
Expert Systems are simply custom-written computer programs that are "expert" in some narrow problem area, and embody (to a certain extent) a true human expert's knowledge, experience and problem-solving strategies. Expert Systems have been used in many problem areas, such as medicine, chemistry, geology, meteorology, computer systems, etc. Expert Systems can generally be used in problem areas that: do not require common sense to solve are well understood data that is input to the expert system can be described objectively Human expertise is scarce Expertise needed in many locations, or in hostile environments there are considerable advantages in reaching an accurate answer quickly

For Decision Support	For Decision Making
To remind a human expert of issues to consider, alternatives to explore, etc. that the human expert may have missed in his decision making. This type of Expert System is commonly found in the area of medicine.	To aid a person in problem solving an area that he/she is unfamiliar in, or is inexperienced in. This type of expert system is commonly found in industrial systems.

Human Experts	Expert Systems
Skills and knowledge can deterriorate over time Training human experts is an expensive and lengthy process that may not even guarentee good results Susceptible to emotional and psychological factors that can impair decision making Scarce and typically command high salaries	Provides permenant expertise Artificial expertise avialable from expert systems is easily reproduced and transferred, simply by duplicating the computer program Provides consistent and reproducible results Expert Systems are relatively cheap to operate and maintain

Exactly how expert systems perform their problem solving depends largely on how their expertise is internally represented. Some of the more common methods by which expert systems internally represent their expertise are:

XCON is a commercial expert system that configures VAX computers for DEC and is the biggest and most mature rule-based expert system in operation. From a customer's order XCON decides what components are needed to assemble a complete computer system, and also determines the spatial relationships between the components. It then outputs this information as a set of diagrams to technicians who physically assemble the computer system.

PROSPECTOR aids geologists in their search for ore deposits. Given field data about a geological region, it can determine the propability of discovering a range of ore deposits, including sandstone uranium, massive sulfide, carbonate lead/zinc, etc. Its expertise is based on geological rules which form models of ore deposits, and a database of known rocks and minerals. In 1980, PROSPECTOR analyzed geological data from a site near Mount Tolman in eastern Washington and predicted the existence of molybdenum (an ore) in a particular location. Subsequent drilling by a mining company confirmed PROSPECTOR's prediction.

Limitations of expert systems

Cannot easily adapt to new/unusual situations; not creative
Do not learn by experience
Not good at representing spatial knowledge
No common sense
Expensive and time-consuming to develop