OffshorePlan – Complementary application of mathematical and discrete-event models to solve complex planning and control problems in offshore construction logistics

• Applicants: Dr.-Ing. Michael Lütjen; Professorin Dr.-Ing. Helena Szczerbicka
• Subject Area: Production Systems, Operations Management, Quality Management and Factory Planning
• Term: since 2018
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 411010215

Offshore construction logistics pose an exceptionally challenging problem in terms of planning and control. To finish construction in time, cost intensive personnel and ships have to be deployed efficiently, despite ever changing, uncertain weather conditions. As offshore wind energy is still considered a new technology, there exist no established methods for the planning and the control for the construction of offshore wind parks. At this, one can differentiate two approaches: event-discrete simulations as well as mathematical or stochastic optimizations. By themselves, both methods provide their own advantages and disadvantages in terms of computational time, level of detail und optimality. As a result, this project aims to investigate new ways for the complementary utilization of both types of methods in the context of offshore construction logistics. Under the basic assumption that despite formal differences, both types of models describe the same elements of the real-world system, the first phase of this project established model transformations, to convert in between or to generate each kind of model with its own level of aggregation/abstraction based on a more basic description or model of the real-world system. On the one hand, this allows project planners to reduce the efforts in developing and maintaining the simulation and optimization models. On the other hand, they can leverage the advantage of both types of models. The second phase aims to increase complexity the regarded problem by including offshore specific characteristics and restrictions of personnel planning as an integrated planning problem. Moreover, the second phase aims to develop a hierarchical framework that can automatically use the advantages of both model formulations. Therefore, at each decision point, the framework will apply newly developed methods to decide if a certain decision should be made using a heuristic, simulation-based optimization or if an optimal, mathematical formulation would be best suited. This new approach intensifies the complementary use of the noted approaches, which have usually been regarded separately in contemporary literature.

• DFG Programme: Research Grants