Vorwerk International & Co. KmG is sponsoring two new scholarships in the Graduate School of Logistics, opening another scholarship class. The two scholarships will start this fall and follow on from an expiring scholarship. This will allow them to build on current results.

Vorwerk’s first fellowship in GSofLog, which expires in September 2021, focused on sales forecasting using machine learning in multi-channel sales. A process model and a first prototype for ML-based forecasts were created. With the two new fellowships, the company now wants to build on these results and add two further perspectives. For this reason, the fellowships have been filled with graduates from the fields of statistics and logistics.

Using the example of small electrical appliances in multi-channel sales, a procedure for determining sales forecasts is to be investigated. On the one hand, company data is to be used as the basis for meaningful forecasts using state-of-the-art analysis methods from machine learning, statistics and time series analysis and incorporating environmental data. On the other hand, a simulation model will help to elucidate distribution behavior in the absence of data quality, highlight causalities and volatility. It will be explored how, with the help of data farming, large stocks of synthetic data obtained from this model can be effectively used to generate sales forecasts and how these contribute to gaining knowledge about the complex behavior of the distribution system.

The developed solutions will be prototypically tested at Vorwerk International in Düsseldorf and in Freienbach (CH). The PhD supervisors are from the Faculty of Statistics, Prof. Markus Pauly, and the Faculty of Mechanical Engineering, Prof. Markus Rabe, at TU Dortmund University.