Kurzfassung: Mehr und mehr Unternehmen setzen auf Shopfloor Management, also die Mitarbeiterführung direkt am Ort der Wertschöpfung, um die Leistung ihrer Produktion zu verbessern und Beschäftigte in Verbesserungsaktivitäten einzubinden. Dabei gibt es aktuell unterschiedliche Auffassungen darüber, wie Shopfloor Management-Systeme aufgebaut sind und Prozesse ablaufen. In diesem Beitrag wird das Darmstädter Shopfloor Management-Modell vorgestellt, welches die Ziele, Voraussetzungen und Handlungsfelder von Shopfloor Management aufzeigt und diese systematisiert.

The Value Stream Method (VSM) is widely used in manufacturing to analyze and redesign value streams. The aim is to improve processes, reduce waste and create a thorough product flow. Despite having many benefits, VSM also comes with disadvantages regarding modern dynamic production environments. It fails to meet the requirement of providing reliable information for a realistic Value Stream Design (VSD) followed by targeted improvement activities. As a result, the VSM is usually subject to uncertainty and relies on expert knowledge. Digitalization, on the other hand, is leading to an increasing availability of production data. The use of data has the potential to support the VSM with targeted data preparation. In this regard, the concept of Digital Twin (DT) offers the capability of providing the required database to systematically collect and condense this data. This paper provides a framework for the Digital Value Stream Twin (DVST). In addition, requirements for the implementation of a DVST in practice will be elaborated.

The value stream method, a key tool in industry to analyze and visualize value streams in production, aims to holistically optimize process steps, reduce waste, and achieve continuous material flow. However, this method primarily relies on data from a single on-site inspection, which is subjective and represents just a snapshot of the process. This limitation can lead to uncertainty and potentially incorrect decisions, especially in industries producing customer-specific products. The increasing digitization in production offers a solution to this limitation by supporting the method through data provision. The concept of the digital shadow emerges as a key tool that systematically captures, processes, and integrates necessary data into a model to enhance traditional value stream mapping. This addresses the method’s shortcomings, especially in heterogeneous IT landscapes and complex value streams. To effectively implement the digital shadow this study identifies concepts of digital shadows and their key components and evaluates them for their relevance in industrial environments using an expert study. Based on the results, a design model is defined. This model entails guidelines to support companies with the practical implementation of the digital shadow of a value stream. Lastly, the model is evaluated on a realistic value stream in a learning factory.

Self-supervised learning has demonstrated state-of-the-art performance on various anomaly detection tasks. Learning effective representations by solving a supervised pretext task with pseudo-labels generated from unlabeled data provides a promising concept for industrial downstream tasks such as process monitoring. In this paper, we present SSMSPC a novel approach for multivariate statistical in-process control (MSPC) based on self-supervised learning. Our motivation for SSMSPC is to leverage the potential of unsupervised representation learning by incorporating self-supervised learning into the general statistical process control (SPC) framework to develop a holistic approach for the detection and localization of anomalous process behavior in discrete manufacturing processes. We propose a pretext task called Location + Transformation prediction, where the objective is to classify both, the type and the location of a randomly applied augmentation on a given time series input. In the downstream task, we follow the one-class classification setting and apply the Hotelling’s $$T^2$$T 2 statistic on the learned representations. We further propose an extension to the control chart view that combines metadata with the learned representations to visualize the anomalous time steps in the process data which supports a machine operator in the root cause analysis. We evaluate the effectiveness of SSMSPC with two real-world CNC-milling datasets and show that it outperforms state-of-the-art anomaly detection approaches, achieving $$100\%$$100 % and $$99.6\%$$99.6 % AUROC, respectively. Lastly, we deploy SSMSPC at a CNC-milling machine to demonstrate its practical applicability when used as a process monitoring tool in a running process.

Due to the ongoing climate crisis and the resulting targets of several countries, as well as the EU, manufacturing companies face the need to address and reduce their greenhouse gas (GHG) emissions. To calculate these emissions, the product carbon footprint (PCF) can be a helpful tool, both to come up with reduction measures internally as well as to communicate it externally as competitive advantage. However, the PCF is subject to uncertainty factors that hinder its use and need to be systematically assessed. For this reason, this paper first collects and clusters relevant uncertainty factors resulting from differences between accounting standards, but also from imprecisions within the standards. Subsequently, the PCF is determined in different scenarios of uncertainty factors, applied to an industrial case study. This is used to be able to put the deviation in the final results into perspective. Recommendations for action are finally derived from the analysis of the literature and the results of the use case.