ABSTRACT

Since the 1980s, human error analysis (HEA) techniques have developed to address an established need in safetycritical industries to identify the human contributions to system performance. Despite the availability of techniques for identifying, analyzing, and assessing equipment-and process-related failures, “human error” was seen as more elusive; a “new disease” seemingly responsible for major disasters. The very high reliability and refinement of mechanical and electronic components, the complexity of systems, the role assigned to the human operator, and latent organizational factors apparently led to a dramatic increase in the human contribution to accident development. While human errors cannot be predicted with the accuracy of hardware failures, many authors agree that human errors are not random. Nagel (1988) notes that human errors seem to be lawful in the sense that they are predictable and that most humans tend to make errors that follow certain patterns under a variety of circumstances. Hence, several well-known techniques or methodologies for the identification and classification of human errors emerged, known as human error analysis (HEA) or human error identification (HEI) techniques. Many of these techniques were influenced heavily by Rasmussen et al.’s (1981) skill-, rule-, and knowledge-based behavior framework and Reason’s (1990) classification of slips, lapses, mistakes, and violations. While these techniques were primarily associated with the nuclear and process industries, HEA was also applied in other new sectors, such as manufacturing, rail, consumer products, public technology, and medicine.