Published March 2015.
Smart data analyzes support the scheduling of component production at Herrenknecht AG. Within a customer order, it is necessary to produce various components. The core components are generated in individual production orders at the corporate headquarters in Schwanau. Component manufacturing includes cost, planning, production and quality data.
For the smart data analysis of the Smart Data Solution Center Baden-Württemberg (SDSC-BW) these data were collected over a period of 6 months. Data from a wide range of systems, such as the Manufacturing Execution System and Enterprise Resource Planning, were included, recording approximately 30,000 production data. The task was to improve existing approaches based on the data situation and to find new approaches. Questions were as follows: “How do plan deviations affect delivery times?” and “Are there further yet unknown correlations between the factors to be examined?”
The data analysis of the SDSC-BW first created a descriptive statistics of the provided data sets. Aspects of data quality, such as the number of missing entries, were considered. After the merge and cleanup of the data sets, production-delaying factors, as well as the examination of possible connections between conspicuous costs and production sequences, were identified using smart data tools from the field of machine learning. Methods of classification and outlier detection were used. The results show that smart data algorithms can help the planning experts at Herrenknecht to further increase the supply accuracy and to reduce buffering times.
For further information, please visit: www.sdsc-bw.de
Data Innovation Community
Herrenknecht AG, Smart Data Solution Center Baden-Württemberg
Smart Data Solution Center Baden-Württemberg
March 2015 – December 2015