FG Automatisierungstechnik
Refine
Document Type
- Conference publication peer-reviewed (18) (remove)
Way of publication
- Open Access (2)
Keywords
- Automotive (1)
- Brandenburg (1)
- Business model innovation (1)
- Digital business model (1)
- Digital innovation hub (1)
- Geschäftsmodellinnovation (1)
- Human-Robot-Collaboration (1)
- Lebenszyklus (1)
- Mensch-Roboter-Kollaboration (1)
- Resilience (1)
Institute
BTU
The introduction of both mobile robots (MR) and AI-embedded mobile robots (AIMR) into the industry is very slow compared to other types of industrial robots (IR) and automation systems. Many scientific articles and studies are focused on the programming and design of MR. At the same time, integration issues, topical problems and related obstacles are almost entirely absent from the scientific literature. The authors of this paper acknowledge that the complete analysis of this area is a very challenging task. Therefore, for the purposes of this study, we focus on the local problem of analysing the introduction of MR and AIMR in small and medium-sized enterprises (SMEs). The authors offer the analysis of the current challenges and trends in the introduction of mobile robots into SMEs. They also propose solutions to these problems based on their own as well as external experience in the design, programming and implementation of mobile robots.
Albeit augmented reality (AR) technologies first have been discovered in the third quarter of the 20th century, their widespread use began just two decades ago. Existing paper trails show that AR has a wide range of industrial application: it simplifies human-machine communication, improves human-machine interfaces (HMI) for quick information exchange in training (including feedback to study the workflow), correction of errors, machine maintenance, assembly assistance etc. However, broader industrial acceptance of AR, prior to all by small and medium-sized enterprises (SMEs), recently faced considerable problems and the expansion of AR solutions does not match the high potential it has demonstrated. That results in a limited practical use, mainly for demonstration and advertising purposes. This short review is to present the state of the art of the industry, challenges that SMEs face in adopting AR technologies, and several practical examples of a (commercial) adoption of AR. Some prospects for further development of AR and its ongoing integration into industry are briefly discussed in the summary.
The transformation of the manufacturing sector towards Industry 4.0 is setting the scene for a major industrial change. Currently, the need for assisting companies in this transformation is covered by a number of maturity models that assess their digital maturity and provide indications accordingly. However, in order to provide operational recommendations to diverse companies, there is a need for making the assessment company-specific. To cope with this challenge, this paper provides an illustration of a new digital maturity assessment approach - 360 Digital Maturity Assessment - which is based on the Problem Based Learning (PBL) model.
This minitrack will address scientific and technical research on how traditional production systems are transformed into next generation smart production systems. It comprises a mix of presentations from leading industries (automotive and aircraft) as well as from well-known research institutions (Fraunhofer society, Germany) and universities (Brandenburg University of Technology Cottbus-Senftenberg, University of Bergamo, Aalborg University).
The minitrack will foster discussions concerning the strategies companies are adopting for managing such transformation, the need for new competences for sustaining it, the implementation of new digital technologies and their impact on the existing production systems.
The advent of the industrial digital transformation and the related technologies of the Industry 4.0 agenda has uncovered new concepts and terminology in the manufacturing domain. Clear definitions represent a solid foundation for supporting the manufacturing research community in addressing this field consistently. This paper addresses this need focusing on the “smart factory”. Starting from a review of the extant literature and integrating it with the outcome of a Delphi study, we propose a new definition of a “smart factory” and discuss its key characteristics. These are related to interconnectivity capabilities and adaptability to the surrounding environment in order to generate and appropriate value. Eventually, such characteristics are exemplified in an empirical context. The aim of this paper is to provide the research community with an updated definition of a smart factory taking both industrial and societal values into account. Furthermore, it may represent a reference for practitioners engaged in the digital transformation of their factories.
Abstract. As product specifications change, manufacturing processes have to adapt. In manual production tasks, the human worker is forced to adapt at the same pace. Fast-changing work tasks lead to high stress and therefore increase failures. Digital assistance systems aim to support the human workforce by providing assembly instructions at the right time and in the right place to reduce the cognitive load. The latest digital assistance systems provide multimodal humanmachine interfaces, such as augmented reality, haptic feedback, and voice control to provide information or react to the user’s input.However, those digital assistance systems require the manufacturing information themselves, which are mostly provided through text-based or graphical programming. Both manufacturing experts and programmers are needed to create a digital assistance system workflow or adapt it to changes. This process is costly, time-consuming, and inflexible. This work presents a gesture recognition based approach for a self-learning digital assistance system. Therefore, assembly gestures are classified based on anatomical grip descriptions. Assembly sequences are recognized and learned by the digital assistance system using machine learning techniques. The learned procedures are used to automatically generate work instructions and guide the worker through the assembly task.
Learning factories constitute a promising approach for the acquisition of specific competencies, especially in terms of a digital transformation of the economy. Respectively, a variety of such factories differing in technology, learning concept, and potential audience have evolved. A precise and recent overview of those does not exist. However, such an overview is required for the implementation of concrete political measures, a future-oriented development of the individual learning factories, and an adequate selection by the audience. For this purpose, the authors investigate the current state of the art of European learning factories in the context of digitization. Thus, the terminology and definition of learning factories are provided. Moreover, using a structured literature review, the factories and their operation mode are outlined. Subsequently, the authors evaluate whether the different factories can build the required competencies among the audience and thus, support a successful digital transformation. Additionally, expert interviews with learning factory operators are performed to obtain profound information on the performance of learning factories. The findings help to assess the pertinency of European learning factories and provide a trace for their future development.
In the industry, connecting machines and tools - also known as the industrial Internet of things (IIoT) - is an essential part of the digital transformation of a company. The aim is to increase the efficiency and predictability of complex processes. In manual and semi-automatic processes, imaging sensors can help to monitor conditions, gives automated feedbacks to a central system, and e.g. provide current information for a digital twin. However, when imaging sensors are integrated into established IIoT platforms, they quickly reach their system limits due to the multidimensionality and high update and data rates. This paper presents a software platform that enables decoupled automated image processing through the abstraction and contextualization of the sensor technology and its data as well as a plugin architecture. Analogous to edge computing, partial processing can already be performed close to the sensor node to condensate data and reduce network loads and latencies. Thereby, all these approaches increase the longevity, flexibility and scalability of multi-sensor systems and associated processing algorithms. Based on the generic structure of the sensor network, the user is provided with an intuitive user interface that is based on IIoT platforms and enables the integration of their processing pipelines even for non-experts, despite the high complexity of the data.
Digital Business Model Innovation in SMEs - Case Studies with DIH Support from Brandenburg (Germany)
(2020)
The fast and safe motion of goods and people is one of the foundations of the modern world. Jet aircraft is the fastest transport at the moment along with high-speed trains. Accordingly, both production and maintenance of aircrafts are an important task of our modern industrial environment as well. Modern aircraft engines require appropriate care and understanding of design and manufacturing. This is even more important, as the production of aerospace engines remains a manual process in many cases with limited data sources. Thus, quality control will need to take into account verification of manufacturing and assembly steps through specific checks and controls whilst implementing additional data sources. Automation of tasks still is at a low level. In this article a review of the challenges with regard to controls, automation and process and technical understanding for aerospace engine production and repair is provided. As this requires the collaboration of many teams and partners, an improvement and step change towards deeper understanding and process efficiency is required. As many operations remain manual, innovations for how humans interact with the technology and collaborate with an industrial environment are needed. The project in this article demonstrates the creation and usage of the proposed solutions for collaboration, troubleshooting and error correction.
Even though digitization is a topic of high interest and promises remarkable benefits for the German manufacturing industry, it is still in its development phase. Thus, necessities arising from it are not predictable, yet. A great obstacle that will occur is the appropriate qualification of employees at all levels, sectors and firm sizes. The integration of new technologies will have radical impact on the role and job profile of all staff members as they alter the requirements across the whole value chain and lead to new processes. Latest studies showed that an insufficient qualification of employees is one of the greatest challenges for a successful digitization. Thus, to regain a globally leading position in the manufacturing industry German companies, especially small and medium sized enterprises (SME), need to qualify their employees appropriately. The LTA-FIT model is an approach that aims to face these challenges.
The usage of convolutional neural networks has revolutionized data processing and its application in the industry during the last few years. Especially detection in images, a historically hard task to automate is now available on every smart phone. Nonetheless, this technology has not yet spread in the industry of car production, where lots of visual tests and quality checks are still performed manually. Even though the vision capabilities convolutional neural networks can give machines are already respectable, they still need well prepared training data that is costly and time-consuming to produce. This paper describes our effort to test and improve a system to automatically synthesize training images. This existing system renders computer aided design models into scenes and out of that produces realistic images and corresponding labels. Two new models, Single Shot Detector and RetinaNet are retrained under the use of distractors and then tested against each other. The better performing RetinaNet is then tested for performance under training with a variety of datasets from different domains in order to observe the models strength and weakness under domain shifts. These domains are real photographs, rendered models and images of objects cut and pasted into different backgrounds. The results show that the model trained with a mixture of all domains performs best.
Digitalization is widely accepted as a principal technological strategy and is expected to vastly impact and transform industry, predominantly by a substantial improvement of the entire value chain. However, its implementation process remains slow, in particular for small and medium-sized enterprises (SMEs), as they tend to have misconceptions about the complexity and expense of digitalization. The purpose of this research is to analyze factors that affect Brandenburg-State, DEU SMEs in their decision making process for an effective implementation of digital technologies in their daily business routine. The results of this study are based on a survey that was conducted in cooperation with fifty SMEs, which currently face problems in the areas of enterprise resource planning (ERP), CAD/CAM, automation, logistics and production/factory planning. Ten of these SMEs have decided to invest in digitalization measures and started the implementation process. Three of them were chosen for the case study to evaluate potential benefits, trade-offs and barriers hampering the implementation of digitalization technologies. These three companies are considered representative items, since their needs and measures investigated were identical to most of the companies of this survey. It was found that the overwhelming majority of the companies regard ERP implementation as their highest priority when investing in the most cost-effective technology.