Bayesian updating of the safety level of concrete elements using information from test results and conformity assessment

Bayesian statistics can be used in order to update the strength distribution of concrete. In practice however, direct strength results from concrete compressive testing are most often not provided by the producer and only indirect information from conformity control can be obtained and used in order to update prior information on concrete strength distributions. In literature, prior information is available on the parameters of concrete strength distributions. These distributions can be updated by using direct information from strength results or by using indirect information which is available through conformity assessment. Due to the filtering effect of conformity control, the average of the strength in the accepted concrete lots increases while the standard deviation in comparison to lots which were offered for inspection. As a consequence, conformity control of concrete has a positive influence on the safety level of concrete structures, which increases in general. It is explained how these effects can be quantified, using a numerical algorithm based on numerical integration and Monte Carlo simulation techniques. The filtering effect of the current EN 206-1 conformity criteria are calculated and analyzed, taking also into account the influence of autocorrelation between consecutive test results