Introduction / overview
As we know, knowledge-based systems (KBS) and multimedia technology (MMT) have been widely used in many industries for years to solve complicated problems. This integration leads to intelligent multimedia, provides a powerful tool for expertise-intensive training. Knowledge-based systems (KBS) attempt to replicate or imitate the knowledge and reasoning processes of human experts on some specialized tasks. It represents perhaps the most matured and fruitful branch of artificial intelligence technologies. Because of their potential to enhance productivity and to augment workforces in many specialized areas where human experts are becoming increasingly difficult to find and retain. Extensive expertise and years of experience is needs in many tasks in construction demand. This training is hardly to be documented in the literature. This training is aim to demonstrate the feasibility and potential benefits of using intelligent multimedia for expertise-intensive training in construction. A KBS is the embodiment within a computer of a knowledge-based component from an expert skill in such a form that the system can offer intelligent advice or take an intelligent decision about a processing function.
A KBS consists of three components such as a knowledge base, an inference engine and a dynamic store holding temporary data. A knowledge base which contains expertise and knowledge about a specific subject domain, represented explicitly in forms of such as facts and rules. An inference engine which uses one or more reasoning mechanism, such as backward and forward chaining to utilize the knowledge. A dynamic store holding temporary data which can explain the reasoning process and the rule used.
Multimedia technology presents information with text, illustrations, photos, narration, sounds, animations, and videos, which has the most successful application in education and training. Multimedia is widely regarded as a value-added technology to educational programmes. There are an increasing number of multimedia applications in the field of education and training. Multimedia is appropriate for these applications primarily because of the flexibility the technology offers over conventional computer-aided learning. It allows non-linear access to information and offers more effective navigation that is geared to learners??™ needs and interests. It also offers greater exploration of relevant and divergent information.
When the above two technologies are integrated, its capability as a training tool is further enhanced. Users of the integrated system can access complex knowledge and timely interactions through all possible sensory means such as sound, video, image, animation, and text. In other words, the integrated system will have the best of both worlds. Because of the synergistic effect of integrating the two technologies, the potential of using such integrated systems for training is very large; this market in the USA alone is estimated to be US$200 billion per year. Findings from empirical research have also demonstrated that integrated instruction systems do as well or better than courses delivered by traditional instruction methods.
Most construction tasks are expertise intensive, demanding extensive training and years of practical experience. Strategic cost estimate is one taste, which provides clients with early advice on the cost implications of their requirements at early design stages. It also helps the architects and other design professionals to address the imbalance of costs among various building elements of projects. Because of the small amount of information available at the early design stage, a large amount of cost data have to be deployed, and a large number of reasonable assumptions have to be made in order to arrive at an appropriate estimate. Owing to the extensive amount of knowledge and expertise required for preparing the estimate at this early stage, both educationalists and trainees are facing difficulties in teaching and learning this topic. This complex and problematic task presents a good case for developing an intelligent multimedia training package.
This paper uses an integrated system for teaching strategic cost estimate as a case study to illustrate the viability and potential benefits of using the intelligent multimedia system as a tool for expertise-intensive training in the construction industry. It reveals that intelligent multimedia can be deployed to meet the challenges of accurately representing the highly complex expertise in the knowledge base, and facilitating the learning process of the trainees through the highly interactive and non-linear access to the explicit and transparent expertise.
The system was verified by seeking comments from the participating experts throughout the development process. Because of the training nature, the system has been validated by a total of 65 second-year students majoring in quantity surveying and construction technology, to ensure that the system is designed properly to fit for its purpose. This was done through an extensive trial run of the system by the students, followed by a questionnaire survey to collect their comments. The validation proved that the questions raised by the system and the associated explanations are highly relevant, the major functionalities provided by the system are very useful, the details of cost breakdown and the report generated by the system are appropriate, and after all, the system is a very good learning tool which provides speedy interactions and is user friendly.
Cost model embedded
The method that did authors use is cost model embedded. Cost model is the important of most construction cost-estimating methods. Its provide advice in a more informed and reliable manner. For example, empirical models, regressive models, and heuristic models are used in the industry but a deterministic model was widely used by quantity surveyors in the local industry because it depends only on the quantity of resources to be used in construction. There are many building elements involved in a building project but only major elements that have large cost impact are selected to arrive at the total cost of the building project. These elements are substructure (including foundation and earth removal), superstructure (including external walls and internal finishing), services, preliminaries, and contingencies.
System structure (KBCES)
Another method use is KBCES. KBCES is knowledge-based cost estimation system, consists of six main modules. The system was developed by using a KAPPA-PC. It is an object-oriented programming package that offers objects and methods as the building blocks of every application. The programming language used inside the package is an interpreted language designed for rapid prototyping. It comes with a source level debugger, and the application can be compiled into C. It offers a GUI development tool and an inference engine for rule-based reasoning. In addition to its set of predefined images, KAPPA-PC applications can use Visual Basic (VBX) controls with a simple plug-in interface, and can access external databases easily and flexibly. These functionalities made the package ideal for development of an intelligent multimedia training system.
The system consists of knowledge of geo-technical information, design criteria, ratios and factors, and cost data for office buildings. By using the deterministic cost model, the system asks a few questions regarding site situation, functionality and appearance of a building. They also implement the required codes and regulations for office buildings, assess the complexity of the project, and decide the standards of finishes required.
Knowledge base and inference nets
The system consists of eight main inference nets that are interrelated. The construction floor area (CFA) is the first figure to be determined after user requirements have been identified. CFA is used to calculate the floor area and the number of storeys above and below ground level. Each of these eight inference nets has been extended to a comprehensive sub-net to address detailed questions and factors relating to a specific area of a project. For example, illustrates the detailed inference net for foundation cost. It shows that factors such as CFA, water problem, and rock problem will affect the type of foundation and the quantity of foundation which, in conjunction with the rate of foundation, will affect the foundation cost.
The system used is primarily a KBS application that employs MMT. This two technologies are working together efficiently to overcome their inherent limitations. The KBS provides the structure, control, knowledge representation and inference mechanism that reflect human expertise in a narrow domain. MMT, on the other hand, offers the greater flexibility, complexity, and depth associated with conceptual relationships and non-linear access to information, which enhances communication with the end-users. For example in the system, when the users are asked to provide requirement for the external wall, various images will be displayed on the screen to illustrate different kinds of external walls available. Another explanation can also be given through voices, even video clips. This feature significantly improved the interface with the end-user.
One of the main strength of the proposed system over the traditional training methods (such as videotapes and training manuals) is the high degree of intelligent interactions, which get the trainees/students closely involved and highly motivated.
Make trainees being motivated by the interactions with the system. This will make trainees see the impact of a changing cost factor on the total cost of project very quickly and efficiently in order to reduces and save the total cost budget. This is not normally possible for the manual training system. Using intelligent multimedia for training will make learning is more interesting and flexible. Trainees can use and learn about construction using the system by their own pace and at preferred and dispersed location. The stimulation provide can optimize trainees??™ learning, attention and performance.
Another strengths is, the learning pattern progress is improved. The trainees can improve their understanding of the concepts and components of knowledge-base systems, an understanding of the development of the information technology and its potential effect on the construction industry.
Although using intelligent multimedia systems for expertise-intensive training in construction is superiority over traditional training methods such as videotapes, but it is required cost to buy the intelligent multimedia systems for training construction. On the other hand, not all trainees are computer literate and know how to use the intelligent multimedia systems.
It can be conclude that using intelligent multimedia systems for expertise-intensive training in construction is superior over traditional training methods such as videotapes. The most significant benefits of using the system are the high degree of interactions between trainees and the system, which motivates learning and the flexibility of learning through the non-linear access to the explicit and extensive amounts of knowledge and expertise. Nowadays, many construction-related training programmes are practical in nature, which require extensive amounts of practical knowledge and years of experiences. As demonstrated through the case study, intelligent multimedia training systems can be used successfully in training in these areas. In some cases, when usability is improved, a training package can be further developed for commercial uses to improve productivity or quality of services, which can offset the development cost of the training system.
Agius, H.W., Angelides, M.C. (1999), “Developing knowledge-based intelligent multimedia tutoring systems using semantic content-based modelling”, Artificial Intelligence Review, Vol. 13 No.1, pp.55-83.
Angelides, M.C. (1995), “Developing hybrid intelligent tutoring and hypertext systems”, The New Review of Hypermedia and Multimedia, Vol. 1 pp.67-106.
Angelides, M.C., Dustdar, S. (1997), Multimedia Information Systems, Kluwer Academic Publishers, Boston, MA, .
Ashworth, A. (1994), “Cost planning”, Cost Studies of Buildings, 2nd ed., Longman Scientific & Technical, London, .
Bainbridge, S. (1991), “Classroom training versus interactive videodisc/CBT training: the final word”, SALT Proceedings: Interactive Instruction Delivery, pp.34-40.
Bielawski, L., Lewand, R. (1991), “Intelligent systems defined”, Intelligent Systems Design: Integrating Expert Systems, Hypermedia, and Database Technologies, John Wiley & Sons, New York, NY, pp.17-62.
Brandon, P., Basden, A., Hamilton, I., Stockley, J. (1988), The Strategic Planning of Construction Projects: Application of Expert Systems to Quantity Surveying, Quantity Surveyors Division of the Royal Institution of Chartered Surveyors in Collaboration with the University of Salford, .
Day, H.I., Berlyne, D.E. (1971), “Intrinsic motivation”, in Lesser, G. (Eds),Psychology and Educational Practice, Scott Foresman, Glenview, IL, .
Demirkan, H., Pultar, M., Ozguc, B. (1992), “A knowledge-based space planning system”, Architectural Science Review, Vol. 35 No.1, pp.3-7.
Eberhart, R.C. (1998), “Computational intelligence: roles in multimedia”, in Selvaraj, H., Verma, B. (Eds),International Conference on Computational Intelligence and Multimedia Applications, World Scientific, Singapore, pp.3-8.
El-Bibany, H. (1996), “Multi-media development software: object-oriented interface-based simulation”, ASCE Journal of Computing in Civil Engineering, Vol. 10 No.4, pp.295-9.