Analytical design models in construction engineering: artificial neural network and gene expression programming practices

Abstract

One important aspect of experimental studies is to supply valuable raw data for the literature. However data at the base of the DIKW (data-information-knowledge-wisdom) pyramid lacks significance unless it has been condensed organized categorized structured and evaluated. In this way the data will be transformed into information. Then processing the information will enable progression to the knowledge stage of the DIKW pyramid and applying this knowledge in action will help to reach the wisdom stage. In this context it is crucial to transform the pure data obtained from the experimental studies into wisdom. Construction engineering a crucial and evolving field of civil engineering advances through experimental research and practical outcomes. Therefore it can be concluded that the core principles of construction engineering are founded on correctly and wisely expressing findings from experimental and practical research. In recent years there have been numerous efforts to gather data from both experimental and practical studies and apply soft-computing techniques to extract valuable insights. In this respect the paper presents empirical design models developed using artificial neural networks (ANNs) and gene expression programming (GEP) noteworthy soft computing methods to solve some engineering problems in the construction field. This chapter aims to explain the significance of evaluating data and transforming it into a clear analytical design model. Following that it examines both soft-computing methods in great detail. Ultimately analytical design models developed to solve construction engineering problems are presented. © 2025 Elsevier B.V. All rights reserved.