Former Research Projects
-
Analysis of Stochastic Flow Production Systems with Artificial Neural Networks
Former Person in Charge: Insa Südbeck
Flow production systems are often used to organize the production of large quantities of homogeneous goods. In flow production systems, different tasks are performed on the workpieces at serially arranged stations. In case of stochastic influences on the processing times at the stations, blocking or starving of parts of the system may be a common occurrence. Thus, to sustain the material flow, expensive buffers are set up between the stations. To find an efficient structure of the flow line, which minimises the costs while generating maximum throughput, fast evaluation procedures are needed. In this approach, we use artificial neural networks to predict the throughput of the overall system.
-
Flexible Projects for the Customer-Specific Remanufacturing of Complex High-Capital Goods
Former Person in Charge: Dr. Carolin Kellenbrink
For the regeneration of complex capital goods such as aircraft engines, there are different remanufacturing modes available. They can be distinguished by the use of resources, the throughput time and the costs. The choice of a regeneration mode determines the functional properties of the regenerated capital good, such as durability or energy consumption. The requirements on the regeneration processes and results - and thus also on the execution modes to be selected - vary depending on the specific good and customer. For the planning of such processes, the regeneration process can be interpreted as a flexible resource-constrained project, in which the project structure and thus the execution modes are determined model-endogenously.
-
Planning Maintenance Processes of Wind Turbines
Former Person in Charge: Dr. Carolin Kellenbrink
Wind turbines are subject to age- and weather-related wear and tear during their service life. Therefore, they must be inspected at fixed intervals and, depending on the condition, be repaired by qualified personnel. The maintenance and repair measures can be implemented in various ways. These so-called regeneration modes differ mainly with regard to organizational aspects. For example, only one team can be deployed, or two teams can work in parallel at one location. The implementation of the regeneration process also depends on the equipment that is available on site. There is a trade-off between lower costs and longer processing times. Next to the spatial dimension, another significant aspect of regeneration planning for wind turbines is the uncertainty of the regeneration processes, which results in particular from the unknown work contents and the weather-dependent accessibility.