Search

Uncoordinated decisions in production networks lead to inefficient structures that endanger a company’s competitiveness. To stay competitive in the future, manufacturing companies must focus their production and migrate their production network into agile structures. This paper presents a methodology for migration planning of production networks. First, strategic priorities and capabilities for the production network are formulated. To develop these priorities and capabilities, production resources are identified and combined to desired network configurations. Second, these configurations are transferred into a stochastic- dynamic optimization model to identify a cost-optimal migration strategy. The approach has been validated in the automation industry.

As vixotrigine (1) entered a later clinical phase for trigeminal neuralgia (Zakrzewska, J. M.; et al. Lancet Neurol. 2017, 16, 291−300), the development of a sustainable late-stage process was requ...

In high-tech production, companies often deal with the manufacture of assemblies with quality requirements close to the technological limits of manufacturing processes. The article shows an approach of a virtual in-line inspection, predicting the products functionality. An artificial neural network (ANN) fed with product characteristics and process data as well as the resulting functional fulfillment of the product is trained for virtual function prognosis. Through the preventive identification of defective products before the final assembly step, components can be recovered and returned to serial production. By optimizing the parameters of the ANN, incorrect classifications are reduced and the practical applicability is ensured. The approach is demonstrated in an industrial application in the automotive industry.

Manufacturing companies operating in global production networks face rising complexities and increasing susceptibilities to disruptions. For coping with disruptions, companies are in need of a holistic, comprehensive disruption management, involving all network actors to find optimal measures. However, today’s disruption management approaches are characterized by intuitive, experienced-based reactions, limiting themselves to solely the production or the logistics perspective and hence not permitting an overarching reaction. Therefore, this paper presents an integrated approach to disruption management, combining the production and logistics perspectives. It incorporates DoE and metamodelling methods in a simulation model to enable efficient, robust decision-making in highly complex environments.

Lean Management focusses on the elimination of wasteful activities in production. Whilst numerous methods such as value stream analysis or spaghetti diagrams exist to identify transport, inventory, defects, overproduction or waiting, the waste of human motion is difficult to detect. Activity recognition attempts to categorize human activities using sensor data. Human activity recognition (HAR) is already used in the consumer domain to detect human activities such as walking, climbing stairs or running. This paper presents an approach to transfer the human activity recognition methods to production in order to detect wasteful motion in production processes and to evaluate workplaces. Using sensor data from ordinary smartphones, long-term short-term memory networks (LSTM) are used to classify human activities. Additional to the LSTM-network, the paper contributes a labeled data set for supervised learning. The paper demonstrates how activity recognition can be included in learning factory training starting from the generation of training data to the analysis of the results.

In high-tech production, companies often deal with the manufacture of assemblies with quality requirements close to the technological limits of manufacturing processes. The article shows a digital DNA approach, integrating a digital product twin into the digital twin of the production to enable improved quality control cycles, at organizational level. A statistical meta-model for real-time application is derived from simulation and experimental data. To optimize its performance, its uncertainty according to GUM is minimized. Applying the meta-model, the cyber-physical production system is optimized regarding technical and economic criteria. The approach is demonstrated in an industrial application in the automotive industry.

Multinational companies deal with production processes in various countries by operating global production networks. These production processes are allocated to production plants with different levels of autonomy regarding strategic and operative decisions. Typically, each plant and the whole network are managed by one or more network managers who have to deal with a decision overload in their daily business. 50% of their decisions are made in less than 9 minutes and only a small amount of decision tasks are dealt with for more than one hour. To reduce this dilemma, it was found that the distribution of decision autonomy can be enhanced. It depends on the company’s strategy and complexity dimensions in global production networks. However, so far there is little evidence on how to better distribute decision autonomy in global production networks in detail. Furthermore, it is not transparent at what level of cetralism a global production network should be managed without cutting the capabilities of production plants. This paper presents a methodology, which examines relevant strategy dimensions and derives guidance on how to distribute decisions in global production networks. First, the network and production strategies of global production networks are classified. Second, relevant complexity dimensions and decisions are introduced. Third, the influence of the distribution of decision autonomy on strategy dimensions is quantified by an impact model. Furthermore, the effect of complexity on the distribution of decision autonomy is quantified by an impact model. Here, the integration of empirical data was used to validate the different influences. Finally, the ideal distribution of decision autonomy for specific production plants in the global production network is derived. The methodology is applied in an industrial use case to prove its practical impact.

An increasingly uncertain and dynamic competitive environment is challenging industrial companies nowadays. Against this backdrop, companies are focusing on their core competences. They organize their production in global production networks. While the competitiveness of production networks could be maintained for a long time by optimizing individual production sites, the overall network is increasingly becoming the focus of attention. In particular, the elimination of redundant production technologies offers the potential to exploit economies of scale, to bundle technology-specific competences and to achieve an increase in efficiency. The purely mathematical optimization models disseminated in research are unable to consider all the sub tasks of planning. For this reason, this article proposes a practice oriented methodology for reallocating production technologies to production locations in global production networks. The procedure consists of three phases: the investigation of current production technology-to-site allocation in the production network, the generation and planning of alternative reallocations as well as the evaluation of reallocations. For testing its practical suitability, the procedure is exemplary applied to the global production network for forging processes of a medical device manufacturer.

The concept of digital twins promises high potentials for product design, manufacturing, user experience and recycling. Thus, digital twins have received increasing interest in academia and industry. However, the actual benefits of digital twins remain in many cases unclear. This article aims to summarize selected recent developments in this field and demonstrate use cases from different phases of the product lifecycle. For that purpose, examples from the design, manufacturing, use and recycling phase are presented. In a subsequent discussion, ideas for new value stream patterns using digital twins are envisioned and research questions are derived.

Companies struggle to overcome the difficulties stemming from the dynamic environment of global production due to the specific conditions in different regions. Particularly, insufficient know-how about a regionalized implementation strategy of smart automation (SmAu) technologies is one significant difficulty for enterprises. Thus, developing a key performance indicator (KPI) oriented, regionalized implementation strategy for smart automation technologies is increasingly important. In this context, a new approach is exposed to systematically investigate and identify the interdependencies among location factors, smart automation technologies, and KPIs. Firstly, the environment consisting of location-related factors, KPIs and smart automation technologies is defined in detail. Further, a Catalog quantifies the influence of different regions in China. Secondly, important aspects to model the qualitative and quantitative interdependencies in a multimethod simulation are introduced. Subsequently, an approach to analyze suitable implementation strategies is presented. A case study based on a production line for digitalized production technology is used to validate the proposed approach.

Composite lightweight components of carbon fiber reinforced plastics (CFRP) provide great potential for the automotive industry. Yet, due to their complexity, there is a high amount of quality defects in their production processes. Thus, this article deals with an approach for in-process metrology of CFRP components by means of an eddy-current sensor array. First, the development of the sensor design and its in-process integration is introduced. Second, a method for the quantitative evaluation of the measurement data as well as the measurement uncertainty is elaborated. Finally, a machine learning approach for the classification of crucial defects is shown.

Micro-transmissions, consisting of micro-gears with a module To enable a time-efficient application of the model for industrial series production, in this article the application of a machine learning approach based on artificial neural networks is investigated. The uncertainty of the model is evaluated according to the principles of the Guide to the Expression of Uncertainty in Measurement (GUM).

Companies are confronted with increasing product quality requirements to manufacture high quality products, close to technological limits, in an economic way. The implementation of adaptive quality control strategies (QCS) in production offers an approach to cope with this challenge. In this paper, new function-oriented QCS by means of selectively assembling multiple components are demonstrated based on a functional product model. The implementation of QCS for fuel injector systems, which must fulfil narrow tolerances regarding the product’s functionality, show benefits in quality and cost-effectiveness. In the approach a functional model of the product and a simulation of the production system are implemented.

In the context of Industry 4.0, the development of maintainable and scalable applications becomes a core activity to master for industrial companies. To offer convincing trainings, learning factories have to coordinate the development of IT and hardware solutions with training concepts. The development of maintainable, modular and stable IT solutions in coherence with the hardware is the basis for good trainings. Especially learning factories face the challenge to work with part-time staff that is unfamiliar with large IT and hardware projects, web technology or PLC programming and development methods. In this paper we present a model how to introduce agile methods step-by-step to teams that have to cope with the challenge of a new project paired with a new technology and development methodology without overloading the team. An adapted teaching concept to introduce these development methods has been developed. For each of the maturity levels, tools and processes are presented as well as criteria to indicate when the team is ready to pass to the next level. The paper also presents the results of an application of the model to the development team at the Learning Factory Global Production at wbk.

Micro-transmissions are used in manifold industrial applications, e.g. the medical industry. They consist of micro-gears with a module ≤200 μm showing large shape deviations compared to their size. Their main function is a proper operation over the required lifetime. Thus, for micro-gears lifetime prognosis dependent on their specific geometrical deviations is particularly important. For this purpose, a method is introduced to enable a suitable function-oriented evaluation of micro-gear measurements. High-precision 3D measurements are processed by finite element simulation to calculate realistic loads, and are correlated with experimental data of lifetime experiments. The uncertainty is evaluated according to GUM and Bayesian statistics.

Bottlenecks in production lines are often shifting and thus hard to identify. They lead to decreased output, longer throughput times and higher work in progress. Go & See is a well-established Lean practice where managers go to the shop floor to see the problems first hand. Mixed reality is a promising technology to improve transparency in complex production environments. Until recently, mixed reality applications have been very demanding in terms of computing power requiring high performance hardware. This paper presents an approach for real-time KPI visualization using mixed reality for bottleneck identification in production lines relying on the bring-your-own device principle. The developed application uses image recognition to identify work stations and visualizes cycle times and work in progress in augmented reality. With this additional information, it is possible to discern different root causes for bottlenecks, for example systematically higher or varying cycle times due to breakdowns. This solution can be classified according to the acatech industry 4.0 maturity model as a level 3 - transparency - application. It could be shown that the identification of bottlenecks and underlying reasons has been improved compared to standard Go & See.

As a result of the increasing interdependencies within global production networks, the importance of an improved coordination of the associated individual processes is continuously rising. In particular, enormous potential is expected from a stronger integration of production and logistics planning and control. While previous approaches have mainly dealt with the integration of production planning and distribution logistics planning, this paper pursues the goal of integrating the procurement logistics and the production perspectives, considering both predictive and reactive components. To achieve this goal, an initial framework for the integrated procurement logistics and production planning and control is developed and exemplified in the context of an aircraft manufacturer within the scope of this paper.

As today manufacturing is not only subject to a single factory, but a network of globally distributed production sites, the goal-oriented encouragement of professional capacities is the motivation for the Learning Factory on Global Production (LGP). In this context, the design of a competency-based and action-oriented didactic and methodological concept is a prerequisite for sustainable learning results and for the development of self-determined problem solving skills. The presented paper gives an overview to the didactic and methodological design approach of the LGP. The integrated modular concept of e-learning and application in the learning factory environment supports self-directed learning and implemented by structuring the teaching/ learning process according to the model of complete action.

Production Management is an important issue for organizations that spend considerable amounts of investment annually on personnel training. Especially in the era of Industry 4.0 and Intelligent Manufacturing, considering the exponential growth of new knowledge and information, personnel need to update and supplement the necessary knowledge. Nevertheless, there is a lack of adequate methodology for executing trainings in the field of production management. This paper aims to develop a method for executing production management training which combines online learning and offline training as well as practical parts by using an Industry 4.0 Innovation Center equipped with model devices. A procedure is to first starting with an E-Learning module containing basic knowledge, accessible on the Learning Platform Moodle. Secondly, an on-line survey is created to collect expectations and requirements. Then the training schedule is carried out for execution of professional training. The training part in the innovation center will contribute to build up the basis for adaptions of the training knowledge to practical need of a company. Lastly the test and evaluation is conducted via virtual team room (Vitero). A case study based on training service provider is used to validate the feasibility of the approach. The derived results are presented and conclusions are discussed.

Multi-variant products to be assembled on mixed-model assembly lines at locations within a production network need to be scheduled locally. Scheduling is a highly complex task especially if it simultaneously covers the assignment of orders, which are product variants to be assembled within a production period, to assembly lines as well as their sequencing on the lines. However, this is required if workers can flexibly fulfill tasks across stations of several lines and, thus, capacity of workers is shared among the lines. As this is the case for final assembly of the Airbus A320 Family, this paper introduces an optimization model for local order scheduling for mixed-model assembly lines covering both assignment to lines as well as sequencing. The model integrates the planning approaches mixed-model sequencing and level scheduling in order to minimize work overload in final assembly and to level material demand with regard to suppliers. The presented model is validated in the industrial application of the final assembly of the Airbus A320 Family. The results demonstrate significant improvement in terms of less work overload and a more even material demand compared to current planning.

Enabled by the development of internet technologies, cyber-physical production systems (CPPS) are expected to open up entirely new possibilities to improve the efficiency of existing assembly systems of industrial companies. Nevertheless, realizing the potential of CPPS still remains a difficult task for small and medium-sized enterprises (SME), given the high variety of improvement possibilities offered by CPPS enabling technologies and the limited resources for their deployment. Hence, it is necessary to develop an implementation strategy of CPPS. Meanwhile, the consideration of location factors could support industrial companies to identify the appropriate CPPS implementation strategy since the location factors highly effect assembly system environment. In this context, a new approach to analyse the influence of location factors on the implementation of CPPS is exposed in this paper, which aims at investigating and identifying of relationships in between. Firstly, an application map of CPPS is generated. Secondly, the manufacturing industry status analyzed and subsequently a catalog of currently important location factors for the assembly systems are identified. Then a qualitative model of a relational analysis is established by an agglomerative hierarchical clustering algorithm. An industrial case study is used to demonstrate the applicability and the validity of the proposed approach.

Based on the fundamental principle of teaching psychology that retentiveness increases if students actively apply learning topics rather than only attend oral or visual presentations, the concept of learning factories becomes more and more popular. Academic education in the field of production science is imparted by means of real-world manufacturing facilities. By applying the manufacturing process of a real product, students or professionals incorporate the learning contents effectively and gain consciousness about their practical implications. Most learning factories are focused on lean manufacturing, lean administration or resource efficiency. As today manufacturing is not only subject to a single factory, but a network of globally distributed production sites, at the wbk Institute of Production Science,currently, a learning factory dealing with the topic of global production is developed. On the one hand, the curriculum of the Learning Factory Global Production (LGP) involves the specifics of local production sites with different location factors, such as different degrees of automation, cost structures and qualification levels, and their effects on the reconfigurability of the production systems. On the other hand, the interaction of the production sites in a globally distributed production network and the strategic configuration of the network are also subject to the curriculum. The manufacturing processes are exemplified by the assembly of an automotive e-motor with transmission in the learning factory on global production. The learning factory is realized in cooperation with the Robert Bosch GmbH.

In high-tech production, companies often deal with the manufacturing of assemblies with quality requirements close to the technological limits. Selective and adaptive production systems are means to cope with this challenge. In this context new measurement technologies and IT-systems offer the opportunity to generate and use real-time quality data along the process chain and to control the production system adaptively. In this article, a holistic matching approach to optimize the performance of selective and adaptive assembly systems is presented and its industrial application within an automotive electric drive assembly is demonstrated.

Companies in the manufacturing industry today are faced with increasing challenges with respect to cost effectiveness, lead time and quality of the production system. Dealing with these contradictory goals, an important task is the selection of suitable solutions for the integration of inspection processes within the process chain, which are necessary to ensure the required production quality. For this, supportive and easily applicable planning techniques are required to analyze and design the configuration of a respective process chain. Value Stream Mapping (VSM) is a state of the art tool which is very often used for this by professionals. It, however, is not capable of addressing the issue of a suitable integration of testing processes within the process chain. Yet, this provides valuable potential to facilitate the identification of effective testing equipment, testing strategies and quality control loops. Therefore, in this article an innovative approach called Quality Value Stream Mapping (QVSM) is presented. Based on the design elements of VSM, it provides a suitable tool for the visualization, analysis and design of quality assurance measures within process chains in manufacturing. The implementation of the developed approach is exemplarily shown for a complex value chain of a manufacturer in the electronic industry.

Companies operating assembly systems in global production networks constantly have to deal with change drivers. For the design of adaptable assembly systems, change drivers can be considered as fluctuating KPIs, such as labor costs, as well as changing KPI targets, such as rising quality requirements. In this paper, a concept for the design of changeable assembly lines with scalable automation is introduced and applied to the Learning Factory Global Production at KIT. The change of the automation level over time is based on an ex ante evaluation and ex post performance assessment of the impact of change drivers.

The development and manufacturing of highly precise micro-mechatronic systems, such as MEMS applications, is a challenging task due to the complexity and variety of their manufacturing technologies, as well as their high quality requirements. Within the context of the product engineering process of micro-mechatronic systems, quality inspection by means of production measurement technology is a crucial factor. This paper presents a survey of the challenges regarding quality inspection of micro-mechatronic systems. Furthermore, a Design for Testability approach for these types of products is described and exemplary applications of its implementation are shown.