Optimizing the Shopping Center - Mix A GIS based Analysis

Bid mix and shop arrangement have been seen from three theoretical perspectives. The potential tenants expect a bid mix, which enables them to take advantage of synergetic effects (shared/suspicious business) (Nelson, 1958) and which minimizes the inter competition ñ respectively makes it tolerable. Similarly, the consumer wants to minimize his transaction costs in the SC ñ as long as it is a matter of a rationally organized visit. If the consumer prefers the adventure, the amenity value will have a greater impact on the shop arrangement. But in the end it is the SC-operator who has to conciliate these demands. In this context he will be anxious ñ according to the theoretic approaches of Carter / Vandell (2005) and Brown (1991) ñ to arrange the diverse shops in a way, so that the visitor frequency is optimized in all parts of the mall. In line with this optimization two main questions appear for the SC-management: Which shops are necessary for an optimal sector mix at all? How should these shops be arranged within the mall in an (frequency) optimal way? It is the second question which will be focused in this article using a GIS-application. Regarding the current use of GIS in real estate research (Culley, 2010; Segerer, 2011), which focused the analysis of locations, a change of perspective takes place: GIS is not used anymore ìaround the real estateî but within. This approach obeys Shun-Te You (2010), who uses GIS to optimize the tenantsí distri-bution within a SC concerning an efficient use of space and visitor control. Basis for the analysis in this paper is a GIS-based model of a shopping center. Within this model not only the geometric principles of different structural level, passers infrastructure and entrances, but also a network of potential paths of consumers through multiple levels are mapped. Specifically collected data on shop mix, the dealersí satisfaction with their own site and the route choice of consumers are the basis for creating inventories and conducting further analysis. These include inter alia the comparison of the self-perception of dealers, the present customer current density and their conversion into sales. Important methods in the process are the application of kernel density estimators for modeling the flow of costumers and the assignment of clusters in reference to coupling. This method makes far-reaching statements on local coupling potentials and the effects of shop-clustering or ñdispersion on diverse target groups. In this context GIS allows not only a detailed analysis of the data, but also an adequate visualization of complex issues. In order to integrate GIS-Systems into the workflow of center managers, operators and investors, the automating of the analysis functions with the ArcGIS-integrated Model-Builder is demonstrated. So the widespread aversion (too expensive/too time-consuming) cannot be confirmed. Mapping models, which are once created and easy to modify, permit a continuous monitoring with flexible process models. Moreover adequate tools can be developed by external providers and accessed via internet on demand.