Quality Checks Logit Human Reliability (LHR): A New Model to Evaluate Human Error Probability (HEP)

Abstract

In the years, several approaches for human reliability analysis (HRA) have been developed. The aim of the present research is to propose a hybrid model to evaluate Human Error Probability (HEP). The new approach is based on logit-normal distribution, Nuclear Action Reliability Assessment (NARA), and Performance Shaping Factors (PSFs) relationship. In the research, shortcomings related to literature approaches are analyzed, especially the limitations of the working time. For this reason, PSFs after 8 hours (work standard) during emergency conditions were estimated. Therefore, the correlation between the advantages of these three methodologies allows proposing a HEP analysis during accident scenarios and emergencies; a fundamental issue to ensure the safety and reliability in industrial plants is emergency Mmnagement (EM). Applying EM methodology, two main aspects are analyzed: system reliability and human reliability. System reliability is strongly related to the reliability of its weakest component. During incidental situations, the weakest parts of the whole system are workers (human reliability) and accidental scenarios influence the operator’s ability to make decisions. This article proposes a new approach called Logit Human Reliability (LHR) that considers internal and external factors to estimate human reliability during emergencies. LHR has been applied in a pharmaceutical accident scenario, considering 24 hours of working time (more than 8 working hours). The results highlighted that the LHR method gives output data more in conformity with data banks than the conventional methods during the stress phase in an accident scenario.

An investigation into accidents in laboratories in universities in China caused by human error: A study based on improved CREAM and SPAR-H

“…The development of HRA has gone through several stages: from first generation methods that considered the effects of Performance Shape Factors (PSFs) relating to the reliability of human behaviour [ [27] , [28] , [29] ], to second generation methods that emphasized the effects of the environment on the reliability of human cognitive understanding [ 30 , 31 ], to third generation methods that focused on analyzing human errors in dynamic scenarios [ 32 ]. Recently, a hybrid model that combined second and third generation methods has been proposed [ [33] , [34] , [35] ]. A PSF was defined as a feature of the organization, task, or environment that specifically lowered or improved behaviour, hence, separately, improving or lessening the probability of human error.…”

The effect of safety climate in workplaces on productive organizational energy of employees: a research in textile industry

“…, 1990). A significant portion of workplace accidents are caused by human error (Etman and Halawa, 2007; Chikudate, 2009; Di Bona et al ., 2021a, b; Bathrinath et al. , 2021; Velmurugan et al ., 2022).…”

National-Standards- and Deep-Learning-Oriented Raster and Vector Benchmark Dataset (RVBD) for Land-Use/Land-Cover Mapping in the Yangtze River Basin

“…That generally results in errors of LULC labeling. To alleviate the issue, some researchers have delved into various aspects, as follows: on one hand, much research adopted quality control methods to reduce human errors [55][56][57][58][59][60][61][62][63][64][65]. For example, Qi et al [45] rely on many technicians to train data labeling several times until reaching a predetermined and reliable confidence score.…”