Mimari Tasarım İçin Modüler ve Dinamik Bir Evrimsel Algoritma

İyi tariflenmiş problem alanlarından planlama ve tasarım gibi kötü-tarifli alanlara doğru geçtiğimizde karşılaştığımız problemlerin karmaşıklığı problem çözme yaklaşımında niteliksel değişiklikler dayatır. Bunun sonucu olarak dinamik problem çözme stratejileri hesaplamalı tasarım çalışmaları için bir gereklilik olarak açığa çıkar. Bu çalışma İçiçe Evrimsel Algoritma (IEA) adında yeni bir çoklu-objektifli Evrimsel Algoritmayı (EA) dinamik yönlere sahip bir problem çözme aracı olarak sunmaktadır. IEA’nın diğer EA’lardan farkı, kullanılan objektiflerden birinin zindelik ilerlemesi duraklayana kadar süreci çeşitli açılardan yönlendirmesi ve daha sonra yönlendirme işlevini bir diğer objektife devretmesidir. Bu şekilde IEA’nın farklı objektifler için farklı ayar ve operatörler kullanması mümkün olmaktadır. Bu sayede IEA problem tanımını işleyişi boyunca dinamik biçimde uyarlayabilmektedir. IEA özel olarak tasarım problemlerinin çizgisel-olmayan, karmaşık karakterine dönük olarak geliştirildiği için bu makalede IEA’nın kendine has özelliklerini kötü-tanımlı, çoklu-modlu ve çoklu objektifli bir problem olan Mimari Plan Düzenlemesi Problemi üzerinden ve kütüphane binaları özelinde sunuyoruz. İlk olarak, IEA’nın işleyiş karakteristiklerini ve başarımını bir sıralama-tabanlı EA versiyonuyla kıyaslayarak yukarıda anlattığımız yönlendirici objektif yaklaşımının sonuçlarını ortaya koyuyoruz. Ardından, işleyiş karakteristiklerini daha derinlemesine yorumlamak üzere IEA’yı popüler bir çoklu-objektifli EA olan NSGA2 ile kıyaslıyoruz. Son olarak spesifik tasarım alanlarına ait bilgiyi işe koşmanın gerekçelerini ve yollarını tartışarak IEA’nın nasıl kullanımlarının olabileceğini ve nasıl daha öte geliştirilebileceğini tartışıyoruz.

A Modular and Dynamic Evolutionary Algorithm For Architectural Design

As we move away from well-defined problem domains, and get closer to more open-ended domains like planning and design, an increasein the complexity of the problems compel the problem-solving behavior to change in a qualitative sense. Consequently, dynamic problem solving strategies appear as one of the requirements for computational design studies. This paper presents a novel multi-objectiveEvolutionary Algorithm (EA) called the Interleaved EA (IEA) as a problem-solving tool, which incorporates dynamic aspects. It is specific toIEA that one of the objectives leads the evolution until its fitness progression stagnates. As such, IEA enables the use of different settingsand operators for each of its objectives, which would be the same for all objectives in a regular EA. This enables the IEA to dynamicallyadapt its problem setting throughout its progression. We present the specificities of the IEA with an application on a design problem.As the IEA has been developed to assist in design problems, it is examined through the “Architectural Layout Design” problem studiedthrough library buildings, exemplifying an ill-defined, multi-modal, and multi-objective problem. We compare the functioning of the algorithm with regard to, first, a regular rank-based version, for demonstrating the effect of the leading objective approach; secondly, witha popular multi-objective EA (i.e., NSGA2). We discuss how and why IEA can be used and developed further to incorporate domain specificunderstanding for multi-modal and dynamic design problems.

PDF

___

Akin, Ö. (2001) “Variants in Design Cognition”, in Eastman, C., Newstetter, W., and McCracken, M. (eds.), Design Knowing and Learning: Cognition in Design Education, 978- 0-08-043868-9, Elsevier, http://doi.org/10.1016/B978- 008043868-9/50006-1.
Akın, Ö. (2009) “Variants and Invariants of Design Cognition”, in McDonnell, J. and Lloyd, P. (Eds), About Designing: Analysing Design Meetings, CRC Press, pp. 171-192.
Back, T., Fogel, D.B., and Michalewicz, Z. (Eds.) (2000) Evolutionary Computation 2: Advanced Algorithms and Operators, IOP Publishing Ltd, Bristol and Philadelphia.
Buchanan, R. (1992) “Wicked problems in design thinking”, Design Issues, Vol. 8, No. 2, pp. 5-21.
Caldas, L.G. (2003) “Shape generation using pareto genetic algorithms”, CAADRIA 2003.
Caldas, L.G. (2005) “Three-dimensional shape generation of lowenergy architecture solutions using Pareto GA’s”, Proceedings of ECAADE’05, Lisbon, September 21-24, 2005, pp. 647-654.
Caldas, L.G. (2006) “GENE_ARCH: An evolution-based generative design system for sustainable architecture”, I.F.C. Smith (Ed.), EG-ICE 2006, LNAI 4200, pp. 109 – 118, 2006.
Caldas, L.G. (2008) “Generation of energy-efficient architecture solutions applying GENE_ARCH: An evolution-based generative design system”, Advanced Engineering Informatics, Volume 22, Issue 1 (January 2008).
Caldas, L.G. and Norford, L.K. (2002) “A design optimization tool based on a genetic algorithm”, Automation in Construction, 11 (2002) 173 – 184.
Caldas, L.G. and Norford, L.K. (2003) “Genetic Algorithms for Optimization of Building Envelopes and the Design and Control of HVAC systems”, Journal of Solar Energy Engineering, August 2003, Vol. 125.
Caldas, L.G. and Rocha, J. (2001) “A generative design system applied to Siza’s school of architecture at Oporto”, In J.S. Gero, S. Chase and M. Rosenman (Eds) CAADRIA 2001, Key Centre of Design Computing and Cognition, University of Sydney, 2001, pp. 253-264.
Damski, J.C. and Gero, J.S. (1997) “An evolutionary approach to generating constraint-based space layout topologies”, In R. Junge (Ed.), CAADFutures 1997, Kluwer, Dordrecht. pp. 855- 864.
Deb, K., Pratap, A., Agarwal, S., and Meyarivan, T. (2002) “A fast and elitist multiobjective genetic algorithm: NSGA-II”, IEEE Transactions on Evolutionary Computation, 6(2), pp. 182– 197.
Gero, J.S., Louis S., and Kundu, S. (1994) “Evolutionary learning of novel grammars for design improvement”, AI EDAM, 8(2):83-94.
Gero, J.S. and Schnier, T. (1995) “Evolving representations of design cases and their use in creative design”, Third International Conference on Computational Models of Creative Design.
Gero, J.S. and Kazakov, V.A. (1998) “Evolving design genes in space layout planning problems”, Artificial Intelligence in Engineering, 12 (1998) 163-176.
Janssen, P. H. (2009) “An evolutionary system for design exploration”, in Proceedings of the International Conference on Computer Aided Architectural Design Futures, Montréal, Canada17th-19th June, pp. 260–272.
Lawson, B. (2004) What Designers Know, Elsevier / Architectural Press, Amsterdam.
Machairas, V., Tsangrassoulis, A., & Axarli, K. (2014) “Algorithms for optimization of building design: A review”, Renewable and Sustainable Energy Reviews, 31, pp. 101–112, http://doi. org/10.1016/j.rser.2013.11.036
Michalewicz, Z. and Schmidt, M. (2007) “Parameter Control in Practice”, in Lobo, F.J., Lima, C.F., and Michalewicz, Z. (Eds.), Parameter Setting in Evolutionary Algorithms, Series: Studies in Computational Intelligence, Vol. 54, XII, Springer-Verlag, Berlin, Heidelberg, 2007, pp. 277-294.
Raphael, B. (2014) “Multi-Criteria Decision Making for the Design of Building Facade”, in Computing in Civil and Building Engineering, ASCE, pp. 1650–1658.
Rittel, H. W. J. and Webber, M. M. (1973) “Dilemmas in a General Theory of Planning”, Policy Sciences, 4, pp. 155-169.
Rodrigues, E., Gaspar, A.R., and Gomes, Á. (2013a) “An approach to the multi-level space allocation problem in architecture using a hybrid evolutionary technique”, Automation in Construction, 35, pp. 482–498, doi:10.1016/j.autcon.2013.06.005
Rodrigues, E., Gaspar, A.R., and Gomes, Á., (2013b) “An evolutionary strategy enhanced with a local search technique for the space allocation problem in architecture, Part 1: Methodology”, Computer-Aided Design, 45, pp. 887–897, doi:10.1016/j.cad.2013.01.001
Rodrigues, E., Gaspar, A.R., and Gomes, Á. (2013c) “An evolutionary strategy enhanced with a local search technique for the space allocation problem in architecture, Part 2: Validation and performance tests”, Computer-Aided Design, 45, pp. 898–910, doi:10.1016/j.cad.2013.01.003
Rosenman, M.A. (1997) “The generation of form using an evolutionary approach”, [online] http://arch.usyd.edu.au (Accessed: September 2009).
Rosenman, M.A. and Saunders, R. (2003) “Self-regulatory hierar chical coevolution”, AI-EDAM, 2003, 17, 273 – 285.
Russell, S. J. and Norvig, P., (2010) Artificial Intelligence: A Modern Approach, Prentice Hall (third edition).
Simon, H. A. (1973) “The Structure of Ill Structured Problems”, Artificial Intelligence, 4 (3): pp. 181–201.
Sönmez, N.O., Erdem, A. (2014) “Design games: A conceptual framework for dynamic evolutionary design”, A | Z ITU Journal of the Faculty of Architecture, Vol. 11, No. 1, 03/2014.
Sönmez, N.O. (2015a) Evolutionary Design Assistants for Architecture, ABE, Architecture and the Built Environment, 5(3), pp. 1–284, http://doi.org/10.7480/abe.2015.3
Sönmez, N.O. (2015b) “Architectural Layout Evolution through Similarity-Based Evaluation”, International Journal of Architectural Computing (IJAC), Vol. 13, No. (3-4), 10/2015, pp. 271–298, http://doi.org/10.1260/1478-0771.13.3-4.271
Turrin, M., von Buelow, P., and Stouffs, R. (2011), “Design explorations of performance driven geometry in architectural design using parametric modeling and genetic algorithms”, Advanced Engineering Informatics, doi:10.1016/j. aei.2011.07.009.
Turrin, M., von Buelow, P., Kilian, A., and Stouffs, R. (2011), “Performative skins for passive climatic comfort: A parametric design process”, Automation in Construction, doi:10.1016/j. autcon.2011.08.001.
Yuan, B. and Gallagher, M. (2007) “Combining Meta-EAs and Racing for Difficult EA Parameter Tuning Tasks”, in Lobo, F. J., Lima, C. F., and Michalewicz, Z. (Eds.), Parameter Setting in Evolutionary Algorithms, Series: Studies in Computational Intelligence, Vol. 54, 2007, XII, Springer-Verlag, Berlin, Heidelberg, pp. 121-142.
Zhang, T., Georgiopoulos, M., and Anagnostopoulos, G.C. (2013) “S-Race: a multi-objective racing algorithm”, in Proceedings of the Fifteenth Annual Conference on Genetic and Evolutionary Computation Conference, ACM, pp. 1565–1572.
Zitzler, E., Laumanns, M., and Bleuler, S. (2004) “A tutorial on evolutionary multiobjective optimization”, in Gandibleux, X., Sevaux, M., Sörensen, K., and T’kindt, V. (Eds.), Metaheuristics for Multiobjective Optimisation, Lecture Notes in Economics and Mathematical Systems, Volume 535, 2004, pp. 3-37.