Applying case-based reasoning to urban planning: A new planning-support system tool

Abstract

Because of the high complexity, uncertainty, and subjectivity of urban planning, it is a domain from which knowledge is often hard to extract and/or represent by the knowledge-representing methods of knowledge-based systems (KBS) which require explicit, definite, and generalized rules of thumb or causal models. Furthermore it is difficult for a user to understand and interact with the black-box inference process of the system. It is more suitable for solving routine problems but lacks the capability of storing case-dependent knowledge which is needed by planners to generate creative solutions or to deal with exceptional problems. Therefore, in the past, there has been little application of KBS in urban planning. In this paper we discuss the applicability of case-based systems-a methodology in which previous examples (cases) are used to solve, evaluate, or interpret a new problem-in urban planning which can help overcome the knowledge elicitation and black-box operation problems of rule-based and model-based KBS. Besides pointing out the advantages of case-based reasoning over conventional KBS in knowledge acquisition and user acceptance, we give an overview of its potentials as a planning-support system and examine some of the problems in using this method.