Knowledge-based behaviour

Knowledge-based behaviour where people have to cope with situations which are new to them and for which they have no routines. This is a fully conscious process of interaction with the situation to solve a problem. 

A further relationship exists between these error Qpes and Rasmussen s [3] classic model of skill, rule and knowledge-based behaviour. Slips and l >ses tend to occur at the skill-based level, whereas mistakes occur at the rule-based and knowledge-based levels. Rule-based mistakes are primarily due to mis lied expertise, where some pre-established plan or problem solution is lied inappropriately. Knowl ge-based mistakes generally occur due to a lack of e q)ertise, where no off-the-shelf solution exists and an individual is forced to work out a plan of action fiom first principles. 

A relatively simple model has been developed by Rasmussen to quantify human error rates based on the level of training (Rasmussen (1979, 1981)). This model divides the behaviour into three basic categories skill-based, rule-based, and knowledge-based behaviours. 

Determine operator behaviour - For each of the tasks identified in Step 2, determine the required operator s behaviours. Three types of behaviours are considered namely, skill-based, rule-based or knowledge-based behaviour. These behaviours are discussed in Sections 9.3. 

As was discussed in Chapter 2 Rasmussen has provided the basic model of human error based on three levels of behaviour skill-, rule- and knowledge-based (S-B, R-B, K-B). This SRK model has been operationalised to describe operator errors in process control tasks by combining it with characteristic task elements, which as a whole cover the entire spectrum of operator subtasks. 

For an estimate of an input quantity that has not been obtained from repeated observations, the associated estimated variance or standard uncertainty is evaluated by scientific judgement based on all available information on the possible of its variability. This is the case of Type B standard uncertainty. The pool of information may include previous data, experience with general knowledge of behaviour of relevant materials and instruments, manufacture s specifications, or data provided in calibration and other certificates or uncertainties assigned to references data taken from handbooks. 

The perception of a perfume depends, in the first place, upon the presence of odorant molecules in the air, and upon their nature and concentration. Most perfume starts off life as a liquid comprising a wide variety of molecules and of a known composition. In general, perfumers do not have a corresponding knowledge of the composition of volatiles in the air above such a mixture, except on those occasions where headspace analysis has provided hard analytical data (see Chapter 12). Perfumers therefore have to build up knowledge bases that summarize the olfactory behaviour of hundreds of ingredients under many different circumstances ... 

Conversely, qualitative models, used in ITS, are designed so that the behaviour of the student could be explained. The knowledge base and the automatic problem solving capabilities of the ITS play an important role because the system compares the student answers to the expert module answers and determines why and how the student reached a certain goal. Qualitative models may be classified in overlay, bug and simulation models [9, 12]. 

Reason goes further and highlights the likely feilure modes at each level of behaviour. For exanq>le skill-based enors occur due to control-mode fiulures of both inattention and over-attention. Rule-based errors can arise fiom the misapplication of rules or the q>plication of incorrect rules. Knowledge-based errors can arise due to selectivity, biased reviewing and a number of other fiictors. 

This approach is well known to us from a different part of the engineering knowledge base thermodynamics. For example, the pressure, P, of a volume, V, of gas is the average of the forces exerted by the individual molecules on the boundary of the volume, and the macroscopic behaviour of the gas is described by the simple equation PV = RT. This relationship between the three parameters P, V, and T holds as they change in time as part of some process, but the change must be slow compared to the timescale on which the microscopic collision processes take place, so that at any point in time, these microscopic processes experience essentially a static or equilibrium environment. 

Rule-based behaviour is at work when the operator does not have the same level of practice at performing the required task, but has a clear knowledge of the procedures. There may be some hesitation in recalling any procedure, the procedure may not be carried out in the proper sequence, or any step may not be performed precisely. 

Dimension Errors based on automatic behaviour Rule-based Knowledge-based errors errors... 

---