Analysis of many civil engineering phenomena is a complex problem due to the participation of a large number of factors involved. Traditional methods usually suffer from a lack of physical understanding. Furthermore, the simplifying assumptions that are usually made in the development of the traditional methods may, in some cases, lead to very large errors. The purpose of this paper is to present a new method, based on evolutionary polynomial regression (EPR) for capturing nonlinear interaction between various parameters of civil engineering systems.
EPR is a data‐driven method based on evolutionary computing, aimed to search for polynomial structures representing a system. In this technique, a combination of the genetic algorithm and the least‐squares method is used to find feasible structures and the appropriate constants for those structures.
Capabilities of the EPR methodology are illustrated by application to two complex practical civil engineering problems including evaluation of uplift capacity of suction caissons and shear strength of reinforced concrete deep beams. The results show that the proposed EPR model provides a significant improvement over the existing models. The EPR models generate a transparent and structured representation of the system. For design purposes, the EPR models, presented in this study, are simple to use and provide results that are more accurate than the existing methods.
In this paper, a new evolutionary data mining approach is presented for the analysis of complex civil engineering problems. The new approach overcomes the shortcomings of the traditional and artificial neural network‐based methods presented in the literature for the analysis of civil engineering systems. EPR provides a viable tool to find a structured representation of the system, which allows the user to gain additional information on how the system performs.
Rezania, M., Javadi, A.A. and Giustolisi, O. (2008), "An evolutionary‐based data mining technique for assessment of civil engineering systems", Engineering Computations, Vol. 25 No. 6, pp. 500-517. https://doi.org/10.1108/02644400810891526
Emerald Group Publishing Limited
Copyright © 2008, Emerald Group Publishing Limited