Data flow and processing in the computational fractal analysis method

One of the few quantitative methods available for the consistent analysis of architectural form is the ‘box-counting’ approach to determining the approximate fractal dimension of a plan or elevation. In its computational form this method has been used to analyze the plans and facades of a wide range of buildings. The data points produced are synthesized by the software into a series of fractal dimension (D) values that are in turn compiled in various ways to produce a series of composite results describing a complete building. Once this process is complete the data may be coded with additional information producing a set of mathematical results that describe the form of a building. This paper offers the first complete description of this important analytical process from the point of view of information flow, algorithmic operations, review options and data magnitude. No previous paper has detailed the full scope of the data used in the computational method, or the way in which various stages produce different types of outcomes. The purpose of this paper is to elucidate the way in which this particular computational method, drawing its inspiration from the complexity in natural systems, may be used to process different types of information and produce various forms of quantitative data to support architectural design and analysis.