From Fractal Urban Pattern Analysis to Fractal Urban Planning Concepts - Computational Approaches for Urban Environments - page 42

￿ Level 3 (places with daily demand or used several times a week such as bakeries,

cigarette/newspaper outlets, schools, butcher's/delicatessens, grocery stores). It
is assumed here that it is the distances within walking distance from home. The
maximum range considered is set to 400 m (Wiel et al. 1997 ).

Since shops and services are not localized in the same place, retail clusters
are introduced including all shops or services lying within 200 m of one another.
Distances less than 200 m improve accessibility, whereas greater distances reduce it.
Accessibility from a given site to a cluster is evaluated by combining the frequency
of use of the different services/shops and the distance on the network. Accessibility
evaluation, based on fuzzy logic, takes into account all existing clusters and for
each one the number of services/shops and the diversity of the offer. By the
underlying logic, two clusters lying at the same distance are better than one; one
big, readily accessible cluster is better than two less accessible ones; if a cluster
is very easily accessible, the existence of another one with low accessibility is not
really important. For higher level facilities, access by public transportation network
is included in the evaluation. It takes into account the timetables of the public
transportation networks such as buses or suburban trains as well as access to stations.
The details of the evaluation model are presented in Tannier et al. ( 2010 , 2012 ).
The distances accepted for acceding to the clusters of different hierarchical levels
may depend on the local context, in particular for higher level facilities, since the
distances to cities offering these types of amenities are not the same for densely
populated metropolitan areas as for low density areas. For evaluating accessibility to
the facilities for daily needs (level 3), walking or cycling distances are used (Tannier
et al. 2006 ).
For green leisure areas, similar categories are introduced. However, two types of
accessibility are considered, the visual access to open landscape for the buildings
and accessibility by the road network. Visual access is evaluated by considering the
number of buildings lying on the boundary of the urbanized areas. To identify urban
borders, consistent morphological envelopes are extracted by a multiscale approach
(Tannier et al. 2011 ). Accessibility to green and natural areas via the transportation
network is measured by defining the specific criteria of accessibility to each type of
space (Tannier et al. 2006 ; Czerkauer-Yamu and Frankhauser 2013 ).
The evaluations are realized for MUP-city for each of the grid squares at each
iteration, usually by using the network distances. For Fractalopolis, the evaluations
are recalculated when changing the position of a square. The evaluation result is
coded by a color which is assigned to the square.

2.7 Conclusion and Outlook

We presented a couple of results about how fractal geometry allows better under-
standing spatial organization of urban fabrics. We saw that fractal approach suits
indeed rather well for describing spatial organization of urban fabrics even if their
shapes seem to be irregular. That such patterns are perceived as irregular is rather