authors 
Brown, A.G.P. and Coenen, F.P. 
year 
2000 
title 
Spatial reasoning: improving computational efficiency 
source 
Automation in Construction 9 (4) (2000) pp. 361367 
summary 
When spatial data is analysed the result is often very computer intensive: even by the standards of contemporary technologies, the machine power needed is great and the processing times significant. This is particularly so in 3D and 4D scenarios. What we describe here is a technique, which tackles this and associated problems. The technique is founded in the idea of quadtesseral addressing; a technique, which was originally applied to the analysis of atomic structures. It is based on ideas concerning Hierarchical clustering developed in the 1960s and 1970s to improve data access time [G.M. Morton, A computer oriented geodetic database and a new technique on file sequencing, IBM Canada, 1996.], and on atomic isohedral (same shape) tiling strategies developed in the 1970s and 1980s concerned with group theory [B. Grunbaum, G.C. Shephard, Tilings and Patterns, Freeman, New York, 1987.]. The technique was first suggested as a suitable representation for GIS in the early 1980s when the two strands were brought together and a tesseral arithmetic applied [F.C. Holdroyd, The Geometry of Tiling Hierarchies, Ars Combanitoria 16B (1983) 211–244.; S.B.M. Bell, B.M. Diaz, F.C. Holroyd, M.J.J. Jackson, Spatially referenced methods of processing raster and vector data, Image and Vision Computing 1 (4) (1983) 211–220.; Diaz, S.B.M. Bell, Spatial Data Processing Using Tesseral Methods, Natural Environment Research Council, Swindon, 1986.]. Here, we describe how that technique can equally be applied to the analysis of environmental interaction with built forms. The way in which the technique deals with the problems described is first to linearise the threedimensional (3D) space being investigated. Then, the reasoning applied to that space is applied within the same environment as the definition of the problem data. We show, with an illustrative example, how the technique can be applied. The problem then remains of how to visualise the results of the analysis so undertaken. We show how this has been accomplished so that the 3D space and the results are represented in a way which facilitates rapid interpretation of the analysis, which has been carried out. 
series 
journal paper 
more 
http://www.elsevier.com/locate/autcon 
references 
Contenttype: text/plain

last changed 
2003/05/15 19:22 
