Assessing urban vulnerability to natural hazards such as earthquakes can be regarded as an ill-structured problem (i.e. a problem for which there is no unique, identifiable, objectively optimal solution). A review of the literature indicates a number of contrasting definitions of what vulnerability means, as well as numerous conflicting perspectives on what should or should not be included within the broad assessment of vulnerability in cities. This paper reports on the findings from a project in which a GIS methodology has been developed to assess urban vulnerability through a spatial analytical procedure. First, we highlight the deficiencies of current GIS approaches to urban vulnerability analysis and discuss the ill-structured nature of the vulnerability problem. We then propose a working definition for vulnerability assessment in which vulnerability is thought of as a spatial decision problem under the conditions of uncertainty. Next, we present a methodology to incorporate this definition into a GIS framework that combines elements from the techniques of spatial multicriteria analysis and fuzzy logic. The application of this methodology is then illustrated with a case study from Los Angeles County. The results suggest that the proposed methodology may provide a new approach for analyzing vulnerability that can add to our understanding of human/hazards interaction.