect consideration of cost when assessing their fairness. However, in an environment with non-negligible transaction cost, e.g., public blockchains, high or unexpected transaction cost might be an obstacle for wide-spread adoption of fair exchange protocols in business applications. For example, as of 2021-12-17, the initialization of the FairSwap protocol on the Ethereum blockchain requires the selling party to pay a fee of approx. 349.20 USD per exchange. We address this issue by defining cost fairness, which can be used to assess two-party exchange protocols including implied transaction cost. We show that in an environment with non-negligible transaction cost where one party has to initialize the exchange protocol and the other party can leave the exchange at any time cost fairness cannot be achieved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Novel Approach for Unsupervised Sensor Error Classification

引用

Procedia CIRP 2024年 130卷 206-213页

作者： Samiha Durnagöz Tobias Eberhardt Felix Fuchs Peter Reimann Thomas Ackermann Maik Fiedler Mathias Mayer Marco F. Huber Alexander Verl Graduate School of Excellence advanced Manufacturing Engineering GSaME University of Stuttgart Nobelstraße 12 70569 Stuttgart Germany Institute of Industrial Manufacturing and Management IFF University of Stuttgart Allmandring 35 70569 Stuttgart Germany AUDI AG NSU-Straße 1 74172 Neckarsulm Germany Fraunhofer Institute for Manufacturing Engineering and Automation IPA Nobelstraße 12 70569 Stuttgart Germany Institute for Control Engineering of Machine Tools and Manufacturing Units ISW University of Stuttgart Seidenstraße 36 70174 Stuttgart Germany J. Schmalz GmbH Johannes-Schmalz-Straße 1 72293 Glatten Germany Institute of Parallel and Distributed Systems (IPVS) Universitätsstraße 38 70569 Stuttgart Germany

The precise detection and classification of sensor errors is crucial for the optimal operation of smart manufacturing systems. Sensor errors, which can manifest in various forms, such as drifts or constant values, significantly challenge data integrity. This work introduces a novel approach that combines unsupervised learning and system simulations to detect and classify relevant sensor errors. Our approach avoids extensive manual labeling and uses only error-free simulated time series data to create initial embeddings of sensor data. This establishes a baseline for normal sensor behavior. To facilitate a precise identification of error types through a classifier, simulated sensor errors are introduced to create context-aware embeddings. We further analyze and compare conventional to advanced embedding methods such as TS2Vec. We apply our approach to a synthetic dataset obtained from a vacuum handling system and accurately identify and classify relevant sensor error types. Our work contributes to advancing operational efficiency and reliability in automated manufacturing processes within the Industry 4.0 framework by improving the detection capabilities for sensor errors.

关键词： Unsupervised learning Sensor errors TS2Vec System simulation

来源：评论

学校读者我要写书评

暂无评论

Cost Fairness for Blockchain-Based Two-Party Exchange Protocols

Cost Fairness for Blockchain-Based Two-Party Exchange Protoc...

引用

IEEE International Conference on Blockchain (Blockchain)

作者： Matthias Lohr Benjamin Schlosser Jan Jürjens Steffen Staab CCRDMT University of Koblenz-Landau Koblenz Germany Technical University of Darmstadt Darmstadt Germany Fraunhofer ISST Dortmund Germany Institute for Parallel and Distributed Systems (IPVS) University of Stuttgart Germany University of Southampton Southampton United Kingdom

ISBN: (数字)9780738104959

ISBN: (纸本)9781665423229

Blockchains can guarantee fairness during the exchange of digital goods such that in a two-party exchange no one is defrauded by a malicious opponent. While several notions of fairness have been discussed in the literature, they all ignore that damage cannot only be incurred by the malicious failure of the exchange, but also by an unfair allocation of transaction costs. To address this issue we: 1. define the novel concept of cost fairness, which 2. builds on the notion of maximum cost matrices that formalize transaction costs in different combinations of benevolent and malicious behavior. 3. We show how limited notions of cost fairness can be achieved by modifying an existing exchange protocol or using a container protocol. In particular, we also provide 4. a tool that let us predict the maximum cost matrix for a specific protocol execution and, thus, gives trade exchange parties the possibility to weigh not only the value of transaction of exchanged goods but also the associated transaction costs.

关键词： Conferences Smart contracts Blockchain Containers Tools Data models Resource management

来源：评论

学校读者我要写书评

暂无评论

A Graph-based Approach to Manage CAE Data in a Data Lake

引用

Procedia CIRP 2020年 93卷 496-501页

作者： Julian Ziegler Peter Reimann Florian Keller Bernhard Mitschang Graduate School of Excellence advanced Manufacturing Engineering (GSaME) University of Stuttgart Nobelstrasse 12 Stuttgart 70569 Germany MANN+HUMMEL Schwieberdinger Strasse 126 71636 Ludwigsburg Germany Institute for Parallel and Distributed Systems (IPVS) University of Stuttgart Universitaetsstrasse 38 70569 Stuttgart Germany

Computer-aided engineering (CAE) applications generate vast quantities of heterogeneous data. Domain experts often fail to explore and analyze these data, because they are not integrated across different applications. Existing data management solutions are rather tailored to scientific applications. In our approach, we tackle this issue by combining a data lake solution with graph-based metadata management. This provides a holistic view of all CAE data and of the data-generating applications in one interconnected structure. Based on a prototypical implementation, we discuss how this eases the task of domain experts to explore and extract data for further analyses.

关键词： data lake data management metadata graph graph database computer-aided engineering simulation data

来源：评论

学校读者我要写书评

暂无评论

Incorporating Economic Aspects into Recommendation Ranking to Reduce Failure Costs

引用

Procedia CIRP 2020年 93卷 747-752页

作者： Vitali Hirsch Peter Reimann Bernhard Mitschang Daimler Truck AG Hanns-Martin-Schleyer-Str. 21-57 683058 Mannheim Germany Graduate School of Excellence advanced Manufacturing Engineering (GSaME) Nobelstr. 12 70569 Stuttgart Germany Institute for Parallel and Distributed Systems (IPVS) Universitaetsstr. 38 70569 Stuttgart Germany

Machine learning approaches for manufacturing usually offer recommendation lists, e.g., to support humans in fault diagnosis. For instance, if a product does not pass the final check after the assembly, a recommendation list may contain likely faulty product components to be replaced. Thereby, the list ranks these components using their probabilities. However, these probabilities often differ marginally, while economic impacts, e.g., the costs for replacing components, differ significantly. We address this issue by proposing an approach that incorporates costs to re-rank an initial list. Our evaluation shows that this approach reduces fault-related costs when using recommendation lists to support human labor.

关键词： decision support predictive analytics quality control End-of-Line testing classification fault isolation failure costs

来源：评论

学校读者我要写书评

暂无评论

Towards Scalable k-out-of-n Models for Assessing the Reliability of Large-Scale Function-as-a-Service systems with Bayesian Networks

Towards Scalable k-out-of-n Models for Assessing the Reliabi...

引用

IEEE International Conference on Cloud Computing, CLOUD

作者： Otto Bibartiu Frank Dürr Kurt Rothermel Beate Ottenwälder Andreas Grau University of Stuttgart Institute for Parallel and Distributed Systems (IPVS) Stuttgart Germany Robert Bosch GmbH Bosch IoT Cloud Stuttgart Germany

Typically, Function-as-a-Service (FaaS) involves state-less replication with very large numbers of instances. The reliability of such services can be evaluated using Bayesian Networks and k-out-of-n models. However, existing k-out-of-n models do not scale to the larger number of hosts of FaaS services. Therefore, we propose a scalable k-out-of-n model in this paper with the same semantics as the standard k-out-of-n voting gates in fault trees, enabling the reliability analysis of FaaS services.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Implementation of the MIALinx User Interface for Future Manufacturing Environments

引用

Procedia CIRP 2019年 81卷 606-611页

作者： Dominik Lucke Frank Steimle Emir Cuk Michael Luckert Matthias Schneider Daniel Schel Fraunhofer Institute for Manufacturing Engineering and Automation IPA Nobelstraße 12 70569 Stuttgart Germany Hochschule Reutlingen ESB Business School Alteburgstr. 150 72762 Reutlingen Germany University of Stuttgart Institute for Parallel and Distributed Systems IPVS Universitätsstraße 38 70569 Stuttgart Germany

The flexible and easy-to-use integration of production equipment and IT systems on the shop floor becomes more and more a success factor for manufacturing to adapt rapidly to changing situations. The approach of the Manufacturing Integration Assistant (MIALinx) is to simplify this challenge. The integration steps range from integrating sensors over collecting and rule-based processing of sensor information to the execution of required actions. This paper presents the implementation of MIALinx to retrofit legacy machines for Industry 4.0 in a manufacturing environment and focus on the concept and implementation of the easy-to-use user interface as a key element.

关键词： manufacturing smart factory Industrie 4.0 manufacturing service bus rules integration user interface

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：