Methode zur Erschließung von Wissen aus Datenmustern in Supply-Chain-Datenbanken (Volume 1) (English shop)

Preview

Table of Contents, PDF (55 KB)
Extract, PDF (210 KB)

A fundamental step towards making the supply chain manageable is the identification of causal relationships that are reflected in logistics transactions. Due to the unmanageable volume of data, the discovery of complex causal relationships cannot be carried out manually. The dissertation presents a method for the discovery of knowledge, such as causal relationships, and discusses the consideration of contextual knowledge in the individual phases of the process model. One focus of the developed method is the integration of model-accompanying verification and validation into a process model of knowledge discovery. Through a novel use of simulation, the work also extends the existing verification options of Knowledge Discovery in Databases. In order to enable the model to be applied even when the data situation is inadequate, concepts of data farming are finally introduced as a methodological element. The practical applicability of the method developed in this work is demonstrated using transaction data from a manufacturer of small electronic devices as well as a data farming model.