This article examines how force sensors must be positioned to detect incorrect positioning of forging blanks. For this purpose, simulative investigations are carried out on a die. Positions for possible sensor placement are applied in a grid pattern. The recorded force values of the respective sensors are analyzed to identify those sensors that are particularly suitable for reliably detecting incorrect positioning.
industry 4.0, digitalisation, process monitoring
Artificial intelligence (AI) can be used to reliably recognise errors, reduce defective parts and increase component quality. In the 'AutoPress' project, researchers have developed a system of sensors and AI that recognises 95 to 98 per cent of all process deviations.
Artificial intelligence, AI, process monitoring, error detection
Aim: We explore the use of the Open Research Knowledge Graph (ORKG) as such an infrastructure by demonstrating how engineers can utilize the ORKG in innovative ways for communication and (re-)use.
Method: For a use case from the Collaborative Research Center 1153 “Tailored Forming”, we collect, extract, and analyze scientific knowledge on 10 Tailored Forming Process Chains (TFPCs) from five publications in the ORKG. In particular, we semantically describe the TFPCs, i.a., regarding their steps, manufacturing methods, measurements, and results. The usefulness of the data extraction topics, their organization, and the relevance of the knowledge described is examined by an expert consultation with 21 experts.
Results: Based on the described knowledge, we build and publish an ORKG comparison as a detailed overview for communication. Furthermore, we (re-)use the knowledge and answer eight competency questions asked by two domain experts. The validation shows a clear agreement of the 21 experts regarding the examined usefulness and relevance.
Conclusions: Our use case shows that the ORKG as a ready-to-use infrastructure with services supports researchers, including engineers, in sustainably organizing FAIR scientific knowledge. The direct use of the ORKG by engineers is feasible, so the ORKG is a promising infrastructure for innovative ways of communicating and (re-)using FAIR scientific knowledge in engineering sciences, thus advancing this research field.
Tailored forming, scientific knowledge, communication, knowledge graph
As the number of product variants continues to grow, the need for flexibility in intralogistics is becoming increasingly apparent. One potential solution to this challenge is the use of cellular automated guided vehicles, which can be variably interconnected depending on the size of the product to be transported. This article presents an optimization model for solving a vehicle routing problem for cellular automated guided vehicles. Furthermore, a recursive method is presented that determines an optimal transport sequence based on the solution of the model. The optimization model is implemented in a specially developed model environment and solved for a dynamic, illustrative use case. Subsequently, logistical target variables are evaluated in order to assess the solution of the optimization model. The exemplary application of the optimization model demonstrates the feasibility of modeling cellular transportation with automated guided vehicles and evaluating its performance based on logistical target variables.
AGV, cellular transport units, optimization model, simulation, logistical target values
This work explores the challenges of fully automating in-house goods transport in environments where industrial trucks like forklift trucks remain necessary due to undefined load carrier positions and shapes. Imitation Learning (IL) is identified as a promising solution for vehicle control in repetitive tasks, yet its application in intralogistics is challenging by the dynamic complexity of industrial trucks and the large dimensional space involved. A Robot Operating System 2 (ROS2) framework is introduced, enabling the acquisition of driving data from both simulation environments and real-world demonstrators. The study also presents a network architecture combining a Convolutional Neural Network (CNN) with a Long Short-Term Memory (LSTM) network, facilitating end-to-end learning from spatial and temporal image data. The framework's effectiveness is evaluated using a dataset of expert driving maneuvers to assess the generalization potential of the IL-trained network in vehicle control in different scenarios. The research aims to demonstrate the utility of the proposed framework for data acquisition and validate IL as a control approach for industrial trucks that require generalization.
Imitation Learning, industrial truck automation, intralogistics, ROS2, load handling
Constantly increasing production volumes and new challenges in production environmentswith the same amount of space are forcing manufacturing companies to deal with the planning of production layouts. The problem often is a non-existent or outdated production layout plan. Autonomous multicopters can help by simplifying layout capture. That's why a voxel-based simulation is investigated to develop and train path planning algorithms with and without artificial intelligence. First, the temporal behavior and the resource utilization of the simulation is investigated. Then, the time factor of simulation is compared to real time and what advantages companies and developers have when using it.
UAS, digital twin, simulation, voxel-based, layout design
It was investigated how a grain refinement can be introduced into a cylindrical rolled piece by means of cross rolling in a flat-jaw design. For this purpose, a non-round shape was rolled in and out again. A theoretical preliminary investigation was used to define a suitable parameter field for experimental evaluation. Metallographic investigations showed that cross-rolling had a positive effect on the microstructure, and IPH will inform the client immediately if it becomes apparent that the estimated costs will be exceeded. The execution of work that leads to the estimated expenditure being exceeded shall only take place after consultation with the client.
Cross-Wedge Rolling, fine grain
A major advantage of flashless precision forging is the saving of resources by avoiding excess material. Especially materials with good flow properties, such as aluminum, sometimes lead to the formation of thin flash. Seals have been developed and experimentally tested to prevent thin flash, i.e. the flow of material into tool gaps. The study presented shows the effectiveness of different sealing materials and the ideal sealing geometry.
thin flash, Sealing concept, precision forging, forming technology, process optimization
It was investigated how a grain refinement can be introduced into a cylindrical rolled part with cross rolling in a flatjaw design. For this purpose, a non-circular shape was rolled in and out again. A theoretical preliminary investigation was used to define a suitable parameter field for experimental evaluation. Metallographic investigations showed that cross rolling had a positive effect on the microstructure.
process design, forming technology
This article examines the use of point clouds as a geometric data basis for factory planning and compares different mapping techniques for generating these point clouds. Data and information acquisition is a crucial step in factory planning and thus in developing efficient production processes. In this context, different mapping techniques are analysed: photogrammetry (using drones and action cameras) and LiDAR scans (performed both from drones and from the ground).
The methodology and results of this investigation are discussed in detail, highlighting the advantages and disadvantages of each mapping technique. The focus is on comparing the generated point clouds in terms of completeness, recognisability and geometric tolerance. This comparison provides valuable insights into which technique is best suited for the data acquisition of factory planning. The outlook of this paper includes the further development of recording techniques, particularly with regard to autonomously flying drones. In the future, these could enable more efficient and precise data acquisition for factory planning and thus further strengthen the basis for optimising production processes.
Drone, Photogrammetry, LiDAR, Point cloud, Factory planning, Data acquisition
Around half of the currently 30,000 active wind turbines in Germany will reach the end of their service life by 2030, which is generally defined by the manufacturer as 20 years of operation. The most common strategy for the subsequent use of a wind farm is repowering, provided this is (legally) possible at the respective location. One option for dealing with old turbines is to resell them. At the time of repowering, in Germany after an average of around 17 years, the wind turbines usually still have a remaining operating time of several years before critical parts such as generators fail.
This article presents a machine learning model for predicting the resale value of used wind turbines. This model can be used to approximately predict the resale value of comparable wind turbines based on certain input parameters such as the power output or the age of the wind turbines. The model was trained using an adjusted data set from an online trading platform for wind turbines. The necessary pre-processing steps such as the removal of extreme outlier values and the addition or replacement of missing or incorrect wind turbinespecific data from a second data source using a self-developed matching algorithm are presented. Finally, the prediction accuracy of different ML algorithms is tested using test data to find the best method for predicting the resale value of wind turbines.
Renewable energies, wind turbines, resale values, machine learning, data science
The article addresses a method for generating project schedules for factory relocations based on a planned layout concept in reorganisation projects. The layout is modelled in a cell grid with cells of uniform edge length. The factory objects are arranged in this grid for both the actual and the target layout state. Further input parameters relate to the required time and personnel resources according to several operations. A two-stage optimisation algorithm is presented, which first checks a given solution for layout feasibility and then generates and solves a project scheduling problem on this basis. Accordingly, elements of the facility layout problem are integrated into a resource constrained project scheduling problem. The process steps of generating and evaluating of the project schedule are further embedded in a genetic algorithm, which successively improves the solution. The project schedules for relocation are evaluated through a combination of the resulting project duration and the downtimes of all factory objects. The aim of the optimisation is to minimise both objective values considering a weighting factor. The article concludes by validating the method using a practical example.
factory planning, relocation planning, reorganisation, project scheduling, project planning
Increasing the service life and process reliability of systems plays an important role in terms of sustainable and economical production. Especially in the field of energy-intensive bulk forming, low scrap rates and long tool lifetimes are business critical. This article describes a modular method for AI-supported process monitoring during hot forming within a screw press. With this method, the following deviations can be detected in an integrated process: the height of the semi-finished product, the positions of the die and the position of the semi-finished product. The method was developed using the CRISP-DM standard. A modular sensor concept was developed that can be used for different screw presses and dies. Subsequently a hot forming-optimized test plan was developed to examine individual and overlapping process deviations. By applying various methods of artificial intelligence, a method for process-integrated detection of process deviations was developed. The results of the investigation show the potential of the developed method and offer starting points for the investigation of further process parameters.
Process monitoring, Wear, Hot forming, Predictive maintenance, Quality management
Wear due to thermal and mechanical stresses is one of the major causes of forging die failure. The assessment of die condition and the associated die life is usually based on experience. This paper presents a method to objectively predict the remaining life of a forging die. With this method a prediction based on optical measurements can be calculated. Practical tests show the possible applications. In addition, force measurements are performed and analyzed to determine how wear affects the force distribution in the die. The assessment based on optical measurements allows objective statements about the remaining tool life of forging dies. The analysis of the force measurements shows potential for predicting tool life but needs further investigation.
You can view the article here.
Process monitoring, wear, optical measurements, force measurement
A continuously growing number of product variants increases the demands on the flexibility of intralogistics transportation. One way to achieve greater flexibility is the use of cellular automated guided vehicles, which can be variably interconnected depending on the size of the product to be transported. This article explains the characteristics of cellular automated guided vehicles and the relationships between influencing variables of the cellular transport system and economic and logistical target variables.
Intralogistics, automated guided vehicles, cellular transport units
The resilience of a supplier is a fundamental factor in the design of supply chains. In supplier selection, resilience is not yet fully taken into account in the evaluation process. As part of a research project, the aim is to create a basis for this and thus integrate resilience as an evaluation criterion into the supplier selection process. The article aims to create an understanding of the various success factors of resilient supply chains.
Resilience, sustainability, supply chain management, risk management
Automated industrial trucks master difficult driving situations worse than humans – for now. New approaches based on artificial intelligence (AI) are intended to replicate human driving behavior and give automated systems more flexibility.
artificial intelligence, intralogistics, industrial trucks
Twelve interviews were conducted with IT system users and providers in order to investigate the adaptability of IT systems in manufacturing companies. A methodical evaluation of the interviews made it possible to identify factors that influence the service life of IT systems. The evaluation shows that, in addition to the technical characteristics, human and organizational aspects in particular are decisive for the long-term use of IT systems.
Information management, people and technology, software
Tools for implementing a systematic quality management are necessary for the use of material extrusion as an additive manufacturing process for products with high quality requirements. Well-defined quality classes are crucial for ensuring that the requirements for a product can be communicated transparently and that the existing properties can be evaluated. Furthermore, there is a lack of capable measurement equipment for the acquisition of process data during the production process. To address these challenges, the present paper introduces an image processing system that determines quality indicators for individual layers in terms of imperfect surface percentages and the number of imperfections. The central element of the hardware is an adaptive darkfield illumination, which leads to high-contrast images. In addition, five types of layer subareas are identified in a segmentation step. Unsupervised machine learning methods are then used to detect imperfections in each layer subarea. In the segmentation, the current layer can be distinguished from irrelevant image background regions with an F-measure of 0.981. For the layer-wise measurement of the quality indicators, relative measurement errors with standard deviations of 25 to 76.1% are found. After evaluating the capabilities of the image processing system, a proposal for limits of quality classes is derived by monitoring several material extrusion processes. For this purpose, three quality classes for each of the five layer subareas are deduced from the process scatter measured by the image processing system. The results are an important contribution to the industrialization of material extrusion in safety–critical areas such as medical technology or the aerospace industry.
additive manufacturing, material extrusion, quality classes, image processing, Process monitoring
Bees are an important part of local ecosystems. Many companies have beehives set up on their company premises and looked after by beekeepers as an ecological measure. Remote digital monitoring can be a useful way of reducing the workload while ensuring the health and continued existence of the bee colony. The developed prototype and AI-based object recognition offers beekeepers the opportunity to monitor activity levels in the hive, detect intruders such as wasps or recognize a drone brood at an early stage, which occurs when the queen bee dies.
AI, image-recognition, Bees
