Expert Systems for Planning and Spatial Decision Support

Abstract

Expert systems (ES) are a well-established technology for capturing expert knowledge and emulating decision-making, stemming from early research in artificial intelligence. ES represent one of many approaches in GeoComputation (GC) for solving ill-structured or semi-structured problems, of which there are many examples in the geospatial domain. This chapter provides a historical and technical overview of ES development including when ES are appropriate methods to use and what tools are available for their creation. Spatial ES are then considered in more detail, focussing on their integration with geographic information systems (GIS), and an example is provided which was developed to support the process of planning for land consolidation. Here, the ES is one component of a larger integrated planning support system that generates alternative land redistribution plans under a range of potential scenarios. Using an actual land consolidation project in Cyprus as a case study, the results demonstrate that the ES performs very well in relation to a solution generated manually by land consolidation experts while taking only a fraction of the time required. Yet while ES technology is becoming more widely used to solve geospatial problems as reported in the literature, there is still a long way to go before they can be used in more operational settings such as in local or regional planning offices. Two further requirements are the need for a user-friendly and efficient ES shell for integrating GIS with ES for any spatial problem domain and further development of the ES technology itself in terms of knowledge acquisition, representation and self-learning.