CumInCAD is a Cumulative Index about publications in Computer Aided Architectural Design
supported by the sibling associations ACADIA, CAADRIA, eCAADe, SIGraDi, ASCAAD and CAAD futures

PDF papers

Hits 1 to 20 of 41

_id ecaade2015_215
id ecaade2015_215
authors Balakrishnan, Bimal and Oprean, Danielle
year 2015
title Communication, Coordination and Collaboration: Media affordances and Team Performance in a Collaborative Design Environment
source Martens, B, Wurzer, G, Grasl T, Lorenz, WE and Schaffranek, R (eds.), Real Time - Proceedings of the 33rd eCAADe Conference - Volume 2, Vienna University of Technology, Vienna, Austria, 16-18 September 2015, pp. 225-232
wos WOS:000372316000027
summary Advances in digital media are encouraging designers to adopt digital tools during early stages of design ideation as well as to facilitate collaboration in design teams. Collaborative environments for design teams should take into consideration both the multimodal nature of design representation as well as the complexity of team cognition. Collaborative tools that take a “black-box” approach often limit affordances for design ideation and collaboration. We describe here a collaborative environment that we put together using a kit-of-parts approach and underlying theoretical considerations. We also describe systematic usability evaluation of the collaborative environment by constraining select media affordances and qualitatively examining the impact on a team's design process. Preliminary findings were used to improve the environment and lay the groundwork for developing tele-collaborative environments.
series eCAADe
last changed 2016/05/16 09:08

_id a6d8
authors Baletic, Bojan
year 1992
title Information Codes of Mutant Forms
source CAAD Instruction: The New Teaching of an Architect? [eCAADe Conference Proceedings] Barcelona (Spain) 12-14 November 1992, pp. 173-186
summary If we assume that the statements from this quote are true, than we have to ask ourselves the question: "Should we teach architecture as we do?" This paper describes our experience in developing a knowledge base using a neural network system to serve as a "intelligent assistant" to students and other practicing architects in the conceptual phase of their work on housing design. Our approach concentrated on rising the awareness of the designer about the problem, not by building rules to guide him to a solution, but by questioning the categories and typologies by which he classifies and understands a problem. This we achieve through examples containing mutant forms, imperfect rules, gray zones between black and white, that carry the seeds of new solutions.
series eCAADe
last changed 2003/11/21 14:16

_id ecaade2018_405
id ecaade2018_405
authors Belém, Catarina and Leit?o, António
year 2018
title From Design to Optimized Design - An algorithmic-based approach
source Kepczynska-Walczak, A, Bialkowski, S (eds.), Computing for a better tomorrow - Proceedings of the 36th eCAADe Conference - Volume 2, Lodz University of Technology, Lodz, Poland, 19-21 September 2018, pp. 549-558
summary Stringent requirements of efficiency and sustainability lead to the demand for buildings that have good performance regarding different criteria, such as cost, lighting, thermal, and structural, among others. Optimization can be used to ensure that such requirements are met. In order to optimize a design, it is necessary to generate different variations of the design, and to evaluate each variation regarding the intended criteria. Currently available design and evaluation tools often demand manual and time-consuming interventions, thus limiting design variations, and causing architects to completely avoid optimization or to postpone it to later stages of the design, when its benefits are diminished. To address these limitations, we propose Algorithmic Optimization, an algorithmic-based approach that combines an algorithmic description of building designs with automated simulation processes and with optimization processes. We test our approach on a daylighting optimization case study and we benchmark different optimization methods. Our results show that the proposed workflow allows to exclude manual interventions from the optimization process, thus enabling its automation. Moreover, the proposed workflow is able to support the architect in the choice of the optimization method, as it enables him to easily switch between different optimization methods.
keywords Algorithmic Design; Algorithmic Analysis; Algorithmic Optimization; Lighting optimization; Black-Box optimization
series eCAADe
last changed 2018/08/22 13:38

_id acadia17_146
id acadia17_146
authors Black, Conor; Forwood, Ed
year 2017
title Game Engine Computation for Serious Engineering: Visualisation and Analysis of Building Facade Movements as a Consequence of Loads on the Primary Structure
source ACADIA 2017: DISCIPLINES & DISRUPTION [Proceedings of the 37th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) ISBN 978-0-692-96506-1] Cambridge, MA 2-4 November, 2017), pp. 146-153
summary This paper demonstrates the innovative use of game engines as a tool in the analysis and communication of complex structural engineering. It specifically looks at the relationship between a building’s primary structure and its façade. The analysis and visualisations, scripted using the Game Engine Unity3D, focuses on visualising the implications of movements from the primary structure [under various load cases] on the façade. This paper describes the novel process by which Unity3D is utilised to create an applet which imports displacements from structural software and post-processes the data to visualise the complex effect on façade panels according to its support conditions. It demonstrates that visualising facade movements in real-time, as opposed to current, static report-based descriptions, provide access for the comprehension of more complex building systems. This therefore has the possibility to reduce safety factors applied to facade movement joints.
keywords design methods; information processing; game engines; fabrication; simulation & optimization
series ACADIA
last changed 2017/10/17 09:12

_id acadia05_094
id acadia05_094
authors Clayton, Mark J.
year 2005
title How I Stopped Worrying and Learned to Love AutoCAD
source Smart Architecture: Integration of Digital and Building Technologies [Proceedings of the 2005 Annual Conference of the Association for Computer Aided Design In Architecture / ISBN 0-9772832-0-8] Savannah (Georgia) 13-16 October 2005, pp. 94-103
summary The history of computing is expressed through AutoCAD as an accretion of ideas and inventions, each of which was a breakthrough in its time. Learning to use AutoCAD, or any CAD system, is augmented by an understanding of the historical context of its development. In contrast to a “deconstructivist” criticism of AutoCAD that avoids all historical context, this paper discusses the user interface of AutoCAD placed in its historical context by combining facts of history with personal reminiscences. The paper answers mysteries about AutoCAD such as “Why a black screen?”, “Why LISP?”, “Why a command line?”, “Why layers, pens and line types?”, and “Why 2D?” An understanding of context and history is a starting point for understanding, mastering, and improving software.
series ACADIA
last changed 2005/10/25 16:52

_id cf2011_p051
id cf2011_p051
authors Cote, Pierre; Mohamed-Ahmed Ashraf, Tremblay Sebastien
year 2011
title A Quantitative Method to Compare the Impact of Design Mediums on the Architectural Ideation Process.
source Computer Aided Architectural Design Futures 2011 [Proceedings of the 14th International Conference on Computer Aided Architectural Design Futures / ISBN 9782874561429] Liege (Belgium) 4-8 July 2011, pp. 539-556.
summary If we compare the architectural design process to a black box system, we can assume that we now know quite well both inputs and outputs of the system. Indeed, everything about the early project either feasibility studies, programming, context integration, site analysis (urban, rural or natural), as well as the integration of participants in a collaborative process can all be considered to initiate and sustain the architectural design and ideation process. Similarly, outputs from that process are also, and to some extent, well known and identifiable. We are referring here, among others, to the project representations or even to the concrete building construction and its post-evaluation. But what about the black box itself that produces the ideation. This is the question that attempts to answer the research. Currently, very few research works linger to identify how the human brain accomplishes those tasks; how to identify the cognitive functions that are playing this role; to what extent they operate and complement each other, and among other things, whether there possibly a chain of causality between these functions. Therefore, this study proposes to define a model that reflects the activity of the black box based on the cognitive activity of the human brain. From an extensive literature review, two cognitive functions have been identified and are investigated to account for some of the complex cognitive activity that occurs during a design process, namely the mental workload and mental imagery. These two variables are measured quantitatively in the context of real design task. Essentially, the mental load is measured using a Bakan's test and the mental imagery with eyes tracking. The statistical software G-Power was used to identify the necessary subject number to obtain for significant variance and correlation result analysis. Thus, in the context of an exploratory research, to ensure effective sample of 0.25 and a statistical power of 0.80, 32 participants are needed. All these participants are students from 3rd, 4th or 5th grade in architecture. They are also very familiar with the architectural design process and the design mediums used, i.e., analog model, freehand drawing and CAD software, SketchUp. In three experimental sessions, participants were asked to design three different projects, namely, a bus shelter, a recycling station and a public toilet. These projects were selected and defined for their complexity similarity, taking into account the available time of 22 minutes, using all three mediums of design, and this in a randomly manner to avoid the order effect. To analyze the two cognitive functions (mental load and mental imagery), two instruments are used. Mental imagery is measured using eye movement tracking with monitoring and quantitative analysis of scan paths and the resulting number and duration of participant eye fixations (Johansson et al, 2005). The mental workload is measured using the performance of a modality hearing secondary task inspired by Bakan'sworks (Bakan et al.; 1963). Each of these three experimental sessions, lasting 90 minutes, was composed of two phases: 1. After calibrating the glasses for eye movement, the subject had to exercise freely for 3 minutes while wearing the glasses and headphones (Bakan task) to get use to the wearing hardware. Then, after reading the guidelines and criteria for the design project (± 5 minutes), he had 22 minutes to execute the design task on a drawing table allowing an upright posture. Once the task is completed, the subject had to take the NASA TLX Test, on the assessment of mental load (± 5 minutes) and a written post-experimental questionnaire on his impressions of the experiment (± 10 minutes). 2. After a break of 5-10 minutes, the participant answered a psychometric test, which is different for each session. These tests (± 20 minutes) are administered in the same order to each participant. Thus, in the first experimental session, the subject had to take the psychometric test from Ekstrom et al. (1978), on spatial performance (Factor-Referenced Cognitive Tests Kit). During the second session, the cognitive style is evaluated using Oltman's test (1971). Finally, in the third and final session, participant creativity is evaluated using Delis-Kaplan test (D-KEFS), Delis et al. (2001). Thus, this study will present the first results of quantitative measures to establish and validate the proposed model. Furthermore, the paper will also discuss the relevance of the proposed approach, considering that currently teaching of ideation in ours schools of architecture in North America is essentially done in a holistic manner through the architectural project.
keywords design, ideation process, mental workload, mental imagery, quantitative mesure
series CAAD Futures
last changed 2012/02/11 18:21

_id 04c9
authors Crandall, N.F. and Wallace , M.J. (ed.)
year 1998
title Work & Rewards in The Virtual Workplace
source Amacom
summary By now, telecommuting is a well-defined word in the corporate U.S. But how about frontline workplace? Or cyberlink workplace? Consultants Crandall and Wallace make convincing arguments about the efficacies of virtual work, and they outline detailed processes and qualifications for any organization contemplating such a move. In a very logical, almost scholarly, fashion, they define terms, explain implementation, demolish perceived and real obstacles, and prove their points via a few case histories. Yet this is not a cut-and-dried book, for the excitement of dramatic changes to our collective workplaces is captured in the descriptions. Chiat/Day assigns its employees a cell phone and a laptop, period. And at Ross Operating Valve, customers actually lead the creative design process. Job satisfaction? You bet. And a much more productive group of employees. Most important for companies interested in these virtual ideas will be the economics chapter, describing in black and white (and sometimes red) the costs involved.
series other
last changed 2003/04/23 13:14

_id 2f1a
authors Dabney, M.K., Wright, J.C. and Sanders, D.H.
year 1999
title Virtual Reality and the Future of Publishing Archaeological Excavations: the multimedia publication of the prehistoric settlement on Tsoungiza at Ancient Nemea
source New York: The Metropolitan Museum of Art
summary The Nemea Valley Archaeological Project is a study of settlement and land use in a regional valley system in Greece extending from the Upper Paleolithic until the present. Active field research was conducted by four teams between 1981 and 1990. The first component was a regional archaeological survey. Second, and closely related to the first, was a social anthropological study of modern settlement and land use. Next was a team assigned to excavate the succession of prehistoric settlements of Ancient Nemea on Tsoungiza. Last, historical ecologists, a palynologist, and a geologist formed the environmental component of the research. As a result of advances in electronic publishing, plans for the final publication of the Nemea Valley Archaeological Project have evolved. Complete publication of the excavation of the prehistoric settlements of Ancient Nemea on Tsoungiza will appear in an interactive multimedia format on CD/DVD in Fall 2000. This project is planned to be the first electronic publication of the American School of Classical Studies at Athens. We have chosen to publish in electronic format because it will meet the needs and interests of a wider audience, including avocational archaeologists, advanced high school and college students, graduate students, and professional archaeologists. The multimedia format on CD/DVD will permit the inclusion of text, databases, color and black-and-white images, two and three-dimensional graphics, and videos. This publication is being developed in cooperation with Learning Sites, Inc., which specializes in interactive three-dimensional reconstructions of ancient worlds The Nemea Valley Archaeological Project is particularly well prepared for the shift towards electronic publishing because the project's field records were designed for and entered in computer databases from the inception of the project. Attention to recording precise locational information for all excavated objects enables us to place reconstructions of objects in their reconstructed architectural settings. Three-dimensional images of architectural remains and associated features will appear both as excavated and as reconstructed. Viewers will be able to navigate these images through the use of virtual reality. Viewers will also be able to reference all original drawings, photographs, and descriptions of the reconstructed architecture and objects. In this way a large audience will be able to view architectural remains, artifacts, and information that are otherwise inaccessible.
series other
last changed 2003/04/23 13:14

_id c62d
authors Davis, Randall
year 1986
title Knowledge-Based Systems
source Science February, 1986. vol. 231: pp. 957-963 : ill. includes bibliography.
summary First developed two decades ago, knowledge-based systems have seen widespread application in recent years. While performance has been a strong focus of attention, building such systems has also expanded our conception of a computer program from a black box providing an answer to something capable of explaining its answers, acquiring new knowledge, and transferring knowledge to students. These abilities derive from distinguishing clearly what the program knows from how that knowledge will be used, making it possible to use the same knowledge in different ways
keywords AI, knowledge base, systems, theory
series CADline
last changed 1999/02/12 14:08

_id f9d8
authors De Valpine, John and Black, Benjamin
year 2001
title Physically Based Daylight Simulation and Visualization
source Reinventing the Discourse - How Digital Tools Help Bridge and Transform Research, Education and Practice in Architecture [Proceedings of the Twenty First Annual Conference of the Association for Computer-Aided Design in Architecture / ISBN 1-880250-10-1] Buffalo (New York) 11-14 October 2001, pp. 406-407
summary While architects typically agree that daylight is a powerful influence for design, architects rarely collect and use daylighting data to help make informed design decisions. This deficiency exists partially because there are no common tools available to provide useful and accessible data. The objective of this project is to provide accurate daylighting data of a prominent urban building site and present it in a clear way so that the architects can make well informed design decisions that respond to site daylighting conditions and improve architectural performance. An urban 3D computer model was created with AutoCad, a commercial CAD application. Daylight was simulated with Radiance, the physically based rendering engine developed at Lawrence Berkeley National Laboratory. The standard CIE model for clear sky and sun was used to produce over 500 images that represent daylight conditions for three different times of year at 10 minute intervals in both luminance and illuminance formats. The simulation data was packaged for analysis with a unique browser tool that enables the architect to easily cycle through the data to evaluate and compare behavior by time of day and by season. The architect can also toggle between luminance and illuminance format to easily visualize both qualitative and quantitative data. The exploration and use of the simulation data can be applied with sensitivity to inform the design and decision making process for the exterior building site.
series ACADIA
last changed 2002/04/25 17:30

_id 2a5e
authors Does, J. van der and Giró, H.
year 1997
title Design communication and image processing
source Architectural and Urban Simulation Techniques in Research and Education [Proceedings of the 3rd European Architectural Endoscopy Association Conference / ISBN 90-407-1669-2]
summary In the proceedings of the first EAEA conference, 1993, I mentioned our first study focused on refining endoscopic video images of a detailed architectural model and drawings. The study was based on work with 900 subjects, of which 200 were professional architects. It has led to a number of technical improvements. In the second study we compared computer-aided design techniques with two techniques from the first study, endoscopic video recordings and coloured and black and white elevations and perspective drawings. Four different groups of 50 subjects took part in this research. We found that computer images are invariably judged to be of moderate value, while drawings yielded consistently high scores. Endoscopic video recordings of the scale model received high scores as far as emotional response is concerned, and moderate scores when the participants were questioned on the actual content of the recordings.
keywords Architectural Endoscopy, Endoscopy, Simulation, Visualisation, Visualization, Real Environments
series EAEA
last changed 2005/09/09 08:43

_id 819d
authors Eiteljorg, H.
year 1988
title Computing Assisted Drafting and Design: new technologies for old problems
source Center for the study of architecture, Bryn Mawr, Pennsylvania
summary In past issues of the Newsletter, George Tressel and I have written about virtual reality and renderings. We have each discussed particular problems with the technology, and both of us mentioned how compelling computer visualizations can be. In my article ("Virtual Reality and Rendering," February, 1995, Vol. 7, no. 4), I indicated my concerns about the quality of the scholarship and the level of detail used in making renderings or virtual worlds. Mr. Tressel (in "Visualizing the Ancient World," November, 1996, Vol. IX, no. 3) wrote about the need to distinguish between real and hypothetical parts of a visualization, the need to differentiate materials, and the difficulties involved in creating the visualizations (some of which were included in the Newsletter in black-and-white and on the Web in color). I am returning to this topic now, in part because the quality of the images available to us is improving so fast and in part because it seems now that neither Mr. Tressel nor I treated all the issues raised by the use of high-quality visualizations. The quality may be illustrated by new images of the older propylon that were created by Mr. Tressel (Figs. 1 - 3); these images are significantly more realistic than the earlier ones, but they do not represent the ultimate in quality, since they were created on a personal computer.
series other
last changed 2003/04/23 13:50

_id sigradi2018_1349
id sigradi2018_1349
authors Fialho Silva, Carolina
year 2018
title White Cube in Evolution: Lighting and Digital Technology in Beyeler Foundation Architecture
source SIGraDi 2018 [Proceedings of the 22nd Conference of the Iberoamerican Society of Digital Graphics - ISSN: 2318-6968] Brazil, São Carlos 7 - 9 November 2018, pp. 52-59
summary This document aims to analyze the lighting system design at the Beyeler Foundation, which is controlled by digital devices. The domination of light is considered its architectural achievement, which generates a peculiar spatial effect. The discussion draws on the white cube concept, as applied in exhibition spaces, on the notion of device, as proposed by Flusser, and on the black box expounded by Latour's theory. The design is understood as the evolution of the white cube, since it maintains the premise of neutral space, but also remains open to the external environment through a glass roof equipped with computerized shutters.
keywords Museum architecture lighting; Digital technology; White cube; Device; Black box
series SIGraDi
last changed 2019/05/20 09:11

_id b1db
authors Francis, Sabu
year 1999
title The Importance of Being Abstract: An Indian Approach to Models
source Architectural Computing from Turing to 2000 [eCAADe Conference Proceedings / ISBN 0-9523687-5-7] Liverpool (UK) 15-17 September 1999, pp. 101-109
summary Traditional Indian way of life is surrounded by ambiguity. This is in direct contrast to an Aristotelian approach, where polarised stands are always taken. A black and white approach tends to yield results speedily, but exhaustive solutions which can explain complexity are usually brute force procedures. Even so, their conclusions in the end are still suspect. The author believes that rich solutions may exist when we use an 'alternate' or abstract synthesized reality to do our modelling instead of relying on analogies and other direct links to the real world. Models that allow synthesis tend to accept ambiguity. The author presents in this paper an 'unconventional' system to represent architecture which has had some amount of success probably because it started of, on pure abstract grounds that allowed ambiguity instead of basing it on an Aristotelian, analytical model.
keywords Aristotle, Buddha, Representations, Abstract Models
series eCAADe
last changed 1999/10/10 12:53

_id ddssup0207
id ddssup0207
authors Geurts, K., Wets, G., Brijs, T. and Vanhoof, K.
year 2002
title The Use of Rule-Based Knowledge Discovery Techniques to Profile Black Spots
source Timmermans, Harry (Ed.), Sixth Design and Decision Support Systems in Architecture and Urban Planning - Part two: Urban Planning Proceedings Avegoor, the Netherlands), 2002
summary In Belgium, traffic safety is currently one of the highest topics on the list of priorities of the government. The identification of black spots and black zones and profiling them in terms of accident related data and location characteristics must provide new insights into the complexity and causes of road accidents which, in turn, provide valuable input for government actions. Data mining is the extraction of information from large amounts of data. The use of data mining algorithms is therefore particularly useful in the context of large datasets on road accidents. In this paper, association rules are used to identify accident circumstances that frequently occur together. The strength of this descriptive approach lies within the definition of different accident types and the identification of relevantvariables that make a strong contribution towards a better understanding of accident circumstances. An analysis of the produced set of rules, describing underlying patterns in the data, indicates that fiveaspects of traffic accidents can be discerned: collision with a pedestrian, collision in parallel, sideways collision, week/weekend accidents and weather conditions. For each of these accident types, different variables play an important role in the occurrence of the accidents.
series DDSS
type normal paper
last changed 2008/11/01 06:38

_id d9bf
authors Goodchild, N.F., Steyaert, L.T., Parks, B.O., Johnson, C., Maidment, D., Crane, M. and Glendinning, S. (Eds.)
year 1996
title GIS and Environmental Modeling: Progress and Research Issues
source Fort Collins, CO: GIS World Books, pp.451-454
summary GIS and Environmental Modeling: Progress and Research Issues Michael F. Goodchild, Louis T. Steyaert, Bradley O. Parks, Carol Johnston, David Maidment, Michael Crane, and Sandi Glendinning, Editors With growing pressure on natural resources and landscapes there is an increasing need to predict the consequences of any changes to the environment. Modelling plays an important role in this by helping our understanding of the environment and by forecasting likely impacts. In recent years moves have been made to link models to Geographical Information Systems to provide a means of analysing changes over an area as well as over time. GIS and Environmental Modeling explores the progress made to date in integrating these two software systems. Approaches to the subject are made from theoretical, technical as well as data stand points. The existing capabilities of current systems are described along with important issues of data availability, accuracy and error. Various case studies illustrate this and highlight the common concepts and issues that exist between researchers in different environmental fields. The future needs and prospects for integrating GIS and environmental models are also explored with developments in both data handling and modelling discussed. The book brings together the knowledge and experience of over 100 researchers from academic, commercial and government backgrounds who work in a wide range of disciplines. The themes followed in the text provide a fund of knowledge and guidance for those involved in environmental modelling and GIS. The book is easily accessible for readers with a basic GIS knowledge and the ideas and results of the research are clearly illustrated with both colour and black and white graphics.
series other
last changed 2003/04/23 13:14

_id e90c
id e90c
authors Greenfield, Gary R.
year 2004
source Proceedings of the Fourth International Conference of Mathematics & Design, Special Edition of the Journal of Mathematics & Design, Volume 4, No.1, pp. 47-54.
summary We consider the problem of automating the recoloring of designs whose “originals” are defined using color look-up tables. Given a digital image , we assume for each pixel of there exists an integer index such that . If is a color look-up table of size , we say image is colored by if each pixel is assigned color . Given a map from to itself, we recolor by assigning to each pixel the color . We are able to evolve the recoloring maps by representing them as -tuples. We automate their evolution by applying multi-objective optimization. Examples of recolorings of designs that were originally in color and originally in black and white are given. Keywords. Color look-up table, image recoloring, genetic algorithm, evolutionary multiobjective optimization.
series other
type normal paper
last changed 2005/04/07 10:47

_id 653f
authors Hedelman, Harold
year 1984
title A Data Flow Approach to Procedural Modeling
source IEEE Computer Graphics and Applications January, 1984. vol. 4: pp. 16-26 : ill. (some col.). includes bibliography.
summary Computer graphics tasks generally involve either modeling or viewing. Modeling combines primitive building blocks (polygons, patches, etc.) into data structures that represent entire objects and scenes. To visualize a modeled object, its data structure is input to appropriate viewing routines. While a great deal has been done on modeling and viewing with geometric primitives, little has been published on the use of procedural primitives. A procedural model is a step-by-step guide for constructing a representation of an object or process, i.e., a program. It is also a function, a 'black box' with a set of inputs and outputs. Two questions are especially pertinent to the work presented in this article
keywords First, what are the advantages of both data flow methods and procedural modeling? Second, how can such models be used in composition? computer graphics, modeling, information, management
series CADline
last changed 1999/02/12 14:08

_id ecaade2018_247
id ecaade2018_247
authors Ilunga, Guilherme and Leit?o, António
year 2018
title Derivative-free Methods for Structural Optimization
source Kepczynska-Walczak, A, Bialkowski, S (eds.), Computing for a better tomorrow - Proceedings of the 36th eCAADe Conference - Volume 1, Lodz University of Technology, Lodz, Poland, 19-21 September 2018, pp. 179-186
summary The focus on efficiency has grown over recent years, and nowadays it is critical that buildings have a good performance regarding different criteria. This need prompts the usage of algorithmic approaches, analysis tools, and optimization algorithms, to find the best performing variation of a design. There are many optimization algorithms and not all of them are adequate for a specific problem. However, Genetic Algorithms are frequently the first and only option, despite being considered last resort algorithms in the mathematical field. This paper discusses methods for structural optimization and applies them on a structural problem. Our tests show that Genetic Algorithms perform poorly, while other algorithms achieve better results. However, they also show that no algorithm is consistently better than the others, which suggests that for structural optimization, several algorithms should be used, instead of simply using Genetic Algorithms.
keywords Derivative-free Optimization; Black-box Optimization; Structural Optimization; Algorithmic Design
series eCAADe
last changed 2018/07/24 10:22

_id caadria2018_126
id caadria2018_126
authors Khean, Nariddh, Kim, Lucas, Martinez, Jorge, Doherty, Ben, Fabbri, Alessandra, Gardner, Nicole and Haeusler, M. Hank
year 2018
title The Introspection of Deep Neural Networks - Towards Illuminating the Black Box - Training Architects Machine Learning via Grasshopper Definitions
source T. Fukuda, W. Huang, P. Janssen, K. Crolla, S. Alhadidi (eds.), Learning, Adapting and Prototyping - Proceedings of the 23rd CAADRIA Conference - Volume 2, Tsinghua University, Beijing, China, 17-19 May 2018, pp. 237-246
summary Machine learning is yet to make a significant impact in the field of architecture and design. However, with the combination of artificial neural networks, a biologically inspired machine learning paradigm, and deep learning, a hierarchical subsystem of machine learning, the predictive capabilities of machine learning processes could prove a valuable tool for designers. Yet, the inherent knowledge gap between the fields of architecture and computer science has meant the complexity of machine learning, and thus its potential value and applications in the design of the built environment remain little understood. To bridge this knowledge gap, this paper describes the development of a learning tool directed at architects and designers to better understand the inner workings of machine learning. Within the parametric modelling environment of Grasshopper, this research develops a framework to express the mathematic and programmatic operations of neural networks in a visual scripting language. This offers a way to segment and parametrise each neural network operation into a basic expression. Unpacking the complexities of machine learning in an intermediary software environment such as Grasshopper intends to foster the broader adoption of artificial intelligence in architecture.
keywords machine learning; neural network; action research; supervised learning; education
series CAADRIA
last changed 2018/05/17 07:08

For more results click below:

this is page 0show page 1show page 2HOMELOGIN (you are user _anon_246132 from group guest) CUMINCAD Papers Powered by SciX Open Publishing Services 1.002