5 Conclusion
Although a great deal of research has been pub- lished in the context of expert system develop- ment, very little attention has been given to the efficiency and effectiveness of the development process. In order to examine all relevant facets in the process of knowledge engineering, the soft systems concept is introduced. Applying the soft systems concept on expert systems develop- ment can be a useful way of thinking about the overall situation of the development and mainte- nance of the expert system. It provides a frame- work for visualising internal and external fac- tors as an integrated entity. This concept also allows recognition of the proper place and func- tions of all associates. Through the soft sys- tems concept, the major problems that may oc- cur in the development life cycle can be encoun- tered by the relevant parties. The methodol- ogy introduced in this paper can be regarded as a strategic model for expert system develop- ment while other existing knowledge-based sys- tem development methodologies are considered as tactical approaches. In other word, the soft systems approach help establishing the develop- ment environment and the technical methodolo- gies. These two kinds of methodologies should be complementing each other and i co-exist to overcome both soft and hard problems. The success of a knowledge-based system develop- ment depends on the determination of mea- sures of performance for individual development processes. However, a limited effort has been contributed to measures of performance for the ESDS. We are currently investigating this is- sue, so that the control mechanisms for evalu- ating the effectiveness of the ESDS can be es- tablished.
References