Use Of Expert Systems Research Articles

Real-time expert systems benchmark for use in data acquisition

A set of rules is proposed to serve as a benchmark to evaluate real-time expert systems for use in data acquisition. The rules are focused on three kinds of issues: including time into a reasoning process, interfacing to external objects, and real-time presentation capabilities. Examples of implementations for G2 and RTworks expert systems are given. >

Expert System Usage, Hardware, Access and Features

Examines the relationships between the frequency of expert system use and the system′s hardware, access location, and features. Also examines the relationship between the expert system′s hardware and access location. The study is empirical, using a survey of marketing executives who work within marketing organizations employing expert systems. The findings include that the hardware type appears to influence the frequency of expert systems use. Daily use is dominated by mainframe computers, while weekly and monthly use is dominated by the microcomputers. Further, the frequency of expert system use increases with access availability and decreases as the expert system becomes less available. The dominant feature of these expert systems is the ability to perform what‐if‐analysis. When access location and hardware type are examined, the dominant hardware is the microcomputer. Further, particular hardware types tend to dominate specific access locations.

Method for generating rules for an expert system for use in controlling a plant

Method for generating rules for an expert system for use in controlling a plant

An expert system for the conceptual design of bridges

The design of any structure can be said to consist of two stages: the first being the conceptual design stage, in which the overall form of the structure is decided upon. The second stage consists of a more detailed structural analysis, during which calculations are carried out to verify the conceptual design choice and determine component dimensions. Both parts of the design process are of importance, and yet in terms of the research effort expended, the second stage of structural analysis has received far more attention. The probable reason for this emphasis is that the behaviour of structures under load is appropriate to numerical analysis, whereas conceptual design is more subjective, relying heavily on the experience of the engineer, and therefore does not lend itself to formal expression. Bridges are a prime example of the type of structure where conceptual design is of importance, and this paper describes the development of an Expert System for use in the conceptual design of a bridge. The information has been obtained from ‘expert’ engineers, with a wealth of experience in bridge design, and this paper discusses the methods used to elicit the expert knowledge from these engineers. Ultimately, the completed system will provide a source of information about this experience which is as yet unavailable to the young engineer. In addition, the work described has thrown light on the subject of conceptual design itself, an area poorly documented to date. It is shown that the use of an Expert System would provide a representation of the way in which the conceptual design process is carried out, as well as providing a useful advisory facility for the inexperienced designer.

A prototype expert system for hazard and operability studies

This paper describes the development of a prototype expert system for use in HAZard and OPerability (HAZOP) studies. The work is motivated by the fact that HAZOP studies are inherently logical exercises utilizing a process engineering knowledge base. The availability of an effective tool to aid engineers in design and operation would be of significant value. The knowledge domain required for HAZOP studies is wide. Hence the system must be built in modular form in order to be efficient. It must also be built in such a way that it is easy to incorporate additional knowledge into the system. The prototype developed in this work has been based on a PC version of Prolog. A basic inference engine, available in Prolog, has been enhanced and tailored into a user-friendly, interactive environment. Details of the knowledge organization and representation are discussed and an example session given. Further work is required to improve the system for serious industrial use through the addition of further knowledge modules, improved input—output facilities and optimization of the program.

Knowledge Processing for Construction Management Data Base

The objectives of project control systems go beyond documentation to recognition of problems and evaluation of their causes. Through the use of broad information systems, a construction manager today is able to assemble a wide variety of data about his projects or programs. Retrieval, analysis, and interpretation of the meaning of these data, however, usually require the user to have detailed knowledge about the structure and content of the data base and the use of a data retrieval language, and have further programming skills to allow the computer to perform analyses. The work reported in this paper has three goals: to ease retrieval with near‐natural language query capabilities; to acquire and accumulate data access and analysis knowledge; and to exercise such knowledge to find patterns in project data. This paper reports on the development of a knowledge‐based query system, Expert‐MCA, developed by coupling the techniques of data‐base management systems, natural language processing, and knowledge‐based expert systems for use in the U.S. Army military construction program (MCA). The system architecture is presented, followed by a description of how knowledge is represented and processed in answering English‐like queries. Finally, examples are used to explain its capabilities in more detail.

An expert system for early detection of cancer of the breast

In order to study the relevancy of exploring complex expert systems for use by the lay public, an expert system for the early detection of cancer of the breast has been developed. The system consists of a conversion between a microcomputer and a woman who is anxious about breast cancer. The conversation is divided into two parts: one is listening to the woman's symptoms regarding the breast then giving advice, the other is an explanation of breast cancer and how to detect it in its early stages. After listening to the woman's symptoms, the computer presents its conclusion and suggests courses of action the woman should take. The system is written in Prolog. It was tried by women in a small town at a festival for health, by female patients who visit our hospital, and received a positive reaction.

An expert system for use in the prescription of electronic augmentative and alternative communication devices

A prototype expert system (PES) has been developed to prescribe electronic augmentative and alternative communication (AAC) devices based on device selection technique for a nonspeaking population. An expert system offers many advantages to a variety of problem-solving activities and, in particular, to AAC device prescription. The appropriate device selection technique depends on the individual's capabilities and needs. Rules representing these relationships were developed and programmed on an IBM″ Personal Computer AT using the M.1®expert system development tool. The performance of the PES was evaluated by comparing computer-generated recommendations for selection technique to 10 case histories of nonspeaking individuals. Independently, a speech-language pathologist rated the performance of the PES using the same case histories. Results of this study demonstrate successful construction of the PES to the extent that its recommendations agreed with those in the case histories and of the speech-language pat...

Construction Techniques for Factory Floor Expert Systems

This paper addresses the specific problems encountered when building an expert system for use in a manufacturing process. These problems include: general unfamiliarity of the experts with expert systems, the diffused nature of the expertise, the low availability of the experts, and the users fear of being replaced or constrained by the computer system. The paper addresses these problems by presenting specific strategies developed in the construction of an expert system for use in controlling a plating process for electronic components.

Expert system focuses on industrial corrosion

St. Louis-based Materials Technology Institute of the Chemical Process Industries has launched a major project to develop an expert system for use by designers and plant engineers in their battle w...

Keynote Papers: Basic Concepts in Design for Producibility

“Are there basic principles and concepts that can put the design for producibility on a more scientific basis?” That is the subject of this keynote paper. To meet the needs of the society, products must be designed and manufactured. This design/manufacturing process involves at least five steps: the formulation of the product requirements (i.e., the generation of ideas for the product and for manufacturing processes, execution of manufacturing operations, and the development of marketing strategies for the final product. Some basic concepts have been advanced to address various aspects of the complex product life cycle design issues. The most subjective element of the design process which is difficult to rationalize is the formulation of functional requirements from a set of perceived needs. In order to facilitate the second element of the design process, i.e., idea generation, morphological techniques and knowledge based expert systems have been proposed. Once the ideas are generated, design decisions must be made to choose the best among those ideas proposed. There are two approaches in making design decisions: algorithmic and axiomatic. Prescriptive design procedure for assembly, which has been generated to lower the assembly cost, and knowledge based expert systems for use in CAD/CAM are examples of an algorithmic approach. On the other hand, the Design Axioms have been introduced to establish the general scientific basis for making design decisions in evaluating the requirements. The consequent theorems and corollaries provide specific criteria for “design for producibility.” The axiomatic approach provides the general conceptual framework and decision making basis, while the algorithmic approach facilitates the design process in solving specific problems.

Getting started in library expert systems research

In the 1960s, information science researchers pioneered in the design of computer-based document storage and retrieval systems. These efforts were crowned with success, and online systems are now in common use as reference tools. Today the new information science frontier is to design, develop, and test expert systems for use in libraries and other information centers. At UCLA we are exploring the applicability of artificial intelligence and expert systems for modeling the cognitive processes involved in cataloging. Specifically, Zorana Ercegovac, a doctoral student, is designing a prototype expert system in the limited domain of map cataloging that will seek to employ the reasoning used by expert catalogers in applying AACR2 rules. It is anticipated that the research results will shed some light on the way catalogers reason and conceptualize the structure of a catalog entry. The project is still in its initial stages, and in this presentation one can only indicate the design choices that need to be made, the reasons for the decisions made, and the problems encountered.

EMP HARDENING TOPOLOGY EXPERT SYSTEM (HARD TOP)

ABSTRACT This paper develops the architecture for an expert system for use in the EMP hardening of complicated systems, such as aircraft and ships. The best form for the topology knowledge base is one that mimics the topology data base: graphic. A specific graph model is explored as a candidate data base structure, and examples of the use of this data base are shown.

A consultant's view of expert systems

There is enormous potential for using expert systems, once the tools and expertise are available for developing the systems. After building an expert system for use in process control plants, several useful lessons have been learnt. There are various theoretical problems still to be overcome, but the major obstacle to the commercialization of expert systems is the practical problem of building cost-effective, usable systems. To solve this, tools will become more powerful and cheaper, while systems will become vertically-oriented, containing some application specific knowledge from the start. Artificial intelligence products will provide the means whereby computers will become useful to a whole new population of users.