THE ROLE OF EXPLANATIONS IN AN INTELLIGENT ASSISTANT SYSTEM

This paper sets out to illustrate the importance of transparency withi n software support systems and in particular for those intelligent ass istant systems performing complex industrial design tasks. Such transp arency (with the meaning 'clear' or 'easy to understand') can be achie ved by two distinct strategies that complement each other: (i) The des ign of intelligible systems that would avoid the need for in depth exp lanation. (ii) The flexible generation of those definitions or aspects of the system or domain that remain ambiguous. The paper illustrates that for the generation of useful explanations going beyond a simple j ustification of a problem solving trace, specific explanatory knowledg e must be acquired. By itself the problem solving techniques are not s ufficient. A new approach to acquire and model explanatory knowledge f or software systems is presented. The new four-layer explanatory model can be used to determine the range of explanation suitable for a give n systems domain. This model has been successfully used for the develo pment of an explanation component for the design assistant system ASSI ST that supports factory layout planning, in itself a complex design t ask. (C) 1997 Elsevier Science Limited.