The early stages of manufacturing innovation are characterized by ambiguity and uncertainty, often named as the fuzzy-front end of innovation. The practice of model-based systemsengineering (MBSE) in manufacturing ca...
详细信息
ISBN:
(纸本)9798350358810;9798350358803
The early stages of manufacturing innovation are characterized by ambiguity and uncertainty, often named as the fuzzy-front end of innovation. The practice of model-based systemsengineering (MBSE) in manufacturing can aid the reduction of ambiguity, by capturing technical requirements and system specification in a centralized modeling environment, sharable by all relevant stakeholders. However, challenges often arise when attempting to achieve this type of collaborative work environment between the various distinct technical and less/non-technical teams, with particular knowledge sets, professional backgrounds, and motivations. Furthermore, manufacturing system design can greatly benefit from early and extensive experimentation, which can be achieved through the building of digital models and performance of simulation runs, to access different considerations and compare alternatives relating to system architecture, material flow and production planning. Despite simulation technologies' high maturity level and the large number of tools available, their effective implementation in the manufacturing sector continues to be very limited, with low adoption levels being often attributed to a persisting lack of the necessary competencies, along with difficulties in the integration of modeling practices within the enterprises' workflows. This paper exemplifies the use of discrete event simulation as a centralized tool for supporting the early stages of manufacturing system design for an alkaline water electrolyser system, laying the foundations for future work on the development of a centralized engineering model and utilization of MBSE to achieve domain interoperability.
The mechanical-electrical-hydraulic systems are developing towards automation and intelligence, which raise higher requirements for accurate and smart fault diagnosis of the core component in these systems. Traditiona...
详细信息
In the modern digital landscape, data plays a crucial role in the competitiveness and efficiency of organizations. data governance, which involves managing and ensuring data quality, faces increasing challenges due to...
详细信息
Decentralised, modular production with the aim of individualised products leads to a more flexible production setup which, however, also influences the handling of faults and failures. Since faults occur rarely compar...
详细信息
ISBN:
(数字)9783031744853
ISBN:
(纸本)9783031744846;9783031744853
Decentralised, modular production with the aim of individualised products leads to a more flexible production setup which, however, also influences the handling of faults and failures. Since faults occur rarely compared to nominal behavior of Cyber-Physical Production Modules (CPPM), it is difficult in common manufacturing environments and even harder in skill-based production to gain experience and knowledge about faults and the context they occur in. Hence, leveraging knowledge and data from multiple CPPM proves beneficial, facilitating the storage of acquired information regarding faults and their context in federated knowledge bases. However, although different approaches tackle the communication between distributed knowledge bases, the use for distributed knowledge-based fault detection and diagnosis in skill-based production environments remains mainly unseen. In this paper the focus lies on the development of a communication scheme that enables automated communication between fault detection and fault diagnosis components for a decentralised control setup to make distributed knowledge about faults accessible. This includes the definition of fault detection and fault diagnosis components and their offered services which encapsulate different forms of knowledge representations. For the communication between the components, a unified model is elaborated, and the required information is identified. An integration in a holonic manufacturing system of SmartFactory(KL) is presented and an outlook for further research is given.
As a multi-staged digital manufacturing process, Additive manufacturing (AM) inherently benefits from data analytics (DA) decision-making opportunities. The abundance of data associated with the various observations a...
详细信息
ISBN:
(纸本)9780791887295
As a multi-staged digital manufacturing process, Additive manufacturing (AM) inherently benefits from data analytics (DA) decision-making opportunities. The abundance of data associated with the various observations and measurements taken throughout the design-to-product transformation creates ample opportunities for iterative, process improvements. To best formulate and address these opportunities, knowledge needs to be strategically and deliberately managed for efficient DA development. However, knowledge in AM is broad and comparatively sparse, making it difficult to create robust DA solutions. Also, existing methods for knowledge management in AM are often case-dependent. To address such challenges, this paper proposes a novel framework to manage case-independent knowledge for AM data analytics. The proposed framework consists of two phases: a knowledge-identification phase and a knowledge-representation phase. A knowledge architecture is defined to provide a reference for discovering knowledge that facilitates AM data analytics. In the knowledge identification phase, the architecture is used to facilitate the identification of actionable knowledge relevant to a specific DA use case. In the knowledge representation phase, ontologies are used for representing and linking that identified knowledge. A case study of application scenarios demonstrates how actionable knowledge is identified, represented, and managed by the framework. The framework enhances efficiency of AM data analytics development and enables knowledge sharing, understanding and reuse in AM data analytics activities.
In modern manufacturing enterprises, the issue of temporary leave and job rotation of frontline production employees affects production scheduling. To ensure safety and production continuity, companies often use subst...
详细信息
ISBN:
(纸本)9798350386783;9798350386776
In modern manufacturing enterprises, the issue of temporary leave and job rotation of frontline production employees affects production scheduling. To ensure safety and production continuity, companies often use substitute workers. However, due to the subjectivity of managers, the substitute workers may not have the required knowledge, skills, and abilities as the requirements of the position. To address this issue, an AHP-GA-BP (Analytic Hierarchy Process-Genetic Algorithm-Backpropagation Neural Network) model is proposed. This model is based on the KSAO (knowledge, skills, abilities, other indicates that can have an impact on work) model and combines the Analytic Hierarchy Process (AHP), Genetic Algorithms (GA), and Backpropagation Neural Network (BP). AHP is used to evaluate and rank the characteristics, obtaining accurate and reasonable weights for the features. Then, the sample data is further optimized and input into the GA-BP model for training. The model is trained and tested using real data from a digital system in a certain process manufacturing company. Experimental results show that the AHP-GA-BP model achieves a 20.09% and 74.85% improvement in prediction accuracy compared to the GA-BP model and BP neural network, respectively, in this specific task.
With the proposed"Made in China 2025" strategy, it is especially vital to transition from traditional manufacturing to high-quality, efficient intelligent manufacturing (IM). In China's manufacturing ind...
详细信息
Deep learning methods have been widely used and proved effective in defect prediction in Additive manufacturing (AM) to ensure process stability and part quality. However, the success of deep learning models depends h...
详细信息
ISBN:
(纸本)9780791888117
Deep learning methods have been widely used and proved effective in defect prediction in Additive manufacturing (AM) to ensure process stability and part quality. However, the success of deep learning models depends heavily on meticulous training, which usually requires a large homogenous dataset. This poses a challenge for the AM industry where many small- and medium-sized enterprises (SMEs) play a crucial role. On one hand, AM parts are usually customized or one-of-a-kind, and the process settings change frequently, resulting in heterogeneous datasets that usually vary by each print. On the other hand, SMEs are usually constrained by time and budget. As a result, they tend to focus on a limited number of process settings. This often leads to insufficient data collection, making SMEs difficult to properly train deep learning models independently. Therefore, there is a need to learn from the similarities in the physics of AM processes and the defect formation mechanisms, consequently enabling the potential knowledge sharing to learn across different AM scenarios. However, unique challenges in knowledge sharing arise from privacy concerns. Each design or print potentially contains sensitive proprietary information, such as process parameters, part geometries, and quality specifications. Such information may not be shared across different entities within the industry. In this context, Federated Learning (FL) emerges as a promising solution to this data scarcity and privacy challenge. FL is an innovative machine learning method that facilitates collaborative machine learning model training across multiple clients without sharing their locally stored data. In this paper, a FL framework is developed to predict section-wise heat emission, a critical process signature during the Laser Powder Bed Fusion (LPBF) for collaborative knowledge sharing across different manufacturing entities, without direct transfer of sensitive information. The framework learns the relationship b
Tool lifecycle management is critical to improving productivity and reducing costs in manufacturingsystems. To make efficient and accurate tool management decisions, it is necessary to have access to multiple related...
详细信息
ISBN:
(纸本)9798350358513;9798350358520
Tool lifecycle management is critical to improving productivity and reducing costs in manufacturingsystems. To make efficient and accurate tool management decisions, it is necessary to have access to multiple related data across software systems. However, the interpretation, integration, and analysis of multi-source data is challenged by structural and semantic heterogeneity. To address this challenge, this study proposes an ontology-based method for integrating multi-source heterogeneous data. The first step involves developing an ontology model and SysML model to create a standardized and semantically consistent description of material, information, processes, and their interactions within the manufacturing system. Next, an automatic extraction tool is developed for multi-source data instances based on the constructed models. The tool converts data instances into a structurally and semantically consistent format and stores them, along with the relationships, in a graph database. A case study is presented using an aeroengine fuel nozzle production line as an example to demonstrate data integration. Additionally, a decision aid application is developed for tool selection, a typical scenario in the tool lifecycle, to validate the proposed methodology's contribution to tool lifecycle management.
Quantum computing (QC) shows the potential to trigger a paradigm shift for numerous industries. As an emerging technology, methodological support for designing and developing QC-based applications is lacking. This pap...
详细信息
ISBN:
(纸本)9780998133171
Quantum computing (QC) shows the potential to trigger a paradigm shift for numerous industries. As an emerging technology, methodological support for designing and developing QC-based applications is lacking. This paper presents the results of a case study applying consortium research in order to perform a requirements engineering process for two QC-based applications in the manufacturing industry. The results show the differences between requirements engineering for QC applications and conventional software applications. The major findings point to the need for QC knowledge and best practices for a successful requirements engineering process and elaborate on the main differences between QC application- and software application requirements.
暂无评论