Description:
To manage perfectly an efficient and effective supply chain of continuous and undisturbed flow of goods is needed. To achieve this identification, location and sensor technologies must be implemented to generate state data of the logistics objects. However, the amount of information overstrains the operational logistics planner and the information systems have to face enormous data streams. Data mining methods are useful to cope with such big data streams, and they are well developed in the literature. But these methods are not often applied to logistical state data. Without knowledge of the processes, the results of the algorithms cannot be understood. Therefore, the objective of this work is to introduce a general concept to model and to analyse logistical state data, in order to find irregularities and their causes and dependences. This work shows that it is possible to use data mining methods on logistical state data to filter irregularities and their causes.