The executive agency of the Ministry of Infrastructure and Water, called Rijkswaterstaat, has the mission to develop and maintain the road network, the water network and the water systems in the Netherlands. The aim is to create procedures that facilitates the safety of the people operating and using these systems and to reduce the risk of an error being made. With the use of various methods, the risk is aimed to be as low as possible. However, to reduce the risk in these systems, it requires a combination of man and machine. Within Rijkswaterstaat, there was a method created to establish the reliability of the man in storm surge barriers, called OPSCHEP model. This method was based on a Human Reliability Analysis (HRA) method called THERP. The OPSCHEP model was only appropriate to use on storm surge barriers within the Netherlands. When this model was applied on other objects, this lead to conservative results. Therefore, Rijkswaterstaat is looking for another HRA method that would be appropriate to use on other objects. In this research, the HRA methods are applied on the Volkerak sluice complex in the Netherlands. As this is one of the largest sluice complex in Europe and therefore, important for the economics in the Netherlands. This leads to the following problem formulation:

Which HRA method is most suitable to be applied for a task performed by the operator on the Volkerak complex?

Three HRA methods were investigated: THERP, SPAR-H, and CREAM. Before the methods could be used, a task analysis for the operator at the Volkerak sluice needed to be identified. In this task analysis, the steps and actions were identified that the operators execute when ships want to pass the sluice from the south side to the north side. With the use of the three methods, each action that the operator executes there is a certain probability that could lead to an error. An error is defined as a significant delay in the process of the ships going from the south side to the north side of the sluice. In each of these actions, factors could influence the performance of the operator, which are included in the three HRA methods. The probabilities of these methods are compared to the results from experts. Based on these results and the review of the three methods, a suitable HRA method was identified to be used for the operator of the Volkerak complex.

The results identified that the CREAM method was the most suitable for Rijkswaterstaat to use on the Volkerak sluice complex. As the method was easy to use and the results showed more than just the probability of a human error. Also, the comparison of the results showed that CREAM is a good method to use.