Cumincad : CUMINCAD Papers : Search Results

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
References

Hits 1 to 20 of 219

Reformat results as: short short into frame detailed detailed into frame

_id	ea2f
authors	Heisserman, Jeff A.
year	1991
title	Generative geometric design and boundary solid grammars
source	Carnegie Mellon University, Department of Architecture
summary	This thesis explores the automatic generation of solid models based on a grammatical paradigm. It introduces a formalism, boundary solid grammars, for this purpose. In this formalism, a set of geometric rules is applied to an initial solid model to generate a language or family of solids. A rule may match on a portion of the boundary of a solid, and then modify the solid or add new solids. Genesis is presented as an implementation of the formalism. A number of grammars have been constructed to demonstrate the concepts and usefulness of the formalism. These grammars generate simple geometric forms including snodakes, recursive octahedra, “fractal” mountains, and spirals. Another grammar generates stereo lithography support structures. Queen Anne houses have been characterized with a more extensive grammar. Grammars are also being developed to generate housings for small computers and structural designs for high rise buildings. The thesis introduces the unary shape operations and a new paradigm for solid modeling, The unary shape operations take models that may have self-intersections, interpret the models consid- ering the given geometry and face orientations, and produce valid models. Local operations, the unary shape operations, and Boolean operations are used together within a valid modeling scheme. The thesis introduces a new boundary representation for manifold and nonmanifold solids, the generalized split-edge representation. It describes generalized Euler operations which manipulate the topology of the nonmanifold representation. Finally, the thesis presents a form of the Euler- Poincare equation that characterizes the relationship between elements of nonmanifold surfaces of solids.
series	thesis:PhD
last changed	2003/02/12 22:37

_id	f14c
authors	Sariyildiz, Sevil
year	1991
title	Conceptual Design by Means of Islamic-Geometric-Patterns within a CAAD-Environment
source	Delft University of Technology
summary	The starting point in this research was to develop a 3D grammar theory on top of existing 2D Islamic-geometric-patterns, trying to rescue their fundamental geometry contents to be applied in contemporary architecture without compromising any architectural style. As it is self evident the architectural design process consists of clearly distinct stages namely conceptual design, materialisation and further completion. A this conceptual stage the innovative item of the research deals with pattern grammars on 3D complex geometrical patterns, considering them as polyhedra and polytopes, for their use as an underlayer to a concept design, like architects use 2D rectangular and triangular grids by the conventional way. Handling these complex 3D patterns requires a special environment which is possible with CAAD. Within the CAAD environment, the handling of these complex patterns is easily done by means of 3D tools, because the 3D tools permit the user to make any possible manipulations and geometrical transformations in an easier way in space. To a geometrical patterns, there is some attention paid during the last 50 years by some scholars. The most complex geometrical patterns are highly developed in Islamic architecture because it is forbidden in Muslim religion to use man's portraits or sculptures of human beings in the religious buildings. All these approaches to complex patterns are analysed and studied as 2D elements. The question was how could we consider them in 3rd dimensions and use them instead of 2D underlayer, as 3D underlayers in the conceptual phase of the CAAD design. Pattern grammar is a generally employable aid (underlying pattern) for conceptual and material designs. On the basis of rules of symmetry and substitution, ordering principles have been worked out, which can be used for formal design methods as well as detailing systems (e.g. modular coordination). Through the realization of a pattern grammar a wider range of underlying patterns can be offered and a choice from these can be made in a more fundamental manner. At a subsequent stage the collection of "empty boxes" can be filled with (architectural) elements in such a way that another option is created between either filling up the boxes completely, filling them partly, or filling them in such a way that they overflow. It is self-evident that underlying patterns can also be used for details and decoration in a design. Concerning the materialisation of the concept design, within the 3D CAAD environment, substitution methods are partially developed. Further theoretical developments concerning the materialisation phase constantly backed up through feed-back with specialist matters (such as e.g. by means of expert systems, decision-support systems), must be worked out. As feed-back of the research, the possibilities of the design with 3D patterns have been tested and the procedures are explained. () Working with 3D patterns gives a designer more inspirations to develop new ideas and new concepts and gives the opportunity to handle the complexity. () The formal, structural and symmetrical qualities of geometrical patterns has a positive influence on the industrialisation of the building components. (*) Working with 3D tools which are able to handle complex geometry have a result because of the accuracy of the information, that there has hardly been a mistake made during the preparation and the assembly of the building components. This has also positive results concerning the financial aspects of the building process.
series	thesis:PhD
email
last changed	2003/02/12 22:37

_id	ga9921
id	ga9921
authors	Coates, P.S. and Hazarika, L.
year	1999
title	The use of genetic programming for applications in the field of spatial composition
source	International Conference on Generative Art
summary	Architectural design teaching using computers has been a preoccupation of CECA since 1991. All design tutors provide their students with a set of models and ways to form, and we have explored a set of approaches including cellular automata, genetic programming ,agent based modelling and shape grammars as additional tools with which to explore architectural ( and architectonic) ideas.This paper discusses the use of genetic programming (G.P.) for applications in the field of spatial composition. CECA has been developing the use of Genetic Programming for some time ( see references ) and has covered the evolution of L-Systems production rules( coates 1997, 1999b), and the evolution of generative grammars of form (Coates 1998 1999a). The G.P. was used to generate three-dimensional spatial forms from a set of geometrical structures .The approach uses genetic programming with a Genetic Library (G.Lib) .G.P. provides a way to genetically breed a computer program to solve a problem.G. Lib. enables genetic programming to define potentially useful subroutines dynamically during a run .* Exploring a shape grammar consisting of simple solid primitives and transformations. * Applying a simple fitness function to the solid breeding G.P.* Exploring a shape grammar of composite surface objects. * Developing grammarsfor existing buildings, and creating hybrids. * Exploring the shape grammar of abuilding within a G.P.We will report on new work using a range of different morphologies ( boolean operations, surface operations and grammars of style ) and describe the use of objective functions ( natural selection) and the "eyeball test" ( artificial selection) as ways of controlling and exploring the design spaces thus defined.
series	other
more	http://www.generativeart.com/
last changed	2003/08/07 17:25

_id	ga0010
id	ga0010
authors	Moroni, A., Zuben, F. Von and Manzolli, J.
year	2000
title	ArTbitrariness in Music
source	International Conference on Generative Art
summary	Evolution is now considered not only powerful enough to bring about the biological entities as complex as humans and conciousness, but also useful in simulation to create algorithms and structures of higher levels of complexity than could easily be built by design. In the context of artistic domains, the process of human-machine interaction is analyzed as a good framework to explore creativity and to produce results that could not be obtained without this interaction. When evolutionary computation and other computational intelligence methodologies are involved, every attempt to improve aesthetic judgement we denote as ArTbitrariness, and is interpreted as an interactive iterative optimization process. ArTbitrariness is also suggested as an effective way to produce art through an efficient manipulation of information and a proper use of computational creativity to increase the complexity of the results without neglecting the aesthetic aspects [Moroni et al., 2000]. Our emphasis will be in an approach to interactive music composition. The problem of computer generation of musical material has received extensive attention and a subclass of the field of algorithmic composition includes those applications which use the computer as something in between an instrument, in which a user "plays" through the application's interface, and a compositional aid, which a user experiments with in order to generate stimulating and varying musical material. This approach was adopted in Vox Populi, a hybrid made up of an instrument and a compositional environment. Differently from other systems found in genetic algorithms or evolutionary computation, in which people have to listen to and judge the musical items, Vox Populi uses the computer and the mouse as real-time music controllers, acting as a new interactive computer-based musical instrument. The interface is designed to be flexible for the user to modify the music being generated. It explores evolutionary computation in the context of algorithmic composition and provides a graphical interface that allows to modify the tonal center and the voice range, changing the evolution of the music by using the mouse[Moroni et al., 1999]. A piece of music consists of several sets of musical material manipulated and exposed to the listener, for example pitches, harmonies, rhythms, timbres, etc. They are composed of a finite number of elements and basically, the aim of a composer is to organize those elements in an esthetic way. Modeling a piece as a dynamic system implies a view in which the composer draws trajectories or orbits using the elements of each set [Manzolli, 1991]. Nonlinear iterative mappings are associated with interface controls. In the next page two examples of nonlinear iterative mappings with their resulting musical pieces are shown.The mappings may give rise to attractors, defined as geometric figures that represent the set of stationary states of a non-linear dynamic system, or simply trajectories to which the system is attracted. The relevance of this approach goes beyond music applications per se. Computer music systems that are built on the basis of a solid theory can be coherently embedded into multimedia environments. The richness and specialty of the music domain are likely to initiate new thinking and ideas, which will have an impact on areas such as knowledge representation and planning, and on the design of visual formalisms and human-computer interfaces in general. Above and bellow, Vox Populi interface is depicted, showing two nonlinear iterative mappings with their resulting musical pieces. References [Manzolli, 1991] J. Manzolli. Harmonic Strange Attractors, CEM BULLETIN, Vol. 2, No. 2, 4 -- 7, 1991. [Moroni et al., 1999] Moroni, J. Manzolli, F. Von Zuben, R. Gudwin. Evolutionary Computation applied to Algorithmic Composition, Proceedings of CEC99 - IEEE International Conference on Evolutionary Computation, Washington D. C., p. 807 -- 811,1999. [Moroni et al., 2000] Moroni, A., Von Zuben, F. and Manzolli, J. ArTbitration, Las Vegas, USA: Proceedings of the 2000 Genetic and Evolutionary Computation Conference Workshop Program – GECCO, 143 -- 145, 2000.
series	other
email
more	http://www.generativeart.com/
last changed	2003/08/07 17:25

_id	diss_hensen
id	diss_hensen
authors	Hensen, J.L.M.
year	1991
title	On the Thermal Interaction of Building Structure and Heating and Ventilating System
source	Eindhoven University of Technology
summary	In this dissertation, developments in the field of building performance evaluation tools are described. The subject of these tools is the thermal interaction of building structure and heating and ventilating system. The employed technique is computer simulation of the integrated, dynamic system comprising the occupants, the building and its heating and ventilating system. With respect to buildings and the heating and ventilating systems which service them, the practical objective is ensuring thermal comfort while using an optimum amount of fuel. While defining the optimum had to be left for other workers, the issue of thermal comfort is addressed here. The conventional theory of thermal comfort in conditions characteristic for dwellings and offices assumes steady-state conditions. Yet thermal conditions in buildings are seldom steady, due to the thermal interaction between building structure, climate, occupancy, and auxiliary systems. A literature rewiew is presented regarding work on thermal comfort specifically undertaken to examine what fluctuations in indoor climate may be acceptable. From the results, assessment criteria are defined. Although its potentials reach beyond the area of Computer Aided Building Design, a description is given of building and plant energy simulation within the context of the CABD field of technology. Following an account of the present state-of-the-art, the choice for starting from an existing energy simulation environment (ESPR) is justified. The main development areas of this software platform - within the present context - are identified as: fluid flow simulation, plant simulation, and their integration with the building side of the overall problem domain. In the field of fluid flow simulation, a fluid flow network simulation module is described. The module is based on the mass balance approach, and may be operated either in standalone mode or from within the integrated building and plant energy simulation system. The program is capable of predicting pressures and mass flows in a user-defined building / plant network comprising nodes (ie building zones, plant components, etc) and connections (ie air leakages, fans, pipes, ducts, etc), when subjected to flow control (eg thermostatic valves) and / or to transient boundary conditions (eg due to wind). The modelling and simulation techniques employed to predict the dynamic behaviour of the heating and ventilating system, are elaborated. The simultaneous approach of the plant and its associated control is described. The present work involved extensions to the ESPR energy simulation environment with respect to robustness of the program, and with respect to additional plant simulation features, supported plant component models and control features. The coupling of fluid flow, plant side energy and mass, and building side energy simulation into one integrated program is described. It is this "modular-simultaneous" technique for the simulation of combined heat and fluid flow in a building / plant context, which enables an integral approach of the thermal interaction of building structure and heating and ventilating system. A multi stage verification and validation methodology is described, and its applicability to the present work is demonstrated by a number of examples addressing each successive step of the methodology. A number of imaginary and real world case studies are described to demonstrate application of the present work both in a modelling orientated context and in a building engineering context. Then the general conclusions of the present work are summarized. Next and finally, there are recommendations towards possible future work in the areas of: theory, user interface, software structure, application, and technology transfer.
series	thesis:PhD
last changed	2003/12/15 14:43

_id	870b
authors	Sivaloganathan, Sangarappillai
year	1991
title	Sketching input for computer aided engineering
source	City University, Department of Mechanical Engineering and Aeronautics
summary	The design process often begins with a graphical description of the proposed device or system and sketching is the physical expression of the design engineer's thinking process. Computer Aided Design is a technique in which man and machine are blended into a problem solving team, intimately coupling the best characteristics of each. Solid modelling is developed to act as the common medium between man and the computer. At present it is achieved mainly by designing with volumes and hence does not leave much room for sketching input, the traditional physical expression of the thinking process of the design engineer. This thesis describes a method of accepting isometric free hand sketching as the input to a solid model. The design engineer is allowed to make a sketch on top of a digitizer indicating (i) visible lines; (ii) hidden lines; (iii) construction lines; (iv) centre lines; (v) erased lines; and (vi) redundant lines as the input. The computer then processes this sketch by identifying the line segments, fitting the best possible lines, removing the erased lines, ignoring the redundant lines and finally merging the hidden lines and visible lines to form the lines in the solid in an interactive manner. The program then uses these lines and the information about the three dimensional origin of the object and produces three dimensional information such as the faces, loops, holes, rings, edges and vertices which are sufficient to build a solid model. This is achieved in the following manner. The points in the sketch are first written into a file. The computer than reads this file, breaks the group of points into sub-groups belonging to individual line segments, fits the best lines and identify the vertices in two dimensions. These improved lines in two dimensions are then merged to form the lines and vertices in the solid. These lines are then used together with the three dimensional origin (or any other point) to produce the wireframe model in three dimensions. The loops in the wireframe models are then identified and surface equations are fitted to these loops. Finally all the necessary inputs to build a B-rep solid model are produced.
series	thesis:PhD
last changed	2003/02/12 22:37

_id	83b4
authors	Tan, Milton
year	1991
title	Themes for Schemes: Design Creativity as the Conceptualization, Transformation, and Representation of Emergent Forms
source	Harvard University, Graduate School of Design
summary	Architects, graphic designers, and others frequently develop designs by picking out and transforming subshapes of two-dimensional or three-dimensional shapes. Shape grammars formalize this aspect of design by specifying rules of the form $a /to b$: the left-hand side a describes a type of subshape that may be picked out, while the right-hand side b describes what that type of subshape may become. Designs in the language specified by a shape grammar are derived by recursively applying the shape transformation rules to a starting shape. To apply a shape-transformation rule automatically, a computer system, must have the capacity to recognize instances of the type of subshape specified on the left-hand side of the rule. Sometimes such instances are explicitly input by the designer, and explicitly represented in a data structure: in this case, recognition is a relatively straightforward task. But there may also be 'emergent' instances that were not explicitly input, and are only indirectly represented in the data structure. These emergent instances are potentially numerous, and may be extremely difficult to discover. This thesis focuses on mechanisms for picking out and transforming subshapes. The first three chapters place the issue in its broadest context by arguing that different designers--bringing different knowledge and attitudes to the task--will pick out and pay attention to different subshapes in a drawing. This contention is supported by introducing some of the relevant literature on perception, problem-solving, and creativity. Chapter 4 introduces shape grammars to provide a more formal framework for investigating this topic. Chapter 5 describes the properties and limitations of Topdown--a computer program which supports design by applying the rules of a shape grammar, but does not provide for recognition of emergent subshapes. Chapter 6 introduces ECART, a computer program which supports efficient recognition and transformation of emergent subshapes, and demonstrates how its performance transcends that of Topdown. Examination of the results produced by ECART suggest that a designer's conceptual filter--the repertoire of subshape types that he or she can recognize in a drawing--plays a crucial role in the development of design ideas.
keywords	Computer Graphics; Computer Software; Development
series	thesis:PhD
email
last changed	2003/02/12 22:37

_id	9c37
id	9c37
authors	Coates P, Derix C, Krakhofer S and Karanouh A
year	2005
title	Generating Architectural Spatial Configurations: two approaches using voronoi tessellations and particle systems
source	Proceedings of the Generative Arts conference, Milan, 2005
summary	It was one of the primary goals of the original Master’s programme in Computing and design at UEL in 1991 that we should work towards defining morphological generative processes for the conceptual design of architectural objects. These two papers offer a range of techniques which have been developed by two of this years MSc students (04-05) which show that we are getting close to this. The approaches range from computational geometric approaches (3d parametrics and voronoi diagrams) to emergent spatial organisation using agent based modelling. In many cases the resultant geometry is defined to the point where it can be transferred to advanced evaluation and fabrication systems, thus making this work sufficiently developed to begin to form a useful part in practical design processes.
keywords	morphology, computational geometry, particle systems, physical simulation, voronoi diagrams
series	other
type	normal paper
email
more	http://www.generativeart.com/
last changed	2012/09/20 18:39

_id	9e10
authors	Tobin, Kenneth L.
year	1991
title	Constraint-Based Three-Dimensional Modeling as a Design Tool
doi	https://doi.org/10.52842/conf.acadia.1991.193
source	Reality and Virtual Reality [ACADIA Conference Proceedings / ISBN 1-880250-00-4] Los Angeles (California - USA) October 1991, pp. 193-209
summary	Architectural design knowledge which may take the form of constraints and how it can be incorporated into the solid modeling process is explored and discussed. This theoretical exploration leads to an illustration of a Rinctional general purpose three-dimensional solid modeler which utilizes design knowledge as constraints upon the interactive solid modeling process. From this illustration, it is shown that the incorporation of constraints into ComputerAided Architectural Design can effectively assist in the early stages of the design process. A prototypical application is presented which provides an example in which the definition and implementation of specific design knowledge constrains or characterizes the generative and interactive behavior of user definable threedimensional entities. The basis of this work lies in the ability to create a design space, or building envelope, and to allow the interactive modeling of conceptual elements within, and constrained to, that design space, including the realistic interaction between the entities themselves.
series	ACADIA
email
last changed	2022/06/07 07:58

_id	a40d
authors	Paoluzzi, Alberto and Sansoni, Claudio
year	1991
title	Solid Modeling of Architectural Design with PLASM Language
source	Computer Aided Architectural Design Futures: Education, Research, Applications [CAAD Futures ‘91 Conference Proceedings / ISBN 3-528-08821-4] Zürich (Switzerland), July 1991, pp. 203-224
summary	PLASM (Programming LAnguage for Solid Modeling) is a prototype, high level, user oriented, functional design language currently being developed at the University of Rome "La Sapienza". A PLASM "program" is the symbolic definition of a complex of variational polyhedra depending on some unbound variable, and therefore allows for the description of a whole set of geometric solutions to a design problem. In our view the language should be used, possibly with the assistance of a graphical user interface, both in the first steps of the design process as well in the detailed design. In the paper the guide-lines are shown for the preliminary definition of the syntax of the language. The paper also contains the definition of some new and very powerful solid operators.
series	CAAD Futures
last changed	1999/04/07 12:03

_id	00bc
authors	Chen, Chen-Cheng
year	1991
title	Analogical and inductive reasoning in architectural design computation
source	Swiss Federal Institute of Technology, ETH Zurich
summary	Computer-aided architectural design technology is now a crucial tool of modern architecture, from the viewpoint of higher productivity and better products. As technologies advance, the amount of information and knowledge that designers can apply to a project is constantly increasing. This requires development of more advanced knowledge acquisition technology to achieve higher functionality, flexibility, and efficient performance of the knowledge-based design systems in architecture. Human designers do not solve design problems from scratch, they utilize previous problem solving episodes for similar design problems as a basis for developmental decision making. This observation leads to the starting point of this research: First, we can utilize past experience to solve a new problem by detecting the similarities between the past problem and the new problem. Second, we can identify constraints and general rules implied by those similarities and the similar parts of similar situations. That is, by applying analogical and inductive reasoning we can advance the problem solving process.The main objective of this research is to establish the theory that (1) design process can be viewed as a learning process, (2) design innovation involves analogical and inductive reasoning, and (3) learning from a designer's previous design cases is necessary for the development of the next generation in a knowledge-based design system. This thesis draws upon results from several disciplines, including knowledge representation and machine learning in artificial intelligence, and knowledge acquisition in knowledge engineering, to investigate a potential design environment for future developments in computer-aided architectural design. This thesis contains three parts which correspond to the different steps of this research. Part I, discusses three different ways - problem solving, learning and creativity - of generating new thoughts based on old ones. In Part II, the problem statement of the thesis is made and a conceptual model of analogical and inductive reasoning in design is proposed. In Part III, three different methods of building design systems for solving an architectural design problem are compared rule-based, example-based, and case-based. Finally, conclusions are made based on the current implementation of the work, and possible future extensions of this research are described. It reveals new approaches for knowledge acquisition, machine learning, and knowledge-based design systems in architecture.
series	thesis:PhD
email
last changed	2003/05/10 05:42

_id	2e56
authors	Coyne, Robert Francis
year	1991
title	ABLOOS : an evolving hierarchical design framework
source	Carnegie Mellon University, Department of Architecture
summary	The research reported in this thesis develops an approach toward a more effective use of hierarchical decomposition in computational design systems. The approach is based on providing designers a convenient interactive means to specify and experiment with the decompositional structure of design problems, rather than having decompositions pre-specified and encoded in the design system. Following this approach, a flexible decomposition capability is combined with an underlying design method to form the basis for an extensible and evolving framework for cooperative (humdcomputer) design. As a testbed for this approach, the ABLOOS framework for layout design is designed and constructed as a hierarchical extension of LOOS.’The framework enables a layout task to be hierarchically decomposed, and for the LOOS methodology to be applied recursively to layout subtasks at appropriate levels of abstraction within the hierarchy; layout solutions for the subtasks are then recomposed to achieve an overall solution, Research results thus far are promising: ABLOOS has produced high quality solutions for a class of industrial layout design tasks (an analog power board layout with 60 components that have multiple complex constraints on their placement); the adaptability of the framework across domains and disciplines has been demonstrated; and, further development of ABLOOS is underway including its extension to layouts in 2 1/2D space and truly 3D arrangements. The contribution of this work is in demonstrating an effective, flexible and extensible capability for hierarchical decomposition in design. It has also produced a more comprehensive layout system that can serve as a foundation for the further investigation of hierarchical decomposition in a variety of design domains.
series	thesis:PhD
last changed	2003/02/12 22:37

_id	diss_kuo
id	diss_kuo
authors	Kuo, C.J.
year	1999
title	Unsupervised Dynamic Concurrent Computer-Aided Design Assistant
source	Los Angeles: UCLA
summary	The increasing capability of computer-aided architectural design systems has strengthened the role that the computer plays in the workplace. Due to the complexity of developing new techniques and research, these systems are undertaken mostly by scientists and engineers without significant architectural input (Willey, 1991). The design concept of these systems may be based on a well-defined and well-understood process, which is not yet realized in architectural design (Galle, 1994). The output of such research may not be easily adapted into the design process. Most of the techniques assume a complete understanding of the design space (Gero and Maher, 1987) (Willey, 1991). The description or construction of the design space is always time and space consuming, and the result can never be complete due to the ever-changing nature of architectural design. This research intends to initiate a solution for the above problems. The proposed system is an unsupervised-dynamic-concurrent-computer-aided-design assistant. The “unsupervised” means the learning process is not supervised by the user because it is against the designer's nature to “think-aloud” in the design studio and it also increases the work load. It is dynamic because the size of the knowledge base is constantly changing. Concurrent means that there are multiple procedures active simultaneously. This research focuses on learning the operational knowledge from an individual designer and reapplying it in future designs. A computer system for this experiment is constructed. It is capable of The preliminary result shows a positive feedback from test subjects. The purpose of this research is to suggest a potent computational frame within which future developments may flourish.
series	thesis:PhD
last changed	2003/11/28 07:37

_id	1839
authors	Papamichael, Konstantinos Michael
year	1991
title	Design process and knowledge possibilities and limitations of computer-aided design
source	University of California, Berkeley
summary	An attempt to determine how computers can be used to assist designers resulted in the development of a design theory, according to which design is 'feeling and thinking while acting.' Design is theorized as living through one's imagination, however being continuously affected by real life itself. The design process is decomposed into elementary activities that are characterized with respect to the nature of knowledge requirements and the degree to which they can be specified and delegated to computers. The results are considered as criteria to determine possibilities and limitations of computer-aided design. An integration of a variety of computer applications tools is proposed towards the design and development of a computer-based Design Support Environment (DSE), that is applicable to any design domain. The proposed DSE automates all specifiable and delegable design activities, while assisting with the nondelegable ones through appropriate user interface. A DSE demonstration prototype is also presented in the Appendix. This prototype addresses the design of fenestration and electric lighting systems of office spaces with respect to comfort, energy and cost.
series	thesis:PhD
email
last changed	2003/02/24 20:32

_id	b5be
authors	Stok, Leon
year	1991
title	Architectural synthesis and optimization of digital systems
source	Eindhoven University of Technology
summary	High level synthesis means going from an functional specification of a digits-system at the algorithmic level to a register transfer level structure. Different appli-cations will ask for different design styles. Despite this diversity in design styles many tasks in the synthesis will be similar. There is no need to write a new synthesis system for each design style. The best way to go seems a decomposition of the high level synthesis problems in several well defined subproblems. How the problem is decomposed depends heavily on a) the type of network architecture chosen, b) the constraints applied to the design and c) on the functional description itself. From this architecture style, the constraints and the functional description a synthesis scheme can be derived.Once this scheme is fixed, algorithms can be chosen which fit into this scheme and solve the subproblems in a fast and, when possible, optimal way. To support such a synthesis philosophy, a framework is needed in which all design information can be stored in a unique way during the various phases of the design process. This asks for a design data base capable of handling all design information with a formally defined interface to all design tools. This thesis gives a formal way to describe both the functional representation, the register transfer level structure and the controller and the relations between all three of them. Special attention has been paid to the efficient representation of mutual exclusive operations and array accesses. The scheduling and allocation problems are defined as mappings between these formal representations. Both the existing synthesis algorithms and the new algorithms described in this thesis fit into this framework. Three new allocation algorithms are presented in this thesis: an algorithm for optimal register allocation in cyclic data flow graphs, an exact polynomial algorithm to do the module allocation and a new scheme to minimize the number of interconnections during all stages of the data path allocation. Cyclic data flow graphs result from high level behavioral descriptions that contain loops. Algorithms for register allocation in high level synthesis published up till now, only considered loop free data flow graphs, When these algorithms are applied to data flow graphs with loops, unnecessary register transfer operations are introduced. A new algorithm is presented that performs a minimal register allocation and eliminates all superfluous register transfer operations. The problem is reformulated as a multicommodity network flow problem for which very efficient solutions exist. Experiments on a benchmark set have shown that in all test cases all register transfers could be eliminated at no increase in register cost. Only heuristic algorithms appeared in literature to solve the module allocation problem. The module allocation problem is usually defined as a clique cover problem on a so-called module allocation graph. It is shown that, under certain conditions, the module allocation graph belongs to the special class of comparability graphs. A polynomial time algorithm can optimally find a clique cover of such a graph. Even when interconnect weights are taken into account, this can be solved exactly. This problem can be transformed into a maximal cost network flow problem, which can be solved exactly in polynomial time. An algorithm is described which solves the module allocation problem with interconnect weights exactly, with a complexity O(kn2), where n is the number of operations In previous research, interconnection was optimized when the module allocation for the operations and the register allocation for the variables already had been done. However, the amount of multiplexing and interconnect are crucial factors to both the delay and the area of a circuit. A new scheme is presented to minimize the number of interconnections during the data path allocation. This scheme first groups all values based on their read and write times. Values belonging to the same group can share a register file. This minimizes the number of data transfers with different sources and destinations. Secondly, registers are allocated for each group separately. Finally the interconnect allocation is done. During the interconnect allocation, the module allocation is determined. The value grouping is based on edge coloring algorithms providing a sharp upper bound on the number of colors needed two techniques: splitting read and write phases of values and introducing serial (re-)write operations for the same value, make that even more efficient exact edge coloring algorithms can be used. It is shown that when variables are grouped into register files and operations are assigned to modules during the interconnection minimization, significant savings (20%) can be obtained in the number of local interconnections and the amount of global interconnect, at the expense of only slightly more register area.
keywords	Digital Systems; Digital Systems
series	thesis:PhD
email
last changed	2003/02/12 22:37

_id	9006
authors	Tham, K.W. (Kwok Wai)
year	1991
title	A Model of Routine Design Using Design Prototypes
source	University of Sydney, Department of Architectural and Design Science
series	thesis:PhD
last changed	2003/02/12 22:37

_id	68c8
authors	Flemming, U., Coyne, R. and Fenves, S. (et al.)
year	1994
title	SEED: A Software Environment to Support the Early Phases in Building Design
source	Proceeding of IKM '94, Weimar, Germany, pp. 5-10
summary	The SEED project intends to develop a software environment that supports the early phases in building design (Flemming et al., 1993). The goal is to provide support, in principle, for the preliminary design of buildings in all aspects that can gain from computer support. This includes using the computer not only for analysis and evaluation, but also more actively for the generation of designs, or more accurately, for the rapid generation of design representations. A major motivation for the development of SEED is to bring the results of two multi-generational research efforts focusing on `generative' design systems closer to practice: 1. LOOS/ABLOOS, a generative system for the synthesis of layouts of rectangles (Flemming et al., 1988; Flemming, 1989; Coyne and Flemming, 1990; Coyne, 1991); 2. GENESIS, a rule-based system that supports the generation of assemblies of 3-dimensional solids (Heisserman, 1991; Heisserman and Woodbury, 1993). The rapid generation of design representations can take advantage of special opportunities when it deals with a recurring building type, that is, a building type dealt with frequently by the users of the system. Design firms - from housing manufacturers to government agencies - accumulate considerable experience with recurring building types. But current CAD systems capture this experience and support its reuse only marginally. SEED intends to provide systematic support for the storing and retrieval of past solutions and their adaptation to similar problem situations. This motivation aligns aspects of SEED closely with current work in Artificial Intelligence that focuses on case-based design (see, for example, Kolodner, 1991; Domeshek and Kolodner, 1992; Hua et al., 1992).
series	other
email
last changed	2003/04/23 15:14

_id	ef46
authors	Petrovic, I.
year	1991
title	Integrative Knowledge-Based Design Systems : A View
source	The Computer Integrated Future, CIB W78 Seminar September, 1991. Unnumbered : ill. includes bibliography.
summary	The paper describes a recent project whose objective was to redesign GIMSEX-PERT, an existing architectural knowledge- based design system developed in 1987. Its critical generative problems appeared to be the rigid structure and limited evaluation criteria. The project's outcome is DESTOOLS, based on the 'object-oriented-methodology' inspired by the traditional trial-and-error approach. It includes a set of interchangeable design methods that can be applied interactively by any desired sequence, producing or transforming a GIMS Building System object. Such 'moderately- loose' system structure offers flexibility in use, avoids pitfalls of knowledge-based design systems with rigid structure, and is applicable in design research, education and practice
keywords	knowledge base, design, architecture, methods, systems, education, practice, integration, evaluation
series	CADline
last changed	2003/06/02 13:58

_id	019c
authors	Beyer, Horst A. and Streilein, André
year	1991
title	Data Generation for CAAD with Digital Photogrammetry
source	Computer Aided Architectural Design Futures: Education, Research, Applications [CAAD Futures ‘91 Conference Proceedings / ISBN 3-528-08821-4] Zürich (Switzerland), July 1991, pp. 583-594
summary	The rapid advances in sensor technology and processing hardware make the development of a Digital Photogrammetric System for Architectural Photogrammetry possible. This system is able to acquire images with sufficient resolution for Architectural Photogrammetry. Geometric and topologic information for a CAAD-System can be derived with manual and/or semi-automated methods. This paper describes the current status of such a system which is under development at the Institute of Geodesy and Photogrammetry in cooperation with the Chair of Architecture and CAAD, both at the Swiss Federal Institute of Technology in Zurich.
series	CAAD Futures
last changed	2003/11/21 15:16

_id	2a0e
authors	Jacobs, Stephen Paul
year	1991
title	The CAD Design Studio: 3D modeling as a fundamental design skill
source	McGraw-Hill, New York
summary	Until now, books on CAD aimed at architects have addressed the use of computer-aided design and drafting as a recording tool, a faster means of producing and storing finished working drawings-and not as an adjunctive creative tool for the design process. Without being software specific, this book will guide the professional and student architect and graphics designer in how to use the computer as an electronic modelling tool, exploring graphic and geometric forms and systems with the freedom and speed of the computer. The reader will be led through a progression of design exercises and design problems, learning how to come up with multiple solutions to a given program. Beautifully illustrated throughout, including 10 four-color CAD drawings!
series	other
last changed	2003/04/23 15:14

For more results click below: