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 9622

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

_id	807e
authors	Maver, Thomas W. and Petric, Jelena (Eds.)
year	1994
title	The Virtual Studio [Conference Proceedings]
doi	https://doi.org/10.52842/conf.ecaade.1994
source	eCAADe Conference Proceedings / ISBN 0-9523687-0-6 / Glasgow (Scotland) 7-10 September 1994, 262 p.
summary	ECAADE was established in 1982 with the intention, across Europe, of facilitating the adoption of the Information Technologies - particularly Computer Aided Architectural Design (CAAD) - within the system of architectural education. The Association, in the 12 years of its existence, has grown in its membership (now close to 350) and in its importance. The annual conferences (Delft 82, Brussels 83, Helsinki 84, Rotterdam 85, Rome 86, Zurich 87, Aarhus 89, Budapest 90, Munich 91, Barcelona 92 and Eindhoven 93) now number 12 and this volume records the 70 or so contributions to the Conference held in Glasgow over the period 7-10 September 1994.The proceedings are arranged according to a number of themes. Theories and Ideas, Teaching and Learning, Visualisation, Multi-Media, Virtual Reality, Virtual Design Studios, Functional Analysis, Design Support Systems and Surveys of Activity. The Conference featured 'long presentations'; and 'short presentations'; the length of these presentations is reflected in the two main sections of this text. To preserve the spirit of conference communication and ensure the rapid dissemination of ideas in a fast grown community of polyglot Europeans, no changes to the papers, which were submitted in Apple Mac and/or PC diskettes, have been imposed; you see them as they were submitted and as the authors intended.
series	eCAADe
email
last changed	2022/06/07 07:49

_id	ga0105
id	ga0105
authors	Mueller, Robert Emmett
year	2001
title	VISIC: A Scoreable Keyboard Color Music
source	International Conference on Generative Art
summary	This paper describes a system to create a video color music that can be scored like music and played on a standard electronic keyboard. Here called "VISIC" the program generates a great variety of multi-colored visual shapes that are propelled through video space on a computer screenfor a limited length of time. The shapes or lines or planes interact in a systemized manner until they disappear, much like musical tones or chords. A notation convention allows VISIC ideas to be composed, replayed at will, and preserved for future rendition. A VISIC composer can thereforecreate new VISIC for real-time performance and artistic development.
series	other
email
more	http://www.generativeart.com/
last changed	2003/08/07 17:25

_id	caadria2022_100
id	caadria2022_100
authors	Oghazian, Farzaneh, Brown, Nathan and Davis, Felecia
year	2022
title	Calibrating a Formfinding Algorithm for Simulation of Tensioned Knitted Textile Architectural Models
doi	https://doi.org/10.52842/conf.caadria.2022.1.111
source	Jeroen van Ameijde, Nicole Gardner, Kyung Hoon Hyun, Dan Luo, Urvi Sheth (eds.), POST-CARBON - Proceedings of the 27th CAADRIA Conference, Sydney, 9-15 April 2022, pp. 111-120
summary	This paper presents an optimization-based calibration process for tuning a digital formfinding algorithm used with knitted textile materials in architectural tension structures. 3D scanning and computational optimization are employed to accurately approximate a physical model in a digital workflow that can be used to establish model settings for future exploration within a knit geometric typology. Several aspects of the process are investigated, including different optimization algorithms and various approaches to data extraction. The goal is to determine the appropriate optimization method and data extraction, as well as automate the process of adjusting formfinding settings related to the length of the meshes associated with the knitted textile behavior. The calibration process comprises three steps: extract data from a 3D scanned model; determine the bounds of formfinding settings; and define optimization variables, constraints, and objectives to run the optimization process. Knitted textiles made of natural yarns are organic materials and when used at the industrial level can satisfy DSG 9 factors to promote sustainable industrialization and foster innovation in building construction through developing sustainable architectural systems. The main contributions of this paper are calibrated digital models of knitted materials and a comparison of the most effective algorithms and model settings, which are a starting point to apply this process to a wider range of knit geometries. These models enhance the implementation and further development of novel architectural knitted systems.
keywords	Tensioned Knitted Textiles, Computational Design, Formfinding, Calibrating, Optimization, SDG 9
series	CAADRIA
email
last changed	2022/07/22 07:34

_id	cd92
authors	Pavlidis, Theo and Van Wyk, Christopher J.
year	1985
title	An Automatic Beautifier for Drawings and Illustrations
source	SIGGRAPH '85 Conference Proceedings. July, 1985. vol. 19 ; no. 3: pp. 225- 230. includes bibliography
summary	A method for inferring constraints that are desirable for a given (rough) drawing and then modifying the drawing to satisfy the constraints wherever possible, is described. The method has been implemented as part of an online graphics editor running under the UNIX operating system and it has undergone modifications in response to user input. Although the framework discussed is general, the current implementation is polygon-oriented. The relations examined are: approximate equality of the slope or length of sides, collinearity of sides, and vertical and horizontal alignment of points
keywords	drafting, computer graphics, algorithms
series	CADline
last changed	2003/06/02 13:58

_id	cdrf2022_337
id	cdrf2022_337
authors	Ping Chen, Chang Liu, and Hsin-Hsien Chiu
year	2022
title	Study on Optimization of Building Climate Adaptive Morphology in Cold Regions of China: Case of U-Shaped College Building
doi	https://doi.org/https://doi.org/10.1007/978-981-19-8637-6_30
source	Proceedings of the 2022 DigitalFUTURES The 4st International Conference on Computational Design and Robotic Fabrication (CDRF 2022)
summary	Proper design of building form will facilitate the use of climate environment in order to reduce the reliance of buildings on active equipment. This study takes the cold region of China as the research area, and Jinan city of Shandong province as a typical city in the cold region for specific research. The multi-objective optimization tool based on NSGA-II algorithm is used to optimize the opening angle, length of both sides and floor height of the building, and finally the optimal size range of the university teaching building under the influence of solar radiation heat gain in winter and summer is obtained, and the results show that for the U-shaped university teaching building, the parameters that affect the building performance more in the case of the east side opening are the length of the north side building and the rotation angle of the south side building, and the parameters that affect the performance more in the case of the west side opening are the length of the building on the south side.
series	cdrf
email
last changed	2024/05/29 14:03

_id	3674
authors	Richens, P.
year	1997
title	Image Processing for Urban Scale Environmental Modelling
source	Proceedings of the Intemational Conference Building Simulation 97 - Prague, 163-171
summary	If a map of a city is encoded as a Digital Elevation Model, it becomes amenable to image-processing software, such as the public-domain NIH Image application. Standard techniques can be used to measure plan areas and volumes and simple macros can be devised to measure perimeter length and wall areas. A macro for calculating shadow volumes is elaborated for the simulation of solar gains and daylight, including indirect lighting, leading to the possibility of an image-based urban-scale environmental model.
series	other
email
last changed	2003/04/23 15:50

_id	cc8e
authors	Richens, P.
year	1997
title	Image Processing for Urban Scale Environmental Modelling
source	Proceedings Fifth International IBPSA Conference: Building Simulation ’97 (Prague). International Building Performance Simulation Association
summary	If a map of a city is encoded as a Digital Elevation Model, it becomes amenable to image-processing software, such as the public-domain NIH Image application. Standard techniques can be used to measure plan areas and volumes and simple macros can be devised to measure perimeter length and wall areas. A macro for calculating shadow volumes is elaborated for the simulation of solar gains and daylight, including indirect lighting, leading to the possibility of an image-based urban-scale environmental model.
series	other
email
more	http://www.arct.cam.ac.uk/research/pubs/html/rich97b/
last changed	2000/03/05 19:05

_id	ecaade2011_079
id	ecaade2011_079
authors	Ryu, Han Soo; Lee, Kwon Hyoung; Ryu, Jung Rim; Choo, Seung Yeon
year	2011
title	Development of optimized geometry for low energy super tall office with BIM: Decision on ratio of lateral to longitudinal length and orientation in early design stage
doi	https://doi.org/10.52842/conf.ecaade.2011.811
source	RESPECTING FRAGILE PLACES [29th eCAADe Conference Proceedings / ISBN 978-9-4912070-1-3], University of Ljubljana, Faculty of Architecture (Slovenia) 21-24 September 2011, pp.811-818
summary	There are many researches to make low energy demanding building. Lots of them focus on facility systems and insulation performance of building materials. However, not only systematic solutions but also design solution can reduce building energy consumption. This study focuses on development of optimized geometry for super tall office building in Seoul, Korea. Specifically, ratio of lateral to longitudinal length and building orientation are main topics of this study because these are the most primitive and preceding factors deciding mass design. To analyze the energy efficiency of masses, energy simulation is necessary at the initial design stages. In this study, BIM and BIM based energy simulation tools are arranged to compare the alternatives.
wos	WOS:000335665500093
keywords	BIM; energy simulation; ratio of lateral to longitudinal length; orientation; super tall office building
series	eCAADe
email
last changed	2022/05/01 23:21

_id	cf2013_187
id	cf2013_187
authors	Schnabel, Marc Aurel
year	2013
title	Bi-tonal Architectural Narratives: Design Expressions in Manga
source	Global Design and Local Materialization[Proceedings of the 15th International Conference on Computer Aided Architectural Design Futures / ISBN 978-3-642-38973-3] Shanghai, China, July 3-5, 2013, pp. 187-197.
summary	Bi-tonal, or black and white architectural renderings are a common medium to present an architectural design that goes beyond factual representations of dimensions of length, width and height, materials, structure or technical details. They aim to engage their audience with visual aspects as well as emotional reactions. An architectural presentation that is based on a story-telling approach, akin to comic or manga style, adds intangible expressions, which allows communicating key elements as well as concepts, development and overall aspects of a design easily to laypersons and professional alike. This paper presents how designers develop a narrative of their architectural proposals using a bi-tonal visualization and story-telling methodology that communicates design to a wider audience.
keywords	Digital manga, architectural depiction, visual communication, story-telling
series	CAAD Futures
email
last changed	2014/03/24 07:08

_id	caadria2012_001
id	caadria2012_001
authors	Schnabel, Marc Aurel and Yingge Qu
year	2012
title	Digital manga virtual environment: Bi-tonal visualisation for texture-mapped virtual environments
doi	https://doi.org/10.52842/conf.caadria.2012.619
source	Proceedings of the 17th International Conference on Computer Aided Architectural Design Research in Asia / Chennai 25-28 April 2012, pp. 619–628
summary	Bi-tonal and non-photorealistic architectural depictions can be employed to develop a narrative that engages the reader with not only the visual aspects, but also other sensorial feelings. Architecture is subsequently not only represented through its factual dimensions of length, width and height, but is extended to intangible sensorial realms. This paper presents a modified rendering appearance of the graphical depiction of a Virtual Environments (VE) -software to communicate design akin to Japanese cartoon (manga) depictions whereby we expand the two-dimensional manga-images into a three-dimensional VE. The modified visualization can be used for storytelling and developing a narrative that professionals and laypersons alike easily can access, understand and interact. The bi-tonal depictions offer users to experience both, visual richness of the original model, as well as enhanced design communications.
keywords	Digital manga; virtual environments; architectural depiction; visual communication
series	CAADRIA
email
last changed	2022/06/07 07:57

_id	caadria2013_109
id	caadria2013_109
authors	Schnabel, Marc Aurel; Tian Tian Lo and Yingge Qu
year	2013
title	Manga Architecture – Developing Architectural Narratives
doi	https://doi.org/10.52842/conf.caadria.2013.771
source	Open Systems: Proceedings of the 18th International Conference on Computer-Aided Architectural Design Research in Asia (CAADRIA 2013) / Singapore 15-18 May 2013, pp. 771-780
summary	Bi-tonal and non-photorealistic architectural depictions can be employed to develop a narrative that engages the reader with both visual aspects and other emotional reactions. Using this story-telling approach to communicate architectural design is subsequently not only represented through its factual dimensions of length, width and height, materials, structure or technical details, but is extended to intangible sensorial realms, which gains special value for laypersons and professional alike. This paper presents how architects and laypersons develop a narrative of their architectural design proposals or concepts using a visualisation and story-telling tools that generates designs akin to Japanese cartoons or manga. Our paper presents the methodology, the instruments used and highlights on the base of representative samples of how narrative bitonal depictions of architecture contributes to the overall understanding of an architectural design and how non tangible factors aid the designers in their communication.
wos	WOS:000351496100080
keywords	Digital manga, Architectural depiction, Visual communication, Story-telling
series	CAADRIA
email
last changed	2022/06/07 07:57

_id	7261
authors	Seebohm, Thomas and Chan, Dean
year	2001
title	The Design Space of Schematic Palladian Plans for Two Villa Topologies
doi	https://doi.org/10.52842/conf.acadia.2001.156
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. 156-165
summary	Given the plan topology for the layout of the rooms of a Palladian villa, we ask what is the design space of possible Palladian plans. What does this space look like in terms of dimensions and proportions? Two plan topologies are examined to throw some light on this question. One is that of the Villa Angarano while the other is that of the Villa Badoer. A Maple program was written for each topology to solve the equations for room proportions, to cycle through the possible proportional spacing of the underlying Tartan grid, and to plot out the possible plans in the design space. The programs eliminate from the design space those alternatives that violate basic Palladian constraints, such as no room having an aspect ration greater than 2:1, and such additional constraints that we and other authors have found. A selection of the plotted output of plans in the design space for each topology is presented as well as threedimensional plots showing the number of plan alternatives in different regions of the critical parameter space, namely, the length and aspect ratio of the plan. We believe that this is the first instance of enumerated Palladian plans which goes beyond topology to examine issues of dimension and proportion. One conclusion is that one cannot scale a Palladian plan topology to fit any set of overall dimensions.
keywords	Palladio, Design Space, Villa, Plans, Rule-Based Generation
series	ACADIA
email
last changed	2022/06/07 07:56

_id	caadria2021_194
id	caadria2021_194
authors	Sun, Chengyu, Li, MengTing and Jiang, Hanchen
year	2021
title	Developing an Automatic Code Checking System for the Urban Planning Bureau of Huangpu District in Shanghai
doi	https://doi.org/10.52842/conf.caadria.2021.1.291
source	A. Globa, J. van Ameijde, A. Fingrut, N. Kim, T.T.S. Lo (eds.), PROJECTIONS - Proceedings of the 26th CAADRIA Conference - Volume 1, The Chinese University of Hong Kong and Online, Hong Kong, 29 March - 1 April 2021, pp. 291-300
summary	As Chinese cities entering a so-called organic renewal era, building projects runs with much more constraints from high-density and high-rise surroundings. Such a situation makes the technical review in any urban planning bureau time-consuming and error-prone, which conflicts with the developers profits and citizens rights. This study introduces a preliminary system being developed for the planning bureau of Huangpu District, Shanghai. It has covered 21 code items among 44 computational ones of the local planning codes last year, which automatically generates technical reviews upon developers submissions. Due to the feasible level of BIM application in domestic projects, a set of strategic approaches, such as the standardization of CAD drawings and the reconstruction of an internal building information model, are adopted rather than developing the system on any BIM platform directly. Two examples of technical reviews about distance-checking between buildings and length-checking of facades are demonstrated, in which officers reached confidential judgments in seconds rather than several days conventionally.
keywords	Planning Constraints; Code Checking; 3D Reconstruction; Design Automation; Building Information Model
series	CAADRIA
email
last changed	2022/06/07 07:56

_id	acadia13_261
id	acadia13_261
authors	Tibbits, Skylar; Falvello, Ana
year	2013
title	BioMolecular, Chiral and Irregular Self-Assemblies
doi	https://doi.org/10.52842/conf.acadia.2013.267
source	ACADIA 13: Adaptive Architecture [Proceedings of the 33rd Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) ISBN 978-1-926724-22-5] Cambridge 24-26 October, 2013), pp. 267-268
summary	As disciplines converge and new capabilities are developed for programmable materials and self-assembly across length scales and industrial applications, designers will need new models for understanding the fundamental principles within this new paradigm. This paper outlines the key ingredients for self-assembly through a number of recent projects including the BioMolecular Self-Assembly and Chiral Self-Assembly projects. Further possibilities of non-deterministic self-assembly will be highlighted through asymmetrical units, nucleus models and hierarchical assemblies. Finally, opportunities for high-yield self-assembly and future applications for manufacturing and construction scenarios will be outlined. Self-assembly offers a glimpse into a future world of highly programmable, intelligent materials that promise far more adaptive, resilient and efficient built environments.
keywords	next generation technology, self-assembly, programmable materials, chirality, non-deterministic assembly
series	ACADIA
type	Normal Paper
email
last changed	2022/06/07 07:58

_id	ascaad2022_047
id	ascaad2022_047
authors	Tu, Han; Yang, Chunfeng
year	2022
title	Mindful Space in Sentences: A Dataset of Virtual Emotions for Natural Language Classification
source	Hybrid Spaces of the Metaverse - Architecture in the Age of the Metaverse: Opportunities and Potentials [10th ASCAAD Conference Proceedings] Debbieh (Lebanon) [Virtual Conference] 12-13 October 2022, pp. 713-730
summary	Spatial emotions have played a critical role in visual-spatial environmental assessment, which can be assessed using bio-sensors and language description. However, information on virtual spatial emotion assessment with objective emotion labels and natural language processing (NLP) is insufficient in literature. Thus, designers’ ability to assess spatial design quantitatively and cost effectively is limited before the design is finalized. This research measures the emotions expressed using electroencephalograms (EEGs) and descriptions in virtual reality (VR) spaces with different parameters. First, 26 subjects experienced 10 designed virtual spaces with a VR headset (Quest 2 device) corresponding to the different space parameters of shape, height, width, and length. Simultaneously, the EEG measured the emotions of the subjects using four electrodes and the five brain waves. Second, two labels – calm and active – were produced using EEGs to describe these virtual reality spaces. Last, this labeled emotion dataset compared the differences among the virtual spaces, human feelings, and the language description of the participants in the VR spatial experience. Experimental results show that the parameter changes of VR spaces can arouse significant fluctuations in the five brain waves. The EEG brain wave signals, in turn, can label the virtual rooms with calm and active emotions. Specifically, in terms of VR spaces and emotions, the experiments find that more relative spatial height results in less active emotions, while round spaces arouse calmness in the human brain waves. Moreover, the precise connection among VR spaces, brain waves in emotion, and languages still needs further research. This research attempts to offer a useful emotion measurement tool in virtual architectural design and description using EEGs. This research identifies potentials for future applications combining physiological metrics and AI methods, i.e., machine learning for synthetic design generation and evaluation.
series	ASCAAD
email
last changed	2024/02/16 13:29

_id	ecaade2024_19
id	ecaade2024_19
authors	Vele, Jiri; Melter, Oto; Hvizdal, Ales; Achten, Henri; Citek, David
year	2024
title	Enhancing 3D Concrete Printing Buildability with Non-Planar Layering
doi	https://doi.org/10.52842/conf.ecaade.2024.1.175
source	Kontovourkis, O, Phocas, MC and Wurzer, G (eds.), Data-Driven Intelligence - Proceedings of the 42nd Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2024), Nicosia, 11-13 September 2024, Volume 1, pp. 175–182
summary	This paper explores improvements in 3D Concrete Printing (3DCP) buildability using non-planar layering. Our proof-of-concept experiment validates the assumption that non-planar layering enhances the buildability of overhangs and vaults. To test this, we designed an object comprising two columns with a Gothic arch and overhangs on the upper part, resulting in a continuous wall at the top. The object measures approximately 1.5 meters in length, 400 mm in width, and 700 mm in height. The design was executed to maintain the centre of mass in the centres of the columns across all print layers, mitigating unwanted deformation. The overhang angle ranged continuously from 0 to 70 degrees. The object was modelled in Rhinoceros software, and G-code for both planar and non-planar printing was generated in Grasshopper. Both samples were designed to use the same amount of material, print length, and print time. We printed these using a cementitious mixture with plastic fibres and an additional accelerant mixed in the nozzle on a gantry printer. Throughout the printing process, we conducted careful observation and monitoring to detect any instances of buckling or collapse. Post-print measurements were carried out to evaluate the deformation of the printed objects, revealing a reduction in deformation with non-planar printing. This paper discusses the analysis of results and proposes a workflow for future data preparation for non-planar slicing.
keywords	3D Concrete Printing, Non-Planar Slicing, Buildability Improvement, Vaults
series	eCAADe
email
last changed	2024/11/17 22:05

_id	caadria2017_009
id	caadria2017_009
authors	Yang, Xuyou, Koh, Shawn Jyh Shen, Loh, Paul and Leggett, David
year	2017
title	Robotic Variable Fabric Formwork
doi	https://doi.org/10.52842/conf.caadria.2017.873
source	P. Janssen, P. Loh, A. Raonic, M. A. Schnabel (eds.), Protocols, Flows, and Glitches - Proceedings of the 22nd CAADRIA Conference, Xi'an Jiaotong-Liverpool University, Suzhou, China, 5-8 April 2017, pp. 873-882
summary	Casting is one of the most widely used construction techniques. Complex geometries produced via computational design processes are not easily achievable through traditional rigid formwork and are subject to increase material waste. More suitable casting techniques are required to efficiently represent digital design output. This paper presents a variable fabric formwork developed to work in conjunction with a 6-axis robotic arm for casting doubly curved panels based on hyperbolic paraboloid geometry. The variable formwork is designed to be extendable in length and width so it is able to produce a wide range of outcome within a single formwork. The interface established in the workflow allows the physical formwork and digital design to influence each other. This variable fabric formwork reduces construction waste and is a more sustainable method of casting complex geometries.
keywords	Digital fabrication; Robotic production; fabric casting
series	CAADRIA
email
last changed	2022/06/07 07:57

_id	ijac202321105
id	ijac202321105
authors	Zargar, Seyed Hossein; Jaleh Sadeghi; Nathan C. Brown
year	2023
title	Agent-based modelling for early-stage optimization of spatial structures
source	International Journal of Architectural Computing 2023, Vol. 21 - no. 1, pp. 84–99
summary	Agent-based modelling (ABM) is a complex problem-solving approach that can be employed in early-stage parametric design, and certain design applications may benefit from such a bottom-up strategy. This research investigates the potential of ABM for structural design optimization. A case study of a form-found cantilevered truss is presented that has a doubly curved shape over a regular grid, resulting in individual members with different lengths across the structure. It is hypothesized that an agent-based approach might generate an irregular grid of similar or better structural performance, but with more uniform length of individual elements. This approach could be useful when designing a global structural form from a kit of parts or adaptively reusing a disassembled existing structure with regular member lengths. A series of ABM simulations are conducted with different hyperparameters, and the generated designs are compared to the original formfound shape in terms of structural performance.
keywords	Agent-based modelling, design space exploration, structural optimization, form-finding, spatial structures
series	journal
last changed	2024/04/17 14:30

_id	37c2
authors	Ahmad Rafi, M.E.
year	1999
title	Visualisation of Design Using Animation for Virtual Prototyping
doi	https://doi.org/10.52842/conf.ecaade.1999.519
source	Architectural Computing from Turing to 2000 [eCAADe Conference Proceedings / ISBN 0-9523687-5-7] Liverpool (UK) 15-17 September 1999, pp. 519-525
summary	Although recent technology in time-based representation has vastly improved, animation in virtual prototype design field remains the same. Some designers invest a huge amount of money in the latest visualisation and multimedia technology and yet may create even worse animation. They often cramp sequences resulting in many viewers failing to interpret the design positively as they miss a lot of vital information that explains the design. This paper basically reports the importance of film-making understanding for producing good virtual prototype animation. It will be based on a part of a research project on the use of time-based media in architectural practices. It also includes an empirical analysis of several architectural-based documentary films (including an interview with the film director) and past and present computer animation. This paper then concludes with recommendations of good techniques for making animated visualisation relative to the stage at which the animation is produced for better design decision.
keywords	Virtual Prototype, Animation, Time-Based, Film-Making
series	eCAADe
email
last changed	2022/06/07 07:54

_id	5236
authors	Arciszewski, T., Michalski, R.S. and Dybala, T.
year	1995
title	STAR methodology-based learning about construction accidents and their prevention
source	Automation in Construction 4 (1) (1995) pp. 75-85
summary	This paper presents the results of a feasibility study concerning the application of STAR-methodology-basedmachine learning to construction accidents and their prevention. A ten-stage knowledge acquisition process is presented and its individual stages described. Knowledge about construction accidents was acquired using a collection of 225 examples, based on actual accidents records. Inductive learning with a system based on the STAR-methodology was employed. This system was used in both the generalization and specialization modes of operation. The decision rules obtained are complex, but their interpretation is clear and they seem to be consistent with the present understanding of causal relationships between accident results and various factors affecting them. Also, the rules were verified using average overall and omission empirical error rates, which were calculated as average for three randomly determined sequences of examples. These error rates were calculated for all seven steps in the machine learning process, and were used to construct learning curves for both error rates. The relationships between error rates and the number of examples used for learning are analyzed, and coefficients of linear regression given and discussed. The 225 examples used were found to be grossly insufficient to produce reliable knowledge about accidents and therefore a large study is postulated which would involve the collection of a larger number of construction accident records. In general, our study demonstrated the feasibility of machine learning in acquiring knowledge about construction accidents.
keywords	Construction accidents and their prevention; Knowledge acquisition; Machine learning; Multi-stepmachine learning process
series	journal paper
more	http://www.elsevier.com/locate/autcon
last changed	2003/06/02 09:31

For more results click below: