TY - GEN A1 - Ewert, Uwe A1 - Jaenisch, Gerd-Rüdiger A1 - Osterloh, Kurt A1 - Zscherpel, Uwe ED - Czichos, Horst ED - Tetsuya Saito, ED - Leslie Smith, T1 - Industrial Radiology T2 - Springer handbook of materials measurement methods KW - Fundamentals KW - Particle-Based Radiological Methods KW - Film Radiography KW - Digital Radiological Methods KW - Public Safety and Security PY - 2006 SN - 978-3-540-20785-6 SP - 844 EP - 858 PB - Springer CY - Berlin AN - OPUS4-14473 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Osterloh, Kurt A1 - Jaenisch, Gerd-Rüdiger T1 - Ways to understand and approach rare events T2 - ECNDT 2014 - 11th European conference on non-destructive testing (Proceedings) N2 - As a matter of fact, avoiding unexpected events with an undesired outcome is an element of survival strategies. Such events are encountered unexpectedly mainly because they occur rarely. Both, safety and security measures are the main pillars to prevent them by appropriate inspections. Common tools in both areas e.g. are radiological technologies enabling an insight into objects to detect suspicious features without even touching them. Since any of these measures is linked to efforts, costs or even obstructions of ongoing processes, it needs a rationale to invest into an appropriate activity. A putative objection to take action in this direction always could be the question 'when and how often it could happen'. This gave rise to find approaches how to define 'rare events' and how to deal with them. Since they entail both, the frequency of occurrence and the unpleasantness of the possible outcome make them to have something in common with the definition of risk: a combination of the probability of occurrence of harm and the severity of that harm. Tackling putative consequences is one side of the coin whereas understanding the rareness of an event is the other, an aspect that worries but not always fully understood. As a first step in approaching the subject 'rare events', the putative occurrence rate is considered in terms of probability distribution functions or their cumulative ones, resp. The problems of estimating an incidence of such an event will be tackled subsequently with the problem of assessing the reliability of diagnostic measures. Any numeric approach of dealing with rare events inevitably remains an ill defined or 'ill-posed' problem that needs additional information for a reasonably satisfying solution. Ways to ease this situation can be found in utilizing additional information, also commonly called prior knowledge, that might be introduced via the Bayesian inference or by regularization algorithms. Simplified models will demonstrate how to apply such tools. As a consequence, there are ways helping to avoid unexpected ad 'surprises' by taking adequate measures in due time upon the correct perception of certain indications. T2 - ECNDT 2014 - 11th European conference on non-destructive testing CY - Prague, Czech Republic DA - 06.10.2014 KW - NDT-wide KW - Public security KW - Technical safety KW - Reliability KW - Indications and inspections KW - Compensation of insufficient knowledge PY - 2014 UR - http://www.ndt.net/events/ECNDT2014/app/content/Paper/562_Osterloh.pdf SN - 978-80-214-5018-9 SP - 1 EP - 9 AN - OPUS4-32302 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Osterloh, Kurt A1 - Jaenisch, Gerd-Rüdiger T1 - Rare events - a probability approach or an ill-posed problem? JF - Insight N2 - Rare events are understood to be events occurring once in a while but with dramatic consequences. Their occurrence cannot be predicted precisely, only a probability might be estimated, for example from past experiences. However, it might be rather misleading to attempt to describe them by distribution curves that apply for frequent or repeated observations, such as the Gaussian bell shape. Alternative distributions have been introduced to characterise the intervals at which a certain event may occur. It is the aim of technical safety and public security to prevent adverse events. Detectable indications that are typical for their course and are observable have to be identified before an incident occurs. Since they should be characteristic for such cases, they themselves also constitute rare events. The problem encountered in any detection system is that nothing is perfect. As in medical diagnostics, true indications may be missed or false test responses may pretend to be something that does not exist. Balancing missed indications with false positive calls is achieved with the aid of the so-called receiver operating characteristics (ROC). However, with the aid of Bayes’ inference it can be shown that identifying signs of a rarely occurring indication is like looking for a needle in a haystack, even with an excellent detection Approach with a low miss rate and an even lower probability of false calls. The inclusion of additionally available information may lead to a more effective search strategy. When employing imaging methods for detecting flaws or illicit items, the identification of rare indications can be impeded by blurring noise or overlapping items. The identification of the features sought can be supported by including information on their typical characteristics by regularisation algorithms. The strategy of such an approach is demonstrated in a simplified example with a plain geometric figure (circle) corrupted with structural noise. The shape of the original figure was clearly recovered. In general, search strategies should aim at an indication typical for the event to be prevented; otherwise, alternative approaches have to be considered, including, perhaps, serendipity. KW - Rare events KW - Bayes statistics KW - Ill-posed problems PY - 2016 DO - https://doi.org/10.1784/insi.2016.58.1.46 SN - 0007-1137 VL - 58 IS - 1 SP - 46 EP - 51 PB - British Institute of Non-Destructive Testing CY - Northampton AN - OPUS4-35312 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER -