Subproject T3 Capacity planning within the regeneration of transformers (finished 2018)

Within the transfer project, the engine-specific findings of the SFB 871 regarding the planning and control of regeneration processes with fuzzy load information are generalized and transferred to the regeneration of transformers. A comprehensive damage library and data mining methods will be used to predict a realistic load level of the involved work systems, caused by future regeneration orders. On this basis, capacity planning is carried out and load controlling actions will be conducted. A challenge within the project will be the integration of the regeneration process in the existing production system. In an additional step, the predicted load level data will be used for an offer calculation methodology that will be developed within the project.

Motivation

© Siemens AG
Rail vehicle transformer © Siemens AG

The potential use of complex capital goods, such as rail vehicle transformers, can be increased by carrying out regeneration measures. In this regard, when commissioning a regeneration service provider with a view to recommissioning the regenerated goods, short turnaround times in the regeneration process and high adherence to deadlines are particularly relevant. The basis for the implementation of logistically efficient regeneration processes is, in particular, reliable planning and control of order processing.  However, the condition of the reconditioned goods in terms of the type and scope of reconditioning requirements is usually not fully known at the time of planning, which poses a major challenge, particularly for capacity planning. This applies in particular to the application partner in the transfer project, as the reconditioning orders have to be integrated into ongoing production there and thus also influence the achievement of their logistical goals. High planning quality and process reliability are therefore required to achieve high logistical performance. In the transfer project, this was to be achieved, among other things, by transferring the findings from subproject D1 on forecasting regeneration costs for aircraft turbines to the regeneration process for rail vehicle transformers.

Results

Based on an analysis of the processes for planning and implementing regeneration measures for rail vehicle transformers, the transfer project first reviewed and adapted the generic process model for regeneration. Furthermore, in cooperation with the application partner, a database for damage cases and associated regeneration activities was created, on the basis of which capacity requirement forecasts for the regeneration of rail vehicle transformers were generated. To increase the quality of the forecasts, the application partner also further developed the technical diagnosis without prior disassembly. To support the logistically efficient, joint processing of production and regeneration orders, a comprehensive production control system was also designed and introduced. The implementation of prototype software tools for the automated execution of various data analyses and associated application-specific visualizations also ensures the long-term usability of the results by the application partner.  

Entity block diagram of the physical data model and information fuzziness curve in the regeneration process

Current work and outlook

The findings will be taken into account in the ongoing subproject D1 for the further evaluation of approaches and methods from the field of data mining and data analytics for production planning and control as well as supply chain configuration.

The project is managed by the Institute of Production Systems and Logistics .

Publications

  • Melissa Seitz, Maren Sobotta und Peter Nyhuis (2018): Einsatz von Data Mining im Regenerationsprozess von Schienenfahrzeug-TransformatorenZWF 113 (12), S. 814–818
    DOI: 10.3139/104.112025
All publiactions of the Collaborative Research Centre