Detailed assessments of asphalt and block revetments in WTI2017 : Feasability and alternatives

This report discusses the options for detailed probabilistic and semi-probabilistic assessments of asphalt and block revetments in the WT12017. Models for detailed probabilistic assessments would have to be available by 2016 for being included in the WT12017. Robust probabilistic load and resistance models cannot be developed, implemented and tested before this deadline. The WTI2017 thus cannot contain models for detailed probabilistic assessments of asphalt and block revetments. Probabilistic assessments could still be carried out within the context of advanced assessments ("toets op maat") with prototype resistance models and tailor-made, site-specific load models. Rather than developing load models on a case-by-case basis, a pragmatic work-around based on the Q-variant could be implemented in Hydra-Ring. Whether this would be efficient depends on the load model in a future WTI: if this model is to be a truly time dependent model, it would be inefficient to spend resources on designing and implementing a temporary fix in Hydra-Ring. Semi-probabilistic assessments will make use of the Q-variant for obtaining design loads, Waveimpact (asphalt revetments), Steentoets (block revetments) and a set of safety factors. These safety factors could be derived from ISO-standardized influence coefficients or from a large number of probabilistic calculations. The latter is strongly advised, given the conservatism that would result from the use of standardized coefficients. A calibration exercise for asphalt revetments can be carried out with the probabilistic Waveimpact prototype and wide-ranging, assumed loading conditions. The probabilistic Steentoets prototype is not yet sufficiently robust and efficient for use in a calibration exercise. It is therefore advised to use response surfaces for now and to cross-check the results against a number of Steentoets calculations. It may become possible to calibrate safety factors with Steentoets after further development of the probabilistic prototype. But waiting for an improved prototype would be ill-advised given the WTI2017-timetable.