%A N. Raford
%D 2004
%K Intelligibility, fractured spatial systems, American cities, urban
regeneration, pedestrian movement modelling
%I UCL (University College London)
%X This thesis explores the influence of configuration on movement in fragmented, low intelligibility
spatial systems. Traditional space syntax theory holds that correlation between space and
movement breaks down in such situations, resulting in lower predictability and / or apparently
chaotic behaviour.
This thesis uses the case study of Boston, Massachusetts to test three hypotheses; 1) that space is
indeed influential on fragmented systems, 2) that other variables are less important than space in
predicting movement, and 3) that space – movement correlations are distributed non-uniformly
between areas and user groups. These are tested through the use of multiple regression analysis
and a new technique named “correlation contour mapping,” which outlines the boundaries of
predictability within complex spatial systems.
It will be shown that space plays a significant role in pedestrian movement in Boston, but that it
correlates non-uniformly with different areas and users. An understanding of sub-area definitions
and the effect of overlapping patterns of correlation is therefore necessary in order to fully
comprehend the effects of fragmentary configuration. It is suggested that the use of correlation
contouring may achieve this and might be a useful tool for exploring the interaction of different
user groups in urban space.
This understanding is then used to construct a pedestrian movement model to forecast the effects
of a large-scale urban regeneration project in Boston, known as the “Big Dig”. The implications
of these findings for other fragmented cities are also discussed in a context of successful
urbanisation, based on the work of Lewis Mumford, Jane Jacobs, Kevin Lynch, and Bill Hillier.
%T Movement economies in fractured urban systems:
the case of Boston, Massachusetts
%L discovery1432