Proceedings of the Eighth International Conference on Probabilistic Safety Assessment & Management (PSAM)
Download citation file:
- Ris (Zotero)
- Reference Manager
Although the details of the various human reliability analysis (HRA) methods vary considerably, one feature shared by virtually all methods is the reliance on some type of performance shaping factors (PSFs) for modifying a nominal or base human error probability to account for the various “environmental” or context-specific influences on human performance. Unfortunately, each method has chosen to adopt its own unique list of PSFs. Often these PSFs themselves are composites of other, more basic influences. For example workload, a commonly used PSF can be viewed as a composite of the specific and objective task-burden requirements placed on the human and the relative capability of the human to handle those task requirements. Obviously, a complex mathematical calculation (task-burden) would impose a much greater workload on a person with a relatively low level of education and intelligence, compared to a well-educated, highly intelligent person. Likewise, stress, another commonly used PSF is a composite of not only the environmental circumstances experienced by an individual (or crew) during a particular event, but also the person's ability to cope with such circumstances, which in turn might depend on training and user aids (such as procedures) as well as a particular individuals innate ability to cope with pressure. Therefore, despite identical environmental conditions the variability among individuals will produce variability in the stress levels (and workload). The approach proposed here defines PSFs in a mutually exclusive taxonomic framework such that the various influencing factors are reduced to more basic components. This type of distinction among PSFs provides a more objective determination of how they influence the human error probability. Furthermore, once a base HEP for a nominal situation has been developed, this more objective approach facilitates a relatively robust estimation of a context-specific HEP.
The framework suggested here defines PSFs (albeit somewhat subjectively) according to three distinct categories. These are aimed at capturing not only the expected influences on human performance, but also the variability in the potential conditions that could produce a range of human responses. These three categories are:
1. Population Capability (PC) — defined as the basic ability of a certain population (e.g., NPP control room operators) to handle expected tasks. This includes, but is not limited to intelligence, experience and general education. However, it does not include organizational-sponsored training, which is covered elsewhere.
2. Organizational Environment (OE) — defined as the general work-place environment in which the individuals function. This includes, but is not limited to work-related training, quality of work aids (e.g., procedures and instrumentation), and safety culture.
3. Event Specifics (ES) — defined as the unique details of a specific event or situation. This includes, but is not limited to time constraints (time available to perform a specific action and the time required to perform it), complexity of the required task, and the uniqueness of a specific situation with respect to the expectations associated with particular work-responsibilities (i.e., is it a situation for which the individual has been trained, and are there applicable procedures in place).