A Computational Framework for Analyzing Spatial Behavior in Buildings

Abstract

This paper proposes a computational framework specifically formulated for observing and describing occupant behavior in architectural analysis and design. The framework comprises two main elements. First, it includes an architecture-oriented conceptual delimitation of occupant behavior centered on the spatial dimension of human actions. Second, the framework provides a comprehensive set of computational metrics to describe the spatial behavior of building occupants particularly designed for use with physical sensing technologies and computational data analysis techniques. An experimental case study demonstrates the applicability of the framework in the context of analyzing a multifunctional building space.