ABSTRACT
A generalization process will be involved, whenever the database has not been purposely structured for the production of such maps, as is the usual case in GIS. The firm condition, that most GIS data sooner or later must meet also various mapping requirements, forces the production units to strive for computer-based solutions for generalization. Aggregation of individual data within larger reference units is another case of model-oriented generalization. The system user selects the object and reference data, including an initial minimal symbolization, analyses the combination visually on the display in view of the following generalization process. In parallel, and as a complement to this visual inspection, propose a series of modules for a statistical analysis of the selected data, again with respect to the generalization strategy to follow. The production of always up-to-date multiple scale versions of maps on the bases of scale-independent databases, one single maintenance procedure, would require a huge, well-conceived and consistent system of efficient generalization facilities.
