Join Expo Room
During the conference you can access the expo room and meet project members for discussions, Q&A and more. Join using the following link:
- 12:30 - 13:00 (CEST): Digital Reality in Zero Defect Manufacturing
- 13:00 - 13:30 (CEST): Dynamic Value Stream Analysis - Using real time analytics to boost factory efficiency
- 13:30 - 14:00 (CEST): VariFlow - Variability Management in directed Material Flow Systems using Machine Learning Methods
Digital Reality in Zero Defect Manufacturing
The EU project “QU4LITY” focuses on the realization of application-oriented and data-driven Zero Defect Manufacturing (ZDM). The research focus of the IPS is the development of interoperable digital infrastructures for the ZDM by researching industrial applications of data analysis. An essential aspect is the reduction of error costs of automatic process control by reliably anticipating the resulting product quality during the production process. For the realistic representation of this objective, the integration of Big Data Analytics and AI (Artificial Intelligence) methods is being researched at Siemens AG’s Amberg electronics plant (EWA). A special focus is placed on the high variety of products variants and the small number of analysable defect components, which are common properties of the production of electronic components.
Dynamic Value Stream Analysis - Using real time analytics to boost factory efficiency
The TWINCULATORS are a software implementation of the scientific method of dynamic value stream analysis (DWSA). This is a cloud-based software solution that processes large amounts of data in real time and adapts to the user’s production system in a scalable manner. Four modularly designed calculators consisting of analysis functions enable users to get an overview of current states of processes, products and production systems in real time. Thus, production planning and control is based on the current real situation of the factory.
VariFlow - Variability Management in directed Material Flow Systems using Machine Learning Methods
Variability Management in directed Material Flow Systems using ML-Methods
Competitive factors through globalization, more individualized customer requirements as well as fluctuating customer demand lead to diverse variability influences in the daily production routines of companies. These variability influences directly affects the order processing process. In order to cope with these influences, the VariFlow research initiative focuses on the use of innovative machine learning methods in the field of production planning and control. Sales and requirements planning (PrABFlow), production program planning (PrOPFlow), order release (PrAFFlow) and bottleneck identification (PrEPFlow) as subfields of production planning and control are dealt with in separate research projects.