Data Mining

The advancing digitization leads to new possibilities for the design and digital support of business processes. In particular, non-R&D-intensive, mostly small and medium-sized enterprises, face great challenges in realizing these potentials. In the context of this article, various application scenarios are outlined. A detailed example of a non-R&D-intensive company shows how the procurement can be supported by the analysis and forecasting of relevant data, e.g. process data or the availability and costs of components, as well as the creation of the offer.

The importance of data science for production and logistics continues to grow because more data are available due to Industry 4.0-Applications used for process and system optimization. In addition, the improved methods and tools for data analysis enable an easier processing of application-specific issues. This article is the second part relating to data science in production and logistics. While the first article dealt with the definition of terms and the potential of data analysis, the article at hand is dedicated to the application of data science in production and logistics by means of various application examples.

Efficient operation and maintenance (O&M) processes are important success factors to reduce the operating costs in the operation phase of industrial assets. Therefore, companies use different maintenance strategies that are well known from literature. Due to uncertainties in occurrence of faults and the duration of process fulfilment, strategies even supported by new technologies do often not fit to the specific real-world challenges. The focus of this paper is to discuss a new concept where operation data analysis is used to support decisions for logistics maintenance processes.

Customers in online shops return at least half of the placed orders. This huge amount of return shipments results in high costs e.g. from a logistic point of view. To predict the return rate based on customer data and order information, machine learning techniques can be applied which are able to learn a powerful model for return prediction based on historical order data. This article introduces a hands-on approach for successfully applying machine learning in real world processes and shows a case study to predict the return shipment probability in an e-commerce scenario.

Today’s standard information technologies in production logistics allow the storage of large data sets that have a huge variety of different parameters. The also increased complexity of production processes leads to complex dependencies between different decisions variables. Therefore, this paper deals with the application of computer-based methods of analysis for the efficient evaluation of multi-criteria dependencies.

In the planning phase of cyclically interlinked production lines neural networks are used to learn and forecast the characteristics of these systems. To increase the results of learning and to accelerate the training this paper presents a method, based on genetic algorithms, that reduces the attributes to describe the behaviour of the production lines.