An Optimization Model for Short-Term Routing and Scheduling of Offshore Wind Maintenance

Abstract

Maintenance costs constitute a significant portion of the total costs for offshore wind investments. Consequently, a substantial amount of research aims to mitigate these costs. Studies targeting short-term decision-making primarily concentrate on finding the most cost-effective routes for the maintenance vessels while scheduling as many maintenance tasks as possible. This thesis suggests an alternative approach where all maintenance tasks are considered optional. Instead of minimizing costs, the optimization model we propose maximizes expected profit. The motivation is to establish a more dynamic relationship between short-term decision-making and long-term strategy. We formulate the problem of selecting routes for maintenance vessels as an integer linear program. Further, we use a mixed integer linear programming sub-problem to generate routes via a column generation algorithm. We have developed several instances for testing the model, which we make available for subsequent research. Our proposed model provides optimal solutions for some of the problem instances where the sub-problem can be solved with exact methods. We also present a meta-heuristic for the sub-problem, capable of finding good solutions to problem instances considering up to 60 maintenance tasks. Lastly, we find that the column generation method outperforms a more straightforward solution algorithm.