Resolving Graphic Conflict in Scale Reduced Maps : Refining the Simulated Annealing Technique

Introduction In previous work, the authors show the potential for iterative improvement techniques to be used as part of an automated map generalisation solution ([1], [2]). In particular, they present a simulated annealing algorithm that controls operations of displacement, deletion, reduction and enlargement of multiple map objects in order to resolve graphic conflict arising as a consequence of scale reduction. The algorithm adopts a trial position approach in which each of n discrete polygonal objects is assigned k candidate trial positions that represent the original, displaced, deleted, reduced and enlarged states of the object. This gives rise to a possible k distinct map configurations; the expectation is that some of these configurations will contain reduced levels of conflict. Each configuration has an associated overall cost, which can be computed. This overall cost combines both conflict cost (i.e. the extent to which acceptable minimum clearances between map objects are violated) and modification cost (i.e. the extent to which the map has been altered). Finding the configuration with least overall cost by means of an exhaustive search is not practical for realistic values of n and k. However, it has been shown that near optimum solutions can be found by using simulated annealing to direct a search through a subset of the configurations; thus effective resolution of graphical conflict can be achieved.