Decision Making under Uncertainty for Construction Management of Offshore Wind Assets

Abstract

Offshore wind is expected to be one of the important contributors to the energy transition towards a more renewable and sustainable energy future. This can be clearly seen from the amount of investments over the past years as well as from the substantial upcoming offshore wind projects in the years to come. Many technological implementation challenges have already been addressed, but the number of new challenges will continue to increase. Especially, as the industry continues moving further offshore with larger wind turbines and as the existing offshore wind farms will approach the end of their service lives. Therefore, the need for improved asset management modelling over the entire service life from design towards decommissioning will continue increasing to support better data driven decision making under uncertainty.

For this and in particular for the construction management of offshore wind assets, in this thesis new models and methods have been developed to support this enhanced decision making. These decisions are subject to various types of risks and uncertainties, varying from environmental uncertainties, supply chain disruptions and stochasticity of construction activities’ duration. Therefore, these should be properly taken into account in construction management models using performance and/or expert data from past construction projects.

In this thesis two types of data availability have been distinguished: (i) where sufficient relevant performance data is available and (ii) where relevant past performance data is rather limited. In the first case, statistical methods are used, such as Copula functions to model the dependence between metocean variables and Bayesian Networks to model the dependence between subsequent construction activities. In the second case, expert knowledge and data are used to quantify the uncertainty using a mathematical aggregation method for expert judgments (i.e. Cooke’s classical modelling). The different methods have been applied to several test cases to investigate the associated cost and time impact. As a result of this research, different tools and an open-source software
were developed. These also can be used in different fields of application using this proper mathematical expert judgment aggregation modelling.

Finally, it can be concluded that the state-of the art developments within this thesis substantially contribute to decision making under uncertainty, so that construction management strategies are optimized and thereby the offshore wind energy assets life cycle value is maximized.