Abstract
The purpose in this paper is to explore the consequences of weight variations in a given set of demand points on the following p-median results: the average and maximum distance between demand and supply, the geographical pattern of supply points, and the amount allocated to every supply site. By using a real-world case study (intraurban location of public libraries), two different estimations of demand associated with the same set of demand and potential supply sites are compared. Results show only small differences in the p-median outputs in spite of rather large discrepancies in the distributions of demand weights. This invites further and systematic research about the robustness of p-median results with respect to variations in the geographical distribution of demand.
Get full access to this article
View all access options for this article.