A computational model for accommodating spatial uncertainty: Predicting inhabitation patterns in open-planned spaces

Abstract

In the past, a range of computational models have been developed for analysing the social implications of spatial patterns and types. While such models are typically focussed on macro-patterns, often in cellular or linearly-organised spaces, few models exist for predicting where people will cluster within complex environments. One reason for this relates to the inherent uncertainty associated with spatial attributes and consequently of human spatial behaviours. The present paper draws on the concept of fuzzy spatial objects to develop an approach to handle such uncertainty in architecture. Focussing on large, open plan spaces, where the configuration of space does not define strict patterns of usage, the paper proposes a computational model for predicting patterns of spatial inhabitation. This new model relies on the theory of fuzzy sets to propose the existence of a “fuzzy architectural spatial object, (FASO)” which is comprised of spatial units with degrees of membership that reflect the possibility of a person being present in a sub-space or involved in a sub-function within a larger space. This model calculates and visualises the FASOs using a fuzzy inference engine and represents the space as distributed possibilities of presence according to the given data. After describing the model the paper demonstrates its application in the prediction of patterns of usage within a major exhibition space, and then presents a check of the efficacy of this prediction against the actual inhabitation of the space.