This title appears in the Scientific Report :
2019
Modellers meet practitioners: On 20% SFSF selection probability and random physical inventory verifications
Modellers meet practitioners: On 20% SFSF selection probability and random physical inventory verifications
Due to Germany’s phase out of nuclear energy production, discussion on current and future inspection regimes for long term spent fuel storage facilities (SFSFs) with static inventory has revived recently. Incurrent practice, in some States, randomly scheduled inspections (RSIs) of SFSFs are performe...
Saved in:
Personal Name(s): | Krieger, Thomas (Corresponding author) |
---|---|
Avenhaus / Burr | |
Contributing Institute: |
Nukleare Entsorgung; IEK-6 |
Imprint: |
2019
|
Conference: | ESARDA 41st Annual Meeting, Symposium on Safeguards and Nuclear Material Management, Stresa (Italy), 2019-05-14 - 2019-05-16 |
Document Type: |
Proceedings |
Research Program: |
Joint Programme on the Technical Development and Further Improvement of IAEA Safeguards between the Government of the Federal Republic of Germany and the International Atomic Energy Agency Nuclear Waste Management |
Publikationsportal JuSER |
Due to Germany’s phase out of nuclear energy production, discussion on current and future inspection regimes for long term spent fuel storage facilities (SFSFs) with static inventory has revived recently. Incurrent practice, in some States, randomly scheduled inspections (RSIs) of SFSFs are performed such that in each year any particular SFSF of the State is selected for RSIs with a required selection probabilityof (say) 20%. Three questions arise from such a requirement: First, how are the SFSFs selected for the RSIs and when are the RSIs performed? Second, how many – a fixed or expected number of RSIs – have to be performed to achieve the required selection probability? Third, if a fixed or expected number of RSIs assure the 20% selection probability, is this number equal to 20% of the number of SFSFs in the State? Regarding future inspection regimes for static long term SFSFs, instead of a regular physical inventory verification (PIV) only random PIVs are being considered. Here the question is to calculate the probability that a PIV is performed in any year, under the requirement that at least one PIV has to be scheduled within (say) 5 years?This paper addresses all these questions, and instead of a typical conference paper the format is a fictitious meeting between modellers and practitioners, e.g. statisticians and inspectors. This way both partys’ ideas and concepts and the ways of thinking are made explicit, and the surprises that can happen on both sides in the course of those meetings are illustrated. |