We present a method for selecting good locations, each of which is close to desirable facilities such as stations, warehouses, promising customers' house, etc. and is far from undesirable facilities such as competitors' shops, noise sources, etc. Skyline query, which selects non-dominated objects, is a well known method for selecting small number of desirable objects. We use the idea of skyline queries to select good locations. However, locations are two dimensional data, while objects in the problem of conventional skyline queries are zero dimensional data. Comparison of two dimensional data is much more complicated than that of zero dimensional data. In this paper, we solve the problem of skyline query for two dimensional data, i.e., areas in a map. Experimental evaluations of the proposed method shows that our approach is able to find reasonable number of desirable skyline areas and can help users to find good locations.
