Difference between revisions of "Expert system"

Latest revision as of 10:17, 15 March 2016

Definition

A data processing system that provides for solving problems in an expert manner within a given field or application area, by drawing inferences with the aid of a knowledge base developed from human expertise. An expert system is a branch of artificial intelligence. See Artificial intelligence and Knowledge base

Description

An expert system is a rule-based system that attempts to reproduce the performance of one or more human experts, most commonly in a specific problem domain and is a traditional application and/or subfield of artificial intelligence. The expert system provides a user with advice and can explain why it asks the questions it does, and how it draws the given conclusions. An expert system should be able to handle uncertain information.

Expert systems are most valuable to nuclear organisations which have a high-level of know-how experience and expertise that cannot be easily transferred to other members. They are designed to carry the intelligence and information found in the intellect of experts and provide this knowledge to other members of the organization for problem-solving purposes.

There are generally three individuals having an interaction with expert systems. The main one is the end-user - the individual who uses the system for its problem solving assistance.

In the building and maintenance of the system there are two other roles: the problem domain expert who builds and supplies the knowledge base providing the domain expertise, and a knowledge engineer who assists the experts in determining the representation of their knowledge, enters this knowledge into an explanation module and defines the inference technique required to obtain useful problem solving activity.

Usually, the knowledge engineer will represent the problem solving activity in the form of rules which is referred to as a rule-based expert system. When these rules are created from the domain expertise, the knowledge base stores the rules of the expert system.

What does a 'Rule-based system' mean?

In expert system technology, the knowledge base is expressed with natural language rules IF ... THEN ... For examples : "IF it is living THEN it is mortal" "IF his age = known THEN his year of birth = date of today - his age in years"

This formulation has the advantage of speaking in everyday language which is very rare in computer science (software is coded). Rules express the knowledge to be exploited by the expert system. There exists other formulations of rules, which are not in everyday language, understandable only to computer scientists. Each rule style is adapted to an engine style. The whole problem of expert systems is to collect this knowledge from the experts. There are methods but almost all are usable only by computer scientists.

Typical areas of expert system applications:

• Troubleshooting technology issues (e.g. “Equipment does not work correctly, what to do?”);
• Medical diagnostics (e.g. "I have a pain in my lower abdomen, what should I do?");
• Product selection (e.g. "Which device is optimal for my system?)

Expert System Architecture

An expert system has 3 components:

• The Knowledge Base - where the information is stored in the expert system in the form of facts and rules (basically a series of IF statements). This is where the programmer writes the code for the expert system.

• The User Interface - where the user interacts with the expert system. In other words where questions are asked, and advice is produced. As well as the advice that is output, the user interface can output the justification features of an expert system. This is either How justification - where the system justifies its reasoning for providing a piece of advice or Why justification - where the system justifies why a particular question is being asked.

Justification allows the user piece of mind about why a question is asked or a piece of advice is provided, and can increase their confidence in taking such advice. It also makes it easier for the programmer of the system to ensure that it works correctly as it will flag up areas where the expert system provides advice that is not intended by the programmer.

• Inference Engine - this applies the facts to the rules and determines the questions to be asked of the user in the user interface and in which order to ask them. This is the 'invisible' part of the expert system, which is active during a consultaion of the system (when the user chooses to run the program).

An expert system can use 2 different methods of inferencing - Forward Chaining and Backward Chaining.

A Backward Chaining system (a goal driven system) works with the system assuming a hypothesis of what the likely outcome will be, and the system then works backwards to collect the evidence that would support this conclusion. Expert systems used for planning often use backward chaining.

A Forward Chaining expert system (a data driven system) simply gathers facts (like a detective at the scene of a crime) until enough evidence is collected that points to an outcome. Forward chaining is often used in expert systems for diagnosis, advise and classification, although the size and complexity of the system can play a part in deciding which method of inferencing to use.

Expert System Technology in Nuclear Industry

There are several hundred nuclear expert systems developed by electric utilities, umbrella R&D organisations, private companies, nuclear vendors, universities and governmental organisations.

The major areas of expert system technology use in nuclear industry are the following:

• engineering tools

• capturing human expertise

• plant design

• facility management

• maintenance planning

• interactive and real - time diagnostics

• decision support

• emergency response

• operator models and control

Most of the appeal of expert systems to the nuclear industry originates with the structure of those systems. Expert systems are very simple entities consisting of a knowledge base, an inference mechanism, and a user interface.For many nuclear applications, one must also add a component for the real-time acquisition of data. The function of the expert system is first to identify the current plant condition and then, via its inference mechanism, ti compare the predecessor clauses of each production rule against the observed plant status. The major advantage of this approach is that the knowledge base and inference mechanism are separate.

For the nuclear industry this means that, as the plant's layout is changed or as new regulations are imposed, the knowledge base can be updated without incurring the need to revise the inference mechanism.

Another feature of the expert systems approach that the nuclear industry finds appealing is the capability of the methodology to generate an explanation for its conclusions. Specifically, once a particular action has been identified as being appropriate, the system can print out a statement to the effect that such an action is required because the observed conditions exist. Moreover, it can cite the relevant supporting regulations. Another benefit to the nuclear industry from this explanatory aspect of expert systems technology is that it facilitates the preparation of the written justifications that must be maintained as documentation for most decisions.

Third major advantage to the use of expert systems within the nuclear industry is that much tedious work can be eliminated. For example, checking planning maintenance and scheduling activities against the applicable quality assurance standards and surveillance requirements is a process that is normally performed by skilled, experienced personnel. An expert system can do much of the drudgery and leave skilled personnel free to address those few questions that really merit their attention.

Expert systems are also of interest to the nuclear industry for their potential impact on plant operations. Analytic methods are currently the techniques for addressing technical problems. Expert systems may create another option that, if used in parallel with existing approaches, might result in more efficient methods for problem solving.

Problems that are "time critical" for NPP operation are candidates for expert systems. These are problems that may occur at any hour of the day or night and that must be quickly resolved in order to maintain system operation. An example is turbine-generator diagnostics. The availability of an expert system may avert the need for a plant shutdown pending the arrival from off-site of skilled analysts.

Expert systems technology has the potential to make a significant contribution to the reliable operation of nuclear power stations, but it necessary to note that application of the expert systems technology is not without the possibility of failure.

Common pitfalls

It is necessary to take into account the following:

• If the problem cannot be solved by human experts using all available resources, then it cannot be solved by an expert system. Expert systems are just tools to assist those already familiar with the problem. Expert system are not a substitute for human.

• An extensive knowledge base is essential to the functioning of an expert system. If the necessary knowledge is not available, then the expert systems approach should not be attempted.

• A problem should not be considered for solution using an expert system unless there are recognized human experts in the field. Moreover, problems amenable to an expert system are those that would take a human expert several hours to solve.

• Sometimes the implementation of an expert system can require a substantial change in work habits, and therefor it may be met with high resistance.

Related articles

Information technology

Yellow pages

References

1. John A. Bernard, Takashi Washio. Expert Systems Applications Within the Nuclear Industry.
2. Expert system on Wikipedia