PIANC Working Group 140 investigated the state-of-the-art of the semi-probabilistic design concepts for hydraulic structures in Europe, USA and China. By a questionnaire typical elements of the semi-probabilistic design procedures were investigated for ultimate and serviceability limit states. These are, e.g., the determination of characteristic load or resistance values, typically used actions for hydraulic structures, partial safety factors for actions and resistances, model factors, combination factors, configurations of limit state functions, existence and number of consequence classes, target reliability indices and robustness criteria.
rnBy literature the different methods were worked out and described. Differences by methods and figures are shown and analysed. Only a few countries have special additional regulations for hydraulic structures, although the basic codes are preferentially developed for buildings. Some of the investigated issues identifying differences are:

    rnrn

  • in loads (hydraulic, seismic, geotechnical, wind, uplift, etc.) and their distributionsrn
  • in load combinationsrn
  • in strengths (steel, concrete, etc.) and their distributionsrn
  • in limit states for navigation and flood control structuresrn

rnAnother concern was the investigation of statistical distributions for relevant actions and resistances. With these findings the understanding of the partial safety concept, as well as probabilistic analyses are enabled. Especially the action of water, a key action for hydraulic structures, has been dealt with more intensely, with some examples.

rnTo identify the differences, the WG reported practical examples for several navigation and control structures. To facilitate acceptance by the design community and to aid in education, the report included examples using the semi-probabilistic procedures. These examples are:

    rnrn

  • a steel gate from navigation lockrn
  • U-frame lock wall (reinforced concrete)rn
  • a gravity flood wall (plain concrete)rn

rnDifferences in the design outcome, a measure for the safety, was often moderate between the different design concepts. As much as possible the findings have been compared and evaluated.

rnrnThe report finishes with a conclusion and recommendations for best practice. rn