Human error has been cited as a primary cause or contributing factor in disasters and accidents in industries as diverse as nuclear power (e.g., the Three Mile Island accident), aviation (see pilot error), space exploration (e.g., the Space Shuttle Challenger Disaster and Space Shuttle Columbia disaster), and medicine (see medical error). Prevention of human error is generally seen as a major contributor to reliability and safety of (complex) systems.
Human error means that something has been done that was "not intended by the actor; not desired by a set of rules or an external observer; or that led the task or system outside its acceptable limits". In short, it is a deviation from intention, expectation or desirability. Logically, human actions can fail to achieve their goal in two different ways: the actions can go as planned, but the plan can be inadequate (leading to mistakes); or, the plan can be satisfactory, but the performance can be deficient (leading to slips and lapses). However, a mere failure is not an error if there had been no plan to accomplish something in particular.
Human error and performance are two sides of the same coin: "human error" mechanisms are the same as "human performance" mechanisms; performance later categorized as 'error' is done so in hindsight: therefore actions later termed "human error" are actually part of the ordinary spectrum of human behaviour. The study of absent-mindedness in everyday life provides ample documentation and categorization of such aspects of behavior. While human error is firmly entrenched in the classical approaches to accident investigation and risk assessment, it has no role in newer approaches such as resilience engineering.
There are many ways to categorize human error.
The cognitive study of human error is a very active research field, including work related to limits of memory and attention and also to decision making strategies such as the availability heuristic and other cognitive biases. Such heuristics and biases are strategies that are useful and often correct, but can lead to systematic patterns of error.