Abstract
In this paper a monitoring and locational decision support system for retail activity is presented. This system uses point-based customer densities as a set of data instead of area-based densities which have been used in most of the previous studies. Bicubic spline gridding and interpolation methods are applied to convert sample data to grid data and to convert grid data to grid data, respectively. Retail trade areas are represented as a three-dimensional surface in colour graphics to monitor the small changes. This system can also search for an optimal location for an intended store and assess its spatial impacts upon the existing stores by means of a spatial interaction model with bicubic splines.