Fakultät für Wirtschaftsingenieurwesen
Refine
Document Type
Language
- English (9) (remove)
Has Fulltext
- no (9)
Is part of the Bibliography
- no (9)
Keywords
- Industry 4.0 (3)
- sustainability (2)
- Bioeconomy (1)
- Forest Wood Value Chain (1)
- Generalization (1)
- Job Shop Scheduling (1)
- Markov Decision Process (1)
- ONEforest (1)
- Production Scheduling (1)
- Reinforcement Learning (1)
The digitalisation of production has a positive impact on manufacturing processes in terms of resource efficiency and environmental impact, particularly in the form of increased efficiency as well as cost and resource savings. However, the use of digitalisation technologies is also associated with efforts such as costs, CO2 emissions, and raw material consumption. When planning or deciding on the digitalisation of manufacturing systems, it is therefore necessary to assess whether these technologies pay off in terms of sustainability over their life cycle. This literature review (based on the PRISMA guidelines) analyses the relevance of sustainability assessment and its methods for the digitalisation of production in research. The review reveals that research focuses on the benefits of digitalisation technologies in manufacturing, while the assessment of efforts and their benefits is in its infancy. There is a need for further research on holistic assessment methods for digitalisation technologies. In particular, there is a lack of assessment methods that consistently link the economic and environmental dimensions of sustainability, and there is also a lack of guidance for the application of assessment methods in production.
Manufacturing industries require the efficient and voluminous production of high-quality finished goods. In the context of Industry 4.0, visual anomaly detection poses an optimistic solution for automatically controlled product quality with high precision. In general, automation based on computer vision is a promising solution to prevent bottlenecks at the product quality checkpoint. We considered recent advancements in machine learning to improve visual defect localization, but challenges persist in obtaining a balanced feature set and database of the wide variety of defects occurring in the production line. Hence, this paper proposes a defect localizing autoencoder with unsupervised class selection by clustering with k-means the features extracted from a pretrained VGG16 network. Moreover, the selected classes of defects are augmented with natural wild textures to simulate artificial defects. The study demonstrates the effectiveness of the defect localizing autoencoder with unsupervised class selection for improving defect detection in manufacturing industries. The proposed methodology shows promising results with precise and accurate localization of quality defects on melamine-faced boards for the furniture industry. Incorporating artificial defects into the training data shows significant potential for practical implementation in real-world quality control scenarios.
The scheduling of production resources (such as associating jobs to machines) plays a vital role for the manufacturing industry not only for saving energy, but also for increasing the overall efficiency. Among the different job scheduling problems, the Job Shop Scheduling Problem (JSSP) is addressed in this work. JSSP falls into the category of NP-hard Combinatorial Optimization Problem (COP), in which solving the problem through exhaustive search becomes unfeasible. Simple heuristics such as First-In, First-Out, Largest Processing Time First and metaheuristics such as taboo search are often adopted to solve the problem by truncating the search space. The viability of the methods becomes inefficient for large problem sizes as it is either far from the optimum or time consuming. In recent years, the research towards using Deep Reinforcement Learning (DRL) to solve COPs has gained interest and has shown promising results in terms of solution quality and computational efficiency. In this work, we provide an novel approach to solve the JSSP examining the objectives generalization and solution effectiveness using DRL. In particular, we employ the Proximal Policy Optimization (PPO) algorithm that adopts the policy-gradient paradigm that is found to perform well in the constrained dispatching of jobs. We incorporated a new method called Order Swapping Mechanism (OSM) in the environment to achieve better generalized learning of the problem. The performance of the presented approach is analyzed in depth by using a set of available benchmark instances and comparing our results with the work of other groups.
Randomization is currently a widely used approach in Sim2Real transfer for data-driven learning algorithms in robotics. Still, most Sim2Real studies report results for a specific randomization technique and often on a highly customized robotic system, making it difficult to evaluate different randomization approaches systematically. To address this problem, we define an easy-to-reproduce experimental setup for a robotic reach-and-balance manipulator task, which can serve as a benchmark for comparison. We compare four randomization strategies with three randomized parameters both in simulation and on a real robot. Our results show that more randomization helps in Sim2Real transfer, yet it can also harm the ability of the algorithm to find a good policy in simulation. Fully randomized simulations and fine-tuning show differentiated results and translate better to the real robot than the other approaches tested
SMEs are increasingly forced to shift to more sustainable manufacturing. Industry 4.0 can support the transformation and foster innovation. But, SMEs need solutions with a low barrier to entry in terms of investment, IT knowledge and capacities. A framework based on value and material flow analysis, low investments and user-oriented IT skills is proposed. As an example, it is implemented in the furniture industry and shows a digital twin in terms of monitoring the energy and material flows. In addition, a product-specific allocation of energy consumption, energy peak shaving and other applications are possible.
Increasing demand for energy-efficient means of transport has steadily intensified the trend towards lightweight components. Thermoplastic glass fiber composites (organo sheets) play a major role in the production of functional automotive components. Organo sheets are cut, shaped and functionalized by injection molding to produce hybrid components, such as those used in car door modules. The cutting process produces a considerable amount of production waste, which has thus far been thermally recycled. This study develops a closed mechanical recycling process and analyzes the different steps of the process. The offcuts were shredded using two shredding methods and implemented directly in the injection-molding process. Using tensile tests and impact bending tests, the material properties of the recycled materials were compared with the virgin material. In addition, fiber length degradation via the injection-molding process and the influence of the waterjet-cutting process on the mechanical properties are investigated. Recycled offcuts are both comparable to new material in terms of mechanical properties and usability, and are also economically and ecologically advantageous. Recycling polypropylene waste with glass fiber reinforcement in a closed loop is an effective way to reduce industrial waste in a sustainable and economical production process.
In countries with large import dependencies, several studies on strategically important metals have classified raw materials as being critical by combining supply and vulnerability risks. In case of availability, the most critical raw materials are metallic ores and industrial minerals. Within the scope of developing resource efficient buildings, innovative building technologies, such as renewable energy systems, are becoming increasingly integral parts of buildings. Since critical raw materials are essential components in building technologies, they will also play an important role in the housing sector.
In this context, it is crucial to consider how sustainability goals in the housing sector are affected by raw material criticality. This paper addresses supply and vulnerability risks of indium, which is contained in thin-film photovoltaics, in order to identify their impacts on future developments in the housing sector. Raw material availability is significantly influenced by the demand and supply of raw materials. To implement risk factors for a criticality assessment in sector-specific material flows, their impacts are analysed and modelled using a dynamic, macro-economic approach. Assuming different energy scenarios in the German housing sector up to 2050, the model reveals the raw material criticality by pointing out supply risks in order to derive recommendations.
The automotive value chain is experiencing significant transformation. Due to changed mobility use patterns and external conditions such as climate change, scarcity of resources and local air pollution, car manufacturers offer connectivity services and are forced to launch electric mobility. Most of the electric vehicles have additional, complementary products and services. As other market participants also engage in this field car manufacturers must decide whether they transform their business model. Therefore, the questions arise which are the determining key criteria for the business models, what are the business segments car manufacturers are engaged and how car manufacturers are preparing for the transformation. By using qualitative content analysis, key criteria for business model analysis are derived. The portfolio analysis of six car manufacturers in the German market unveils that a few car manufacturers are in an advanced state while others follow with a considerable lag, or even, are behind.