Search

This paper provides human resource managers with important guidelines when applying expert systems to human resource domains, by reviewing system characteristics, potential benefits and limitations, and appropriate domains to be selected, based on the literature. Structural and procedural aspects of expert systems development in human resource management (represented by a wheel model) and problem descriptions of each expert system respectively are important research results in this paper. We also show semantic networks or semantic nets used for a knowledge representation methodology on such major human resource management activities as human resource planning, recruiting, compensation and labor-management relations. Knowledge representation is a key component in the development stages of expert systems. [ABSTRACT FROM AUTHOR]

Although MIS professionals had predicted that expert systems would improve productivity enormously, and a number of expert system application success stories have been reported, the expert system revolution has not yet happened. Moreover, there have been many cases, less well publicised, where expert systems have failed. Most problems concerning expert systems failure stem from non-technical issues such as cognitive and psychological problems, rather than such purely technical issues as an inference engine and an expert system shell. Here, the major reasons for expert systems failure and the need to consider human factors are discussed. We then propose human factor principles that can help designers handle most of these non-technical problems and many technical ones elegantly, improving the performance and acceptance of expert systems. Finally, some human factor guidelines for expert systems are presented so that these considerations may be incorporated into expert systems development in a clear and comprehensible manner. [ABSTRACT FROM AUTHOR]