Universidad de Granada Digibug

Repositorio Institucional de la Universidad de Granada >
1.-Investigación >
Departamentos, Grupos de Investigación e Institutos >
Departamento de Urbanística y Ordenación del Territorio >
DUOT - Comunicación Congresos, Conferencias, ... >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10481/18384

Title: Generation of downtown planning-ordinances using self organizing maps. Lecture.
Authors: Abarca-Álvarez, Francisco Javier
Fernández Avidad, Ángel
Issue Date: 2010
Abstract: Facing the stability of typical urban ordinances, which all too frequently become outdated at the moment of their formulation, a new method of work is currently being explored. This innovation is based on feedback from the normative body via the dynamic introduction of preexistences and transformations generated by the same framework as the ordinance. The Ordinance, as a link between the urban project and architecture, should compromise with the purely urban concepts as well as the purely architectural, establishing relationship or network pattern between them. It will be that definition of relational or network pattern that we will call Network-Ordinance. And its adaptation or approximation to the same ordinance, by means of the architectural-urban proposal, will be the normative framework generated from the global, in contrast to the particular and accidental. It will be, therefore, the search for coherencies formalized in pattern that come together and display the most representative ideals and values of the selected area. By means of the objective numeric representation of the values considered useful or representative, as well as those whose interests are unknown, and their application through an artificial neuronal network, with unsupervised and competitive learning of the same type as the SOM (Self-Organizing Map) and, more concretely, Kohonen’s network, the underlying structure of the same data will be discovered and represented in a comprehensible manner. The results obtained are easily interpretable, permitting the recognition of the grouping, in the form of pattern, of the architectural objects represented, allowing the verification of the integration (or not) of a new building or object into those pattern, which conform to the Network-Ordinance and what we have come to call Ecotype. The new objects allowed by this rule will then form part of the same ordinance, becoming integrated in the body of a new neuronal network, and thus acquiring feedback from the ordinance. As a verification, it is proposed that a Network-Ordinance is created for a sector of the historical center of a city in the province of Granada, in order to compare the obtained results with the proposed Ordinance by the city planning.
Sponsorship: Universidad de Granada. Departamento de Urbanística y Ordenación del Territorio.
Description: 10th International Conference on Design and Decision Support Systems in Architecture and Urban Planning. Eindhoven (The Netherlands).
Keywords: Ordinance
Neural network
Self organizing map
URI: http://hdl.handle.net/10481/18384
Rights : Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License
Appears in Collections:RNM357 - Comunicaciones Congresos, Conferencias, ...
DUOT - Comunicación Congresos, Conferencias, ...

Files in This Item:

File Description SizeFormat
Lecture Abarca.pdfLecture115.43 kBAdobe PDFView/Open
Session DDSS Abarca.pdfSlides2.12 MBAdobe PDFView/Open
Recommend this item

This item is licensed under a Creative Commons License
Creative Commons

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.


Valid XHTML 1.0! OpenAire compliant DSpace Software Copyright © 2002-2007 MIT and Hewlett-Packard - Feedback

© Universidad de Granada