CAAD.HBT.ETHZ.CH . MAS0607stu . Papers

ETH Zuerich - Startseite
Professur für CAAD

 


caad d-arch

MAS ETH ARCH/CAAD - 2006/07 | Student Pages
Master of Advanced Studies in Architecture, Specialization in Computer Aided Architectural Design | 065-0005/6
Supervision: Prof. Dr. Ludger Hovestadt, Philipp Schaerer
Chair of CAAD, ETH Zurich






LEE SEONG KI | leese@student.ethz.ch | CV | Papers

An Ontology-based Approach in Design Process Modeling

Lee, Seong Ki Kim, Sung ah

in Proceedings of Annual Conference in Architectural Institute of Korea, Vol. 25, No. 1, Seoul, Korea, pp. 437-440, 2005. (in Korean)

Abstract

The Purpose of design process modeling is to understand the characteristics of the design efficiently through analyses of design process models. This research is also to classify and recognize design activities, to build the taxonomy of design activities and the interactions between activities, to conceptualize the characteristics of design representations, when and how design representation being used within design activities, and to figure out the process until the final design is realized, etc. Finally, from the perspective of information communication technology development, the authors propose a visualization method of the proposed design process model. This research tries to construct design process ontology by applying the ontology methodology to the design process modeling, and its application, To do this, the authors implement the design process ontology using OWL (Web Ontology Language). Also, some practical applications are illustrated through visualization of design process ontology.

Keyword : Design Process, Design Model, Design Activity, Design Representation, Ontology, OWL, Protege, X3D


An Automatic Classification of Architectural Cases using Feature-based Similarity Measurement

Lee, Seong Ki

Master Thesis

Department of Architecture, Graduate School, Sungkyunkwan University,

Suwon, Korea, August, 2006

Abstract

Using computers to support architectural design process is a key issue in Computer-Aided Architectural Design (CAAD) domain. One of the efforts is to use computers for Case-Based Design(CBD) based on Case-Based Reasoning (CBR) paradigm. CBD systems that are developed until now are adapting keyword-based cases indexing, searching, and retrieving method. That means that if user want to find case while designing, he/she must abstract his/her ambiguous or ill-defined design idea into some keywords that are used in the database of CBD system. And if CBD system give user cases that user requested, he/she should recursively browse until he/she find a case. It is the objective of this work to propose a methodology of architectural cases classification based on shape similarity that can be implemented to support users while designing using CBD systems. To achieve this goal, we firstly review various theories which are discussed within CBD, for example, analogical reasoning as a way of design problem solving, typology as a method of finding design alternatives, and so on. General descriptions of CBD systems are followed.

Similarity issues including form similarity, spatial similarity, spatial topology similarity, visibility similarity in architecture are discussed. Theories on feature recognition are discussed as a way of implementing feature-based similarity measurement for cases. And various feature-based similarity assessments using feature recognition techniques are followed as well. Case-studies are used to apply feature recognition techniques to architectural cases. For this, the author proposes and exploits the notion of three kinds of architectural 3D model (outline contour type, space volume type, wall type). And with these models, the author implements an automatic classification of three types of architectural 3D models using Form Feature Recognition (FFD) and a feature recognition tool 3DSearchIT. The author think that the use of feature recognition techniques in CBD systems will bring an efficiency to the design process and will augment the capabilities of human designers. Afterward this methodology can be applied to an efficient architectural cases management, design education.

Keywords : Case-Based Design, Similarity, Feature Recognition

-- LeeSeongKi - 01 Nov 2006

Revision r2 - 02 Nov 2006 - 07:29 - LeeSeongKi
Parents: WebHome > LeeMainPage > CurriculaVitae

This website has been archived and is no longer maintained.