Improvement of accuracy and precision in non-destructive assay of waste from decommissioning of nuclear installations

The decommissioning of waste arising from nuclear installations requires declaration or free-release, i.e. a quantification of the activity content. For this purpose different information like a-priori data, non-destructive assay data and data benches are available, but are rarely all used together for quantification. Thus, the potential in optimising accuracy and precision is by far not exhausted. With increasing information, the activity range can be more and more limited, i.e. the accuracy will be increased. This is demonstrated qualitatively in Fig. 1. On the other hand, the available information is conventionally not sufficient for an exact quantification. Therefore several assumptions are made whose accuracy and quality are not (exactly) known. The impact of this ''lack-of-knowledge'' factor in activity determination on accuracy and precision is mostly neglected or is considered only empirically. The basic reason for these shortcomings is the absence of an appropriate evaluation program, considering all different information as well as quoting assumptions involved by reliable uncertainties. The development, implementation and testing of such an evaluation program is the aim of a project, whose actual status is presented. (orig.)