Benedikt Konrad

Challenges for Data Mining on Sensor Data of Interlinked Processes

In industries like steel production, interlinked production processes leave no time for assessing the physical quality of intermediate products. Failures during the process can lead to high internal costs when already defective products are passed through the entire value chain. However, process data like machine parameters and sensor data which are directly linked to quality can be recorded. Based on a rolling mill case study, the paper discusses how decentralized data mining and intelligent machine-to-machine communication could be used to predict the physical quality of intermediate products online and in real-time for detecting quality issues as early as possible. The recording of huge data masses and the distributed but sequential nature of the problem lead to challenging research questions for the next generation of

Sustainable Interlinked Manufacturing Processes through Real-Time Quality Prediction

Based on a rolling mill case study, this paper discusses how data mining techniques and intelligent machine-to-machine telematics could be used to predict internal quality issues of intermediate products in manufacturing processes. The huge amount of data recorded during processing and the distributed but sequential nature of the manufacturing lead to challenging questions for data mining applications and advanced process control approaches in industries like steel production. Moreover, the discovery for hidden information, knowledge and dependencies in the process data contribute significantly to support avoiding waste of resources and achieving the objectives of zero-defect-production, sustainable and energy-efficient manufacturing processes.

Die Bedeutung von Data-Mining im Kontext von Industrie 4.0

Kurzfassung Moderne Informations- und Kommunikationstechnologien ermöglichen die umfassende Speicherung großer Datenmengen, ihre Auswertung wird jedoch oftmals nicht hinreichend fokussiert. Insbesondere die effektive Nutzung des in den Informationsspeichern vorhandenen Wissens zur prädiktiven Entscheidungs- und Planungsunterstützung ist von höchster Bedeutung. Dieser Beitrag beschreibt am Beispiel drei unterschiedlicher, anwendungsspezifischer Ansätze der Wissensentdeckung die zunehmende Relevanz des Data-Mining im Produktlebenszyklus.

Striving for Zero Defect Production: Intelligent Manufacturing Control Through Data Mining in Continuous Rolling Mill Processes

Steel production processes are renowned for being energy and material demanding. Moreover, due to organizational and technological restrictions in flow production processes, the intermediate product’s internal quality features cannot be assessed within the process chain. This lack of knowledge causes waste of energy and material resources, unnecessary machine wear as well as reworking and rejection costs, when defective products are passed through the entire process chain without being labeled defective. The process control approach presented in this paper provides the opportunity of gaining transparency on quality properties of intermediate products. This aim is achieved by predicting intermediate product’s quality by means of data mining techniques. This approach can be applied in a wide field of production environments, ranging from steel and rolling mills to automated assembly operations. Concerning this concept, the authors derive a methodology for representing different quality properties in a way that it can be applied in the process control. Beyond that, first results of statistical analyses on the quality-related significance of process parameters are disclosed.

Pushing the Limits of Lean Thinking–Design and Management of Complex Production Systems

This paper presents an approach for handling variability in production systems. A classification for variability which includes a differentiation in non-value adding and value adding variability is proposed and indicators for their quantification are derived. It is shown that classical Lean methods fail in make-to-order production because of the misconceived influence of value adding variability. Consequently, different approaches from Data Mining and Factory Physics are applied.

Konzepte zur Effizienzsteigerung von Variantenfließlinien bei hohen Variantenanzahlen

Kurzfassung Die Nachfrage nach individuellen Produkten erfordert von Unternehmen das Angebot eines breiten Produktspektrums. Zwar sind Variantenfließlinien gängige Praxis, jedoch stellt der Umgang mit weiter zunehmenden Taktzeitspreizungen und volatilen Kundenabrufen eine große Herausforderung dar. In dem Beitrag werden zwei Konzepte vorgestellt, die aufbauend auf Methoden des Data Mining sowie des flexiblen Mitarbeitereinsatzes die Linieneffizienz von Variantenfließlinien erhöhen.

Renaissance of Group Technology: Reducing Variability to Match Lean Production Prerequisites

Abstract Although Group Technology was invented to transfer benefits of economies of scale to job-shop production during the 1960s, its underlying methodologies are still relevant for modern production, which follows the lean production paradigm. This paper discusses how Group Technology has developed towards an essential tool in industrial engineering in the last decades and how it is applied in the context of Lean Production. Moreover, the authors present two examples of current research in the field of Group Technology and Lean Production: Levelling of low volume, high mix production and mixed-model assembly line balancing.

Renaissance der Gruppentechnologie

Kurzfassung Obwohl die Gruppentechnologie mit der Motivation entwickelt wurde, in der variantenreichen Kleinserie positive Skaleneffekte zu erzielen, ist sie auch in Zeiten der „Lean Production” von hoher Relevanz. Die Autoren zeigen in diesem Beitrag anhand von Praxisbeispielen und aktuellen Forschungsvorhaben, wie Verfahren der Gruppentechnologie eingesetzt werden, um die Ziele schlanker Produktionssysteme, wie die Reduktion von Verschwendung oder Komplexität, durch Variantenbeherrschung zu erreichen.

Verkehrsforschung in der Produktionsflussanalyse

Kurzfassung Methoden der statistischen Physik werden seit einigen Jahren erfolgreich auf Probleme der Verkehrsforschung angewandt. Da sich starke Analogien zwischen den Produktions- und Verkehrssystemen feststellen lassen, liegt eine Übertragung der Untersuchungsansätze in das jeweils andere Themengebiet nahe. Dieser Artikel stellt einige wichtige Verfahren beider Bereiche vor und zeigt auf, welche Vorteile ihre Kombination in der Produktionsflussanalyse generieren würde.