Fakultät Wirtschaftsingenieurwesen
Refine
Year of publication
Document Type
- Article (75)
- Conference Proceeding (62)
- Part of a Book (24)
- Book (22)
- Bachelor Thesis (2)
- Report (2)
- Other (1)
- Part of Periodical (1)
- Working Paper (1)
Is part of the Bibliography
- yes (190)
Keywords
- manufacturing (2)
- Architecture (1)
- AutoStore (1)
- BEST-FIT (1)
- Building and Construction (1)
- CNC machine tools (1)
- Civil and Structural Engineering (1)
- Cold Chain (1)
- Deep learning (1)
- Energy consumption (1)
Institute
- Fakultät Wirtschaftsingenieurwesen (190)
- Institut Digital Engineering (IDEE) (45)
- Institut für Sustainable Energy Systems (INSYS) (12)
- Fakultät für angewandte Natur- und Geisteswissenschaften (4)
- Fakultät Elektrotechnik (2)
- Fakultät Maschinenbau (2)
- Fakultät Angewandte Sozialwissenschaften (1)
- Hochschulleitung und -verwaltung (1)
- THWS Business School (1)
Time-Series Modelling for Energy Consumption Prediction in CNC Milling with Regenerative Drives
(2026)
Accurately predicting the energy demand of Computerized Numeric Control (CNC) machining processes before production enables the assessment of a product’s CO₂ footprint, the identification of optimization opportunities, and the implementation of energy-aware scheduling strategies. However, forecasting the energy consumption of CNC machines equipped with regenerative drives presents unique challenges, as the energy demand of a given G-command is influenced by the preceding operation. This study investigates the application of time-series Machine Learning (ML) models to better capture these temporal dependencies and improve energy consumption accuracy. A significant variance in repeated measurements was observed during the experimental phase, prompting a comparative analysis of using raw versus averaged energy values as input data. Multiple time-series model architectures, including Long Short-Term Memory (LSTM) and Temporal Convolutional Networks (TCNs), are evaluated for their ability to learn sequential patterns in a 5-axis machining process. The results reveal that while ensemble methods such as LightGBM and Random Forest achieve the highest accuracy and efficiency on the test dataset, sequence-based models demonstrate greater robustness on unseen validation data. Incorporating a small portion of validation data into training further improves ensemble performance, highlighting the trade-off between robustness and efficiency in energy demand prediction.
Research in this article refers to the influencing factors during Strategic IoT architecture decisions in small and medium-sized enterprises. Our results reveal the importance of each factor and its influence on other factors. A selection of influencing factors - mainly from literature - is examined for their significance and relations using Interpretive Structural Modeling. The data for the Interpretive Structural Modeling comes from an interview study with high-quality experts from the German-speaking area. Based on the results, a decision environment for practical decision-makers is built up from the influencing factors found. Subsequently, we provide a tactic to manage the decision-making process via a guide for practitioners. The outcome of our research can help SME practitioners control influencing factors on strategic decisions about the IoT architecture to improve the decision outcome. The authors are unaware of any literature source that provides an approach to handling the influencing factors of strategic IoT architecture decisions in small and medium-sized enterprises.
Purpose of this paper: The global rollout of COVID-19 vaccines has been one of the most significant public health measures in modern history. The logistics systems of the vaccination campaigns were key to both their effectiveness and efficiency. This paper examines the logistical framework conditions of the German COVID-19 vaccination campaign and analyses vaccine distribution in ultra-cold supply chains in order to contribute to a better understanding of the design of challenging logistics systems in the field of tension between public and private actors.
Design/methodology/approach: We use a case study approach to look at the COVID-19 vaccination campaign in Germany where the federal structure gives the 16 federal states (the units of analysis in the case study) a decision-making authority in health-related issues that was also reflected in the various logistics systems that were implemented for the campaign. Primary data was obtained through in-depth interviews and enriched with secondary data from newspapers and other sources. The cases were analysed through the theory lens of the collaborative governance framework.
Findings:
Federalism leads to braking effects where a rapid, coordinated and technically excellent response would be required. Although the vaccination campaign was an overall success, some logistical processes lacked GDP (Good Distribution Practice) approval, so the start was a bit bumpy. The opportunity to develop standard procedures for similar situations in the future was not properly utilised, reducing preparedness to a level below what was possible.
Value: This case study is one of the few empirical works that deal with the logistical systems and processes behind vaccination campaigns. In particular, it sheds light on the interplay between public and private actors in a federal government structure.
Research limitations/implications:
Primary data could only be obtained from four of the 16 federal states. This is mainly due to the fact that dealing with the COVID-19 pandemic in Germany is still an open issue and an unparalleled political minefield.
Practical implications:
The research shows that standardised, GDP-approved and configurable logistics processes together with a strategic inventory of equipment, infrastructure and consumables are key elements to be prepared for similar situations in the future.
Abstract
This data descriptor contains information about an extensive production data set for a five-axis CNC milling process. Three geometrically different products were manufactured and relevant features from the numerical control of the machine were recorded. The recorded manufacturing process contains the preparation of the machine for the next product (changeover) as well as the machining process (production). The experimental manufacturing was organized with the aid of a changeover matrix to ensure that all possible changeover combinations for the three products were considered. The production was repeated five times, resulting in 30 manufacturing sessions and five complete changeover matrices. The data set was recorded in a laboratory environment. A rich feature set including i.e. the NC-code of the products, tool information, and a Jupyter notebook is provided with the data set.
Computerized Numerical Control (CNC) plays an important role in highly autonomous manufacturing systems with multiple machine tools. The necessary Numerical Control (NC) programs to manufacture the parts are mostly written in standardized G-code. An a priori evaluation of the energy demand of CNC-based machine processes opens up the possibility of scheduling multiple jobs according to balanced energy consumption over a production period. Due to this, we present a combined Machine Learning (ML) and Job-Shop-Scheduling (JSS) approach to evaluate G-code for a CNC-milling process with respect to the energy demand of each G-command. The ML model training data are derived by the Latin hypercube sampling (LHS) method facing the main G-code operations G00, G01, and G02. The resulting energy demand for each job enhances a JSS algorithm to smooth the energy demand for multiple jobs, as peak power consumption needs to be avoided due to its expense.
The described data set contains features from the machine control of a five-axis milling machine. The features were recorded during thirteen series productions. Each series production includes a changeover process in which the machine was set up for the production of a different product. In addition to the timestamps and the twenty recorded features derived from Numerical Control (NC) variables, the data set also contains labels for the different production phases. For this purpose, up to 23 phases were assigned, which are based on a generalized milling process. The data set consists of thirteen .csv files, each representing a series production. The data set was recorded in a production company in the contract manufacturing sector for components with real series orders in ongoing industrial production.
Purpose of this paper:
In a production line with capacity constraint resources (CCR) or bottleneck stations, the
production control and the throughput rate are determined by the CCR station. In this
study, we consider a production system with parallel machines and there are many rounds
of reentrants during the production processes, and can be regarded as a reentrant flow
shop process (RFS). In most RFS systems, the release of jobs is a push type without
considering the WIP in each layer of reentrance. To resolve the excessive WIP in an RFS
system, we consider the theory of constraints (TOC) and implement the DBR (drum,
bottleneck and rope) to model RFS scheduling and to define appropriate buffer sizes at the
CCR stations of diverse reentrant layers.
Design/methodology/approach STS:
We obtain production data of a type of semiconductor chip where the photolithography
process is the most frequent reentrant. We designed nine scenarios based on the current
RFS of the case company by applying the TOC and considering the time differences in
parallel machines. A simulation model for the photolithography process was established
using AutoMod simulation. Thirty simulation runs were performed for the mean and
standard deviation. The goal is to identify the best scenario among the nine proposed ones.
Findings:
The simulation results demonstrated that Scenario 9, which involves dynamic buffers and
assigning higher priority to products with more processing reentrance layers, achieved the
highest throughput of finished products. The throughput of finished products in Scenario 9
is 22% higher than in the base scenario. It also shows improvements in maximum product
stack-up and input-output ratio compared to the base scenario.
Value:
The paper studies dispatching rules for reentrant flow lines having parallel machines with
data from the semiconductor industry, and this type of production system is common for
producing a wide range of electronic devices. With limited buffer sizes on the shop floor,
the number of WIPs is significant, especially for reentrant flow shops, to avoid being
clogged with too many WIPs. Implementing DBR modeling from TOC, the nine scenarios
demonstrate the effects of controlling buffer sizes on system performances.
Research limitations/implications:
The nine scenarios are designed based on the case company situations and may not be
readily implemented in other industries. Moreover, the DBR modeling of the buffer sizes
and CCR station schedule control should also be adjusted when employed in other
applications.
Purpose of this paper:
Robotic compact storage and retrieval systems (RCSRS) such as the AutoSto-reTM are very popular these days. A standardized yet flexible concept, plug-and-play control software and manageable investment and operating costs have made AutoStore systems a bestseller and inspired other companies to venture into developing their own RCSRS.
RCSRS in general and AutoStore in particular are highly encapsulated systems in which design, dimensioning and performance analysis during operation are based to a large extent on simulation studies. However, simulation-based per-formance analysis does not cover the actual workload of the system or the specifics of the logistics environment and does not take into account the sys-tem environment, such as the WMS, the middleware that connects the Au-toStore controller to the WMS, or the UnifyAnalytics data analytics tool now provided by AutoStore. In addition, simulation studies do not provide ware-house managers with guidance on tactical decisions such as whether dividing bin locations into compartments or operational decisions (e.g. opening or clos-ing picking ports) to improve overall system performance.
The purpose of this paper is to develop and test a systems theory approach to the performance of RCSRS, using an AutoStore as a case study.
Design/methodology/approach:
Following a systems theory approach, we define the AutoStore system as (a) the RCSRS with robots, storage bins, and picking ports (hardware), (b) the Au-toStore control software, middleware, WMS and UnifyAnalytics service (soft-ware), (c) the pickers and warehouse managers (people) and (d) the logistics and business environment. We then, in a case study, analyze an AutoStore installation in a German company in terms of its performance and correlate productivity losses with parameters of the system. We have purposefully se-lected the case company’s AutoStore for its potential to generalize findings.
Findings:
RCSRS such as the AutoStore are heavily dependent on a steep Pareto distri-bution for good performance. Our case study shows how dividing storage bins into compartments can jeopardize the performance of an RCSRS. If a compa-ny wants to use compartments to create more storage locations in the Au-toStore, it must change its unit of analysis from SKUs to bins in the simulation studies used in the design and dimensioning phase and deploy a middleware that is able to maintain an established ABC distribution of bins in the putaway process. Our case study also shows that warehouse managers have limited options to improve the performance of the system at runtime: they can only switch picking ports on or off. We propose a dashboard fed with live data from the AutoStore to help warehouse managers make this decision.
Value:
This paper is a contribution to the still relatively small literature on RCSRS. It goes beyond previous studies that mainly focused on algorithmic improvement of system details. In particular, it emphasizes the impact of tactical decisions made during the design of the system on the operational performance of the system. Furthermore, it demonstrates the limited operational levers available to warehouse managers and makes a strong case for adopting a systems the-ory approach to RCSRS design.
Research limitations/implications:
So far, our approach has only been tested on one installation of an AutoStore system. It should be tested more thoroughly when applied to other installations and other RCSRSs. Since RCSRS performance is highly dependent on the lo-gistical and business environment, future research should aim to first classify these environments and then perform detailed analysis within each class.
Practical implications:
Companies operating an AutoStore can use our approach as a guide for ana-lyzing the performance of the system. Adopting a systems theory perspective in the sales phase of an RCSRS can improve design decisions and help com-mission “better” systems, i.e., systems that better fit the target company's business and logistics environment and meet the performance suggested in pre-sales simulation studies. We also offer a solution for warehouse managers to understand, from a system performance perspective, when it is advisable to turn picking ports on or off and when not.
Global climate change is a cognitive challenge for many people and often evokes negative associations due to its complexity and interactions with politics, social movements and economic developments. Therefore, the possession of green skills becomes central to the fight against climate change. The European Council conclusions recognize this urgency and underline the need for a transition to green skills. This recognition also extends to higher education, where institutions have a crucial role to play in tackling the climate crisis. Personal Green Skills in Higher Education (PeGSinHE) is an Erasmus+ KA2 project coordinated by Kauno Kolegija (KK, Lithuania), Tampere University of Applied Sciences (TAMK, Finland), Hochschule für Agrar- und Umweltpädagogik (HAUP, Austria), Universidad de Málaga (UMA, Spain) and Technical University of Applied Sciences Würzburg Schweinfurt (THWS, Germany). The strategically designed project aims not only to promote green skills among students and encourage personal behavioral change in line with the Sustainable Development Goals, but also to instill a sense of social responsibility in the partner institutions. The focus is on empowering lecturers at partner universities through innovative teaching and learning methods to effectively impart green skills to students. This report describes the objectives and methodology used to assess environmental and sustainability competencies in the higher education institutions involved in the project. Methodologically, the report uses an assessment template designed to provide a comprehensive overview of best practice and baseline levels of environmental and sustainability competencies. It advocates the involvement of key stakeholders from all five partner Higher Education Institutions to ensure a broad perspective on these practices and competences within their respective countries and organizations. Different methods and perspectives will be used to collect data to enable a holistic understanding of the topic. The joint completion of the assessment template serves as a catalyst for joint discussions on the level of environmental and sustainability competencies and the identification of best practices in each organization. The results show that national implementation strategies are relatively loose, although some competency descriptions set targets for undergraduate degree programs. Challenges faced by higher education staff include resource constraints, particularly lack of time, the need for a deeper understanding of sustainable development and pedagogical tools, and the need for improved opportunities for collaboration. Given the time and resource constraints of this study, the results must be considered preliminary. Nevertheless, they confirm the findings of previous studies.
Computerized Numeric Control (CNC) plays an essential role in highly autonomous manufacturing systems for interlinked process chains for machine tools. NC-programs are mostly written in standardized G-code. Evaluating CNC-controlled manufacturing processes before their real application is advantageous due to resource efficiency. One dimension is the estimation of the energy demand of a part manufactured by an NC-program, e.g. to discover optimization potentials. In this context, this paper presents a Machine Learning (ML) approach to assess G-code for CNC-milling processes from the perspective of the energy demand of basic G-commands. We propose Latin Hypercube Sampling as an efficient method of Design of Experiments to train the ML model with minimum experimental effort to avoid costly setup and implementation time of the model training and deployment.
Introduction:
Entrepreneurs play a central role in economic and social stability, yet the start-up rate in Germany has declined in recent years, possibly due to the stress associated with entrepreneurial endeavors. Stressors such as financial uncertainty and time pressure are prevalent among entrepreneurs and negatively affect their psychological well-being. However, research on stress management strategies among self-employed individuals remains limited.
Methods:
This pilot study conducted a quantitative analysis with 117 self-employed participants in Germany. The study focused on typical entrepreneurial work demands and selected stress coping mechanisms.
Results:
The analysis revealed a significant correlation between quantitative demands and mental exhaustion. Furthermore, a high positive correlation between presenteeism and workload suggests that presenteeism may partially explain the variance in workload. These findings underscore how high job demands can lead to self-endangering behaviors that are detrimental to mental health.
Discussion:
Although no significant moderating effect of proactive coping on the relationship between job demands and mental exhaustion was observed, significant negative correlations between proactive coping and both job demands and mental exhaustion suggest a potential protective role of proactive coping against work-related stress. This study highlights the importance of understanding stress coping strategies among self-employed individuals and their impact on entrepreneurial success and mental well-being. Further research in this area is warranted to develop effective interventions to support the well-being and productivity of self-employed individuals in Germany.
In the context of environmental protection, the construction industry plays a key role with significant CO2 emissions from mineral-based construction materials. Recycling these materials is crucial, but the presence of hazardous substances, i.e., in older building materials, complicates this effort. To be able to legally introduce substances into a circular economy, reliable predictions within minimal possible time are necessary. This work introduces a machine learning approach for detecting trace quantities (≥0.06 wt%) of minerals, exemplified by siderite in calcium carbonate mixtures. The model, trained on 1680 X-ray powder diffraction datasets, provides dependable and fast predictions, eliminating the need for specialized expertise. While limitations exist in transferability to other mineral traces, the approach offers automation without expertise and a potential for real-world applications with minimal prediction time.
