Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 78
PROCEEDINGS OF THE SEVENTH INTERNATIONAL CONFERENCE ON THE APPLICATION OF ARTIFICIAL INTELLIGENCE TO CIVIL AND STRUCTURAL ENGINEERING
Edited by: B.H.V. Topping
Paper 9

Knowledge Based Systems Applications for Dam Engineering, Railway Tunnelling and Structural Design

E.A. Portela+, F. Farinha*, C.S. Domingues$ and L.R. Sousa+

+National Laboratory of Civil Engineering, Portugal
*Algarve University, Higher School of Technology, Portugal
$Polytechnic School of Porto, Portugal

Full Bibliographic Reference for this paper
E.A. Portela, F. Farinha, C.S. Domingues, L.R. Sousa, "Knowledge Based Systems Applications for Dam Engineering, Railway Tunnelling and Structural Design", in B.H.V. Topping, (Editor), "Proceedings of the Seventh International Conference on the Application of Artificial Intelligence to Civil and Structural Engineering", Civil-Comp Press, Stirlingshire, UK, Paper 9, 2003. doi:10.4203/ccp.78.9
Keywords: knowledge based system, dam engineering, tunnelling,.

Summary
Although substantial advances have been achieved in last twenty years, we are still far from taking full advantage of the potential of Knowledge Based Systems (KBS) in real life. Very few KBS have survived their evaluation period and only one third of the initially reported as successes are still in use. The paper reports three KBS applications developed for research purposes, in the field of civil engineering and highlight their achievements and the main issues concerning their future developments to solve real life problems.

The main challenge of these applications were to elicit knowledge and reasoning strategies out of engineering experts, so that a computer model of such expertise could be developed as a means for assisting engineers in the decision making process.

The first application is in the field of dam engineering. As many of dam structures are now over 50 years old, special attention has been dedicated to issues related to ageing. Managing large dams requires complex computing and the effort of multi-disciplinary teams with specialised knowledge. This knowledge is accumulated through a process of learning and understanding how the structure responds to the current and exceptional actions. The main purpose of this application is to assist dam managers in the various safety assessment tasks [1].

The second application centralizes into the main issues related with the safety control of old railway tunnels. Most of the existing underground railway patrimony in Portugal was built in the end of the 19th Century/beginning of the 20th Century. Therefore, as a result of the ageing of these structures and the deficient construction methods used at the time, a significant deterioration in the condition of the tunnels has been observed, which may endanger the safety of these underground structures. The observation of the works becomes an essential issue and must be done in a well-planned systematic way to ensure its safety and serviceability. In order to guarantee the adequate safety level and to assist the establishment of recommendations for diagnosis and rehabilitation, a computational system designed - MATUF was implemented [2].

The third application supports structural design activities. Structural design activities have accompanied, since the beginning, the IT revolution and its later widespread employment directly linked to the use of personal computers [3]. During this first stage, the tasks involved were mainly numeric ones for which objective mathematical models existed. However there are others areas in which the procedures employed are totally dependent on the engineers and were considered, till recently, as difficult or even impossible to computerize, particularly since they were not of a numeric nature and were based on heuristic knowledge (e.g. conformity with building codes, design optimisation). The application chosen is the reinforced concrete analysis and detailing [4]. The main motivation for choosing this activity is based in two relevant facts: 1) there are relatively low numbers of aids for reinforced concrete analysis and detailing; 2) detailing is a complex process that involves activities and decisions that are heuristic in nature and rely on knowledge, experience and judgement.

The experience gained with the three cases studies enable to conclude that the application of KBS technology contributes to enhance the whole process, namely, providing higher levels of economy, robustness and efficiency. These applications were developed to support the decision making process of engineers in different fields but can also be used as a learning tool for students or young engineers.

However, the real-life application of those systems needs further developments. Despite no technological problems were found, their success depends heavily on managerial or economics issues. Furthermore, the development and maintenance of such systems may not satisfy the profit requirements common to most companies. In the future, those systems will probably need to be justified on basis of its strategic impact in terms of gaining competitive advantage rather than cost-effectiveness.

References
1
Portela, E.A., "New methodologies to support the safety control of concrete dams. An expert systems approach" (in Portuguese). PhD. Thesis, Technical University of Lisbon, Lisbon, 1999.
2
Silva Domingues, C., "Safety control of railway tunnels. Development of support methodologies and based knowledge systems" (in Portuguese). MSc Thesis, University of Porto, Porto, 2001.
3
Fenves, S., "Information technology in construction: a personal journey", in "Construction on the information highway", Turk, Z. (Editor), CIB Publication 198, University of Ljubljana, Slovenia, 1996.
4
Farinha, F., "Reflective practice in intelligent CAD systems for buildings", (in Portuguese), PhD Thesis, Technical University of Lisbon, Lisbon, 1997.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £82 +P&P)