Welcome
Digital library of construction informatics
and information technology in civil engineering and construction
 

Works 

Search Results

Facilitated by the SciX project

Hits 1 to 10 of 22

Aminah Robinson Fayek and Krista Marsh

A Decision-Making Model For Surety Underwriters In The Construction Industry Based On Fuzzy Expert Systems

Abstract:

Keywords:

DOI:

Full text: content.pdf (449,537 bytes) (available to registered users only)

Series: w78:2006 (browse)
Cluster:
Class:
Similar papers:
Sound: N/A.


Ciftcioglu O, Durmisevic S, Sariyildiz S

Building design support by hierarchical expert networks

Abstract: "Rapid advances in parallel processing technologies gave essential impetus to intelligent information processing, which became the driving source of an emerging technology known as soft computing. This calls for intelligent systems that are able to process information which may be complex, uncertain even incomplete or contradictory. In this context, neural networks and fuzzy logic are the essential tools. Considering the merits of each approach separately, most suitable computational intelligence method can be used for a specific application. Additionally, the combination of these methods can provide enhanced information processing for decision-making with enhanced reliability. For building design, the computational intelligence system use a knowledge base formed by means of neural network and fuzzy logic (neuro-fuzzy) techniques, from a building design database. The application of such a system to a building design task was preliminarily demonstrated earlier [1]. The present research describes a systematic neural fuzzy modelling of data that form a knowledge base in a hierarchical form (s.figure below). Each sub-knowledge base represents a local expert, being level-one expert and the association of local experts forms a more comprehensive expert that becomes a global domain expert as level-two. The association of the experts is accomplished by means of fuzzy-logic-driven gating network that performs, the information handling as required. Although, the present paper describes two-level hierarchical experts as local and global, the associations can be done in more subtle form, i.e., in more than two steps so that the level of experts can be categorised in multi-level form. In such more complex structures, multi-level experts require related gating network that could similarly be designed. The building design support system with the expert network developed, as a whole, is generic enough for decision-makings with a novel systematic approach concept using appropriate database. Accordingly, the research deals with a particular architectural building design with efficiency and consistency features using the hierarchical expert network system described [1] Ciftcioglu O, Sariyildiz S. and Veer P. v.d., 1998 , Information Ordering for decision support in building design, D and DSS, Design and Decision Support Systems 4th International Conference on Design and Decision Support Systems in Architecture and Urban Planning, Castle Vaeshartelt, Maastricht, July 26-29, The Netherlands"

Keywords:

DOI:

Full text: content.pdf (519,213 bytes) (available to registered users only)

Series: w78:2000 (browse)
Cluster: papers of the same cluster (result of machine made clusters)
Class: class.analysis (0.064229) class.synthesis (0.019630) class.man-man (0.013152)
Similar papers:
Sound: read aloud.

Permission to reproduce these documents have been graciously provided by Icelandic Building Research Institute. The assistance of the editor, Mr. Gudni Gudnason, is gratefully appreciated


Ciftcioglu O, Durmisevic S, Sariyildiz S

Soft computing in construction information technology

Abstract: Purpose: With this paper a data driven model of knowledge representation for use in construction information technology (CIT) is introduced as a novel implementation and it is effectively implemented by artificial intelligence methods. Although for CIT knowledge base systems as a general framework is available where the user can place the information. Such systems are eventually mere data warehouses or in a more sophisticated form they are decision support systems in the form of rule based expert systems. However, in the case of a framework structure to organise the knowledge base is difficult and cumbersome task to establish an effective product. In the case expert systems, the inference is deductive and therefore the effectiveness of the system is limited to the prescribed rules. Therefore in place of high level data base management software like Prolog , the integration of new computational information processing methods and technologies into CIT would be much informative and therefore they are much effective and finally desirable. From the viewpoint of computation, CIT the data are rather soft requiring special methods and techniques to deal with. In this respect, computational intelligence is one of the emerging technologies, which provides CIT with ample possibilities and techniques for the enhancement of CIT products. Computational intelligence is a part of artificial intelligence (AI) and can be defined as a branch of soft computing methodologies including Expert Systems, Fuzzy Logic, Artificial Neural Networks and Evolutionary Computation. Methodology: For the CIT data soft computing methods are invoked. Soft Computing is an emerging approach to computing which parallels the remarkable ability of the human mind to reason and learn in an environment of uncertainty and imprecision. In plain terms, it is the processing of uncertain information with the methods, methodologies, and paradigms of artificial NN, fuzzy logic and evolutionary algorithms. The equivalence of neural networks and fuzzy logic applications is well established. However, the effectiveness of either method is still dependent on the application itself. Each method has its strong merits. However, in general, best performance is obtained when both methods are used in hybrid form. Especially neural system can cope with complex systems while it is relatively difficult for fuzzy systems. On the contrary, it is easier to deal with linguistic variables by fuzzy systems. Such a hybrid model is implemented in the knowledge model accomplished. Results: A novel concept of soft computing in CIT is introduced using actual building design data for design evaluation. The knowledge base contains all the local and global information and their inherent relationships among themselves. The knowledge representation is performed by means of a series of fuzzy systems having their both fuzzy input space and output space. The associations between the spaces are established by learning techniques of AI using the data at hand. Such an 'intelligent' knowledge base can make inference resulting in 'intelligent' due outcomes, which are not explicitly coded, in advance. In other words this is an inductive and computational inference for decision-making compared to conventional knowledge base systems where inference is deductive prescribed by rules. Conclusions: The soft computing in CIT is an important step for processing the relevant effectively and efficiently. In this respect, the paper describes ongoing advanced research and its verifications by actual data at hand.

Keywords:

DOI:

Full text: content.pdf (204,669 bytes) (available to registered users only)

Series: w78:2001 (browse)
Cluster: papers of the same cluster (result of machine made clusters)
Class: class.analysis (0.064489) class.synthesis (0.025964) class.deployment (0.019220)
Similar papers:
Sound: read aloud.

Permission to reproduce these documents have been graciously provided by CSIR Building and Construction Technology. The assistance of the editors, Mr. Gustav Coetzee and Mr. Frances Boshoff, is gratefully appreciated.


Ciftcioglu O

Classification of construction information with fuzzy attributes

Abstract: Construction industry deals with various kinds of diverse information, which have to be synergistically handled. Information can be engineering data from exact sciences as well as linguistic or qualitative data from soft sciences. In such a wide spectrum to deal with the context dependent information optimally in perspective is a formidable task. The loss of benefit from available information is reflected on the cost effectiveness and efficiency of the construction. Referring to the complexity of this information, intelligent technologies can be of important help to assess information in a particular context in perspective for enhanced assessment of the construction process while it is in progress. In this respect, role of intelligent technologies, in particular, fuzzy logic, neural network and evolutionary search algorithms, in dealing with construction information is discussed and exemplified. In particular, in fuzzy logic terms, classification of construction information with fuzzy attributes/semantic labels is described.

Keywords:

DOI:

Full text: content.pdf (287,397 bytes) (available to registered users only)

Series: w78:2003 (browse)
Cluster:
Class:
Similar papers:
Sound: N/A.

Permission to reproduce these papers has been graciously provided by the University of Auckland. The assistance of the editor who provided the full texts and the structured metadata, Dr. Robert Amor, is gratefully appreciated.


Durmisevic S, Ciftcioglu O, Sariyildiz S

An application of neural network in post-occupancy evaluation of underground stations

Abstract: "The architectural and construction design deals very often with the word quality. This term is so vague and broad that the main difficulty arises if one needs to determine its aspects. It is rather simple to deal with the quantifiable building standards. The problem is how to demystify and thereafter integrate this fuzzy concept of quality into design. As an example we will use underground stations as a design problem area for two reasons. First of all, these spaces are rather young structures that have a high potential in the future. The efficiency of underground transport and importance of multiple space usage in the densely built urban areas are only some benefits that these spaces can offer. But yet many realized underground projects were not satisfactory to the users. Second reason lies in a fact that these spaces have their own limitations. Some qualities that are so obvious for the aboveground buildings, such as daylight or view, are rather difficult to obtain in underground spaces. Therefore, in these spaces the word quality is even more sensitive. But the literature that the architects can consult regarding these problems is rather scattered and difficult to obtain. One of the reasons is a lack of detailed documentation on actual applications of the theories followed by the research results and applied techniques. In this paper we used the AI technique, a Neural Network, for data analysis. The main objective of this paper is to develop a Support Model that will enable quality measurement of underground spaces in a systematic way. In order to avoid the ad-hoc design solutions for underground spaces, there is a need for systematic approach to their design. In such way the intuitive approach to problem solving can be minimized. This paper deals with following topics: 1. aspects that determine the quality of space 2. classification of psychological and spatial aspects 3. development of conceptual framework 4. application of Neural Network for post-occupancy evaluation 5. results and endeavor design guidelines First three topics will deal with criteria definition, which were necessary for design of the experimental part of a research. The experimental research, which was carried out at the site of one underground station, provided the necessary data. The main emphasis of the paper will be on Neural Network application (topic 4), which will be used to treat the data gathered on underground station. The main objective is to verify the consistency of the outcomes against the predefined criteria."

Keywords:

DOI:

Full text: content.pdf (1,096,055 bytes) (available to registered users only)

Series: w78:2000 (browse)
Cluster: papers of the same cluster (result of machine made clusters)
Class: class.analysis (0.050429) class.impact (0.013741) class.social (0.008794)
Similar papers:
Sound: read aloud.

Permission to reproduce these documents have been graciously provided by Icelandic Building Research Institute. The assistance of the editor, Mr. Gudni Gudnason, is gratefully appreciated


Durmisevic S, Ciftcioglu O, Sariyildiz S

Knowledge modelling of 'soft' data in architectural design

Abstract: IT Context: Information technologies are at present used in various disciplines to address issues such as information processing, data mining, knowledge-modelling etc. Its final goal is to provide necessary aid to professionals during decision-making process. This raises already few questions such as, what type of data is considered and are there some new emerging technologies that can improve knowledge modelling and therefore provide better decision support to the professionals. Design professionals are very often confronted with soft data that they somehow need to interpret and finally integrate in a design. Situations dealing with the numerical data may occur quite naturally in exact sciences like engineering sciences, life sciences etc. However, the quantities subject to consideration in soft sciences are often qualitative rather than quantitative so that we relate to that type of data as 'soft' data. As an example, in such cases, the quantities may be linguistic so that such quantities have to be somehow expressed in numerical form for treatment by conclusive numerical analysis methods. Objectives: The architectural design task is one example having linguistic qualities as priory design information. This is especially the case when qualities of certain space are discussed, like for example in post occupancy evaluation of the buildings, where the relationship between spatial characteristics and psychological aspects plays an important role. Expressions such as: bright colour, light room, large space are some of these examples and therefore a special method is needed for representation and processing of such vague expressions and concepts. Better understanding of these concepts is necessary so that the knowledge can be modelled in a proper way. Methodology: The analyses are performed by means of soft computing methods. The data subject to analysis and later to knowledge modelling belongs to an underground station that is already being used. For this purpose, the data on psychological aspects are obtained via comprehensive inquiry of the users of underground station. For the analysis, the linguistic information is firstly converted to terms in fuzzy logic domain and after appropriate treatment, the data analyses are carried out and the results are expressed in most comprehensible form for design assessments. Such conversions are referred to as fuzzification and defuzzification, where the data are expressed in numerical form and therefore become convenient for mathematical treatment. Conclusions: Referring to the complexity of task in dealing with the soft data as well as dealing with soft computing, the paper first identifies the source of these complexities referring to the architectural design tasks. Following this, a soft computing analysis method based on one case study will be presented, whereby the focus will be on knowledge modelling. Finally, the results of the analyses together with the conclusions regarding the observed effectiveness of the approach are presented.

Keywords:

DOI:

Full text: content.pdf (242,586 bytes) (available to registered users only)

Series: w78:2001 (browse)
Cluster: papers of the same cluster (result of machine made clusters)
Class: class.retrieve (0.036108) class.impact (0.013187) class.analysis (0.007731)
Similar papers:
Sound: read aloud.

Permission to reproduce these documents have been graciously provided by CSIR Building and Construction Technology. The assistance of the editors, Mr. Gustav Coetzee and Mr. Frances Boshoff, is gratefully appreciated.


H Kim, B Elhamim, H Jeong, C Kim, H Kim

On-Site Safety Management Using Image Processing and Fuzzy Inference

Abstract:

Keywords:

DOI:

Full text: content.pdf (253,485 bytes) (available to registered users only)

Series: w78:2014 (browse)
Cluster:
Class:
Similar papers:
Sound: N/A.


Hirokane M, Furuta H, Nakai S, Mikumo Y

Rough-set-based diagnosis model of slope-failure danger level

Abstract: Slope-failure is caused by complex working of various factors such as geology, rainfall, etc. If the fuzzy theory is applied to all of these factors, results will often be of short reliability. In this paper, we present a method to acquire, removing inconsistencies, the minimum knowledge necessary to deduce reliable results from actual cases where engineers of long experience diagnosed the danger levels. Next comes a method of constructing a knowledge base for an expert system, such acquired knowledge used. Then, a method to check the accuracy and reliability of acquired knowledge and evaluate the knowledge base is described.

Keywords:

DOI:

Full text: content.pdf ( bytes) (available to registered users only)

Series: ecce:1997 (browse)
Cluster:
Class:
Similar papers:
Sound: N/A.


Hsin-Lung Liu, Shin-Hua Lin, and Jia-Ruey Chang

The GA And Fuzzy MCDM Applied In Financial Planning For A Construction Project

Abstract:

Keywords:

DOI:

Full text: content.pdf (590,902 bytes) (available to registered users only)

Series: w78:2006 (browse)
Cluster:
Class:
Similar papers:
Sound: N/A.


Ibbs, C.W., Chang, T.C. and Echeverry, D.

Schedule Generation: A Fuzzy Representation of Trade Interaction Constraints

Abstract:

Keywords:

DOI:

Full text: content.pdf (3,566,879 bytes) (available to registered users only)

Series: w78:1990 (browse)
Cluster:
Class:
Similar papers:
Sound: N/A.


For more results click below:

 

hosted by University of Ljubljana



includes

W78




© itc.scix.net 2003
this is page 1 show page 2 show page 3 Home page of this database login Powered by SciX Open Publishing Services 1.002 February 16, 2003