Optimization of buffer storage size for radioactive waste in the decommissioning of a nuclear power plant

Abstract

During the operation of a nuclear power plant many of its components, systems and structures become radioactive. After the final shutdown of a nuclear power plant is will be decommissioned, i.e. the radioactive substances will be removed from the plant. The decommissioning is a very labour-intensive project, and e.g. the cost estimate for the decommissioning of the Loviisa nuclear power plant amounts 360 M€, comprising a significant part of its nuclear waste management costs. The decommissioning comprises several consecutive work phases, such as dismantling, cutting, characterization, packaging, transport and final disposal of the components, systems and structures of the power plant. A smooth completion of the decommissioning project requires these operations to be arranged taking into account the variability in the work performance in an appropriate way. Besides the management of the work performance, availability of buffer storage space between the work phases can reduce the effect of delays in one work phase on the rest of the work phases.

The thesis considers some waste types according to the decommissioning plan of the Loviisa nuclear power plant, and determines optimal buffer storage size considering the cost of storage and the costs resulting from a deficiency of buffer storage. The variation in the work performance is determined based on literature from comparable applications, and an optimization model is constructed using queueing theory to describe the material flow. The optimization model results in an optimal buffer storage size and the related costs. Several sensitivity analysis cases with regard to model parameters are included in the analysis, and they indicate the material flow variation being one of the dominant factors to determine the optimal buffer storage size. The sensitivity of the result to storage and delay cost parameters is lower. The recommended buffer storage space represents 15 % of the total waste volume focused on in the thesis.

One of the key findings is that in such a labour-intensive project as the decommissioning, underdimensioning of the buffer storage may result in significantly higher costs than overdimensioning. Therefore a slight overdimensioning is recommended due to uncertainties in the work variability data. Furthermore, the analysis carried out in the thesis reveals the need of a management system which is able to respond fast to anomalies in the work performance as well as an enterprise resource planning system which meets the requirements for workforce allocation and material flow bookkeeping, including the special requirements relating to radioactive materials. Based on the analysis, the thesis gives recommendations for further decommissioning planning as well as for the execution of the decommissioning project.