Knowledge-based tasks

Perform task analysis (see Chapter 4) and identify skill, rule or knowledge-based tasks or aspects of tasks (the flow diagram in Figure 2.7 may be used to assist in this classification). 

The practical implications of this experiment are that when evaluating the effects of shift work due to circadian effects, the type of task being carried out by the worker must be taken into account. For example, skill-based tasks would be expected to exhibit the performance changes characteristic of low memory load tasks, whereas performance variations in knowledge-based tasks would be expected to follow the pattern of high memory load tasks. Performance on rule-based tasks may depend on the degree of frequency of use of the rules, which in turn may determine the memory load. If these results were confirmed by further process plant studies, it would have implications for when different types of operation (involving different levels of memory load) should be scheduled to reduce circadian rhythm effects and minimize errors. 

These results are directly compatible with models of human information processing (Rasmussen, 1983). Calculation is, in most cases, a skill-and-rule-based activity and requires relatively little mental effort. In contrast to this, selection and evaluation, as well as the activities around task clarification, are more rule-and knowledge-based tasks that require higher mental effort. According to the stress-strain concept, these phases result in a higher strain level. 

Knowledge acquisition subsystem The task of the knowledge acquisition subsystems is to assemble and upgrade the knowledge base. A major task is to verify the data and check for consistency. 

Definition 2 Knowledge-based systems are computer programs that encode symboHc knowledge about domains and tasks, and solve problems by manipulating this knowledge using quaHtative techniques. 

These two examples hint at a few of the reasons for the importance of knowledge-based systems. A medical faciHty may handle hundreds of infectious disease cases a year. Speedy, accurate diagnosis of these cases, aided by a system such as Mycin, may help the medical faciHty handle more patients, more effectively. Likewise, configuring large computer systems composed of many components can be a time-consuming and error-prone task. 

Alternative technologies should also be considered, and the reasons for not using them should be justifiable. For example, database technology is not the right choice if a task requires reasoning that goes beyond retrieval of stored data based on well-defined criteria. At the same time, many problems that are stated in a symboHc way can be formulated mathematically, and in fact can be better solved numerically. For such problems, knowledge-based systems are not the appropriate answer. 

Eor a number of cognitive or interpretive tasks, there are alternatives to mainstream knowledge-based systems that may be more appropriate, especially if adaptive behavior and learning capabihty are important to system performance. Two approaches that embody these characteristics are neural networks (nets) and case-based reasoning. 

An influential classification of the different types of information processing involved in industrial tasks was developed by J. Rasmussen of the Rise Laboratory in Denmark. This scheme provides a useful framework for identifying the types of error likely to occur in different operational situations, or within different aspects of the same task where different types of information processing demands on the individual may occur. The classification system, known as the skill-, rule-, knowledge-based (SRK) approach is described in a... 

Perform task analysis and classify skill, rule or knowledge-based behaviors involved in the scenario being evaluated. 

Detailed procedures will only be required in unusual situations where the usual rules of thumb do not apply and the worker is likely to be in the knowledge-based mode. In Chapter 4, and case study 3 in Chapter 7, a systematic framework for developing procedures, in which their format and content is based on a detailed analysis of the tasks to be performed and the normal skill level of the person who will perform the tasks, will be described. 

Ramesh, T. S., Davis, J. F., and Schwenzer, G. M., Knowledge-based diagnostic systems for continuous process operations based upon the task framework, Comput. Chem. Eng. 16(2), 109-127 (1992). 

---