Work, especially that in complex, dynamic workplaces, often requires subtle, local judgment with regard to timing of subtasks, relevance, importance, prioritization and so forth. Still, people in Nuclear Industry seem to think safety results from people just following procedures. In the wake of failure it can be tempting to introduce new procedures and an even stricter "rule following culture". None, or at least very little, attention is given to tacit knowledge and individual skills. I am aiming to highlight the inadequacy of putting too much trust in formalization and that reporting and trending of events will contribute to increased learning, an increased nuclear safety and an efficient operational experience. The ability to interpret a situation concrete depends on proven experience in similar situations, analogical thinking and tacit knowledge. In this essay I intend to problematize the introduction and use of so-called Corrective Action Program (CAP) and computerized reporting systems linked to CAP in the Nuclear Industry. What I found out is that the whole industry, from regulators to licensees, seems to be stuck in the idea that the scientific perspective on knowledge is the only "true" perspective. This leads to an exaggerated belief in that technology and formalized work processes and routines will create a safer business. The computerized reporting system will not, as the idea was from the beginning, contribute to increased nuclear safety since the reports is based on the trigger and not the underlying causes and in-depth analysis. Managing safety by numbers (incidents, error counts, safety threats, and safety culture indicators) is very practical but has its limitations. Error counts only uphold an illusion of rationality and control, but may offer neither real insight nor productive routes for progress on safety. The question is why the CAP, error counts and computerized reporting systems have had such a big impact in the nuclear industry? It rests after all, on too weak foundations. The answer is that the scientific perspective on knowledge is the dominating perspective. What people do not understand is that an excessive use of computerized systems and an increased formalization actually will create new risks when people lose their skills and ability to reflect and put more trust in the system than in themselves.

Managing safety at a nuclear power plant is about a complex system with demanding technology under time pressure where the cost of failure is exceptionally high. Swedish nuclear power plants have over the last few years introduced Pre-job Briefing and other so called Human Performance Tools to advert errors and strengthen control. By using the Systemic Resilience Model different views of safety are taken to understand the origin of the signals that leads to a Pre-job Briefing, and how the signal is interpreted, re-interpreted, and presented. The study took place at a Swedish nuclear power plant and included four days of observations and 20 interviewees. The thematic analysis shows a similarity between mentioned origins of Pre-job Briefings and the intended use of Pre-job Briefing. Characteristics of a High Reliability Organisation is shown in practice by a culture of that one will to have a Pre-job Briefing is enough, that sharp end workers is used as a valuable resource for safety and a systematic support to screen jobs over time without influencing non-job related factors. The signals acted upon matched well with the intended, and personnel get several opportunities to evaluate the signals together, striving for best possible circumstances. The Systemic Resilience Model was successfully applied together with a thematic analysis, which strengthens its validity as a holistic model that combines different views of safety in one coherent model. SyRes allowed to present additional themes, leaving the question at what stage SyRes is optimally implemented in a thematic analysis.