Predicting liquefaction-induced lateral spreading by using the multigene genetic programming (MGGP), multilayer perceptron (MLP), and random forest (RF) techniques


KAYA Z., LATİFOĞLU L., UNCUOĞLU E., EROL A., Keskin M. S.

BULLETIN OF ENGINEERING GEOLOGY AND THE ENVIRONMENT, vol.82, no.3, 2023 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 82 Issue: 3
  • Publication Date: 2023
  • Doi Number: 10.1007/s10064-023-03103-9
  • Journal Name: BULLETIN OF ENGINEERING GEOLOGY AND THE ENVIRONMENT
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, IBZ Online, Aerospace Database, Agricultural & Environmental Science Database, Aquatic Science & Fisheries Abstracts (ASFA), CAB Abstracts, Communication Abstracts, Compendex, Environment Index, Geobase, INSPEC, Metadex, Pollution Abstracts, Civil Engineering Abstracts
  • Keywords: Kocaeli earthquake, Lateral spreading, MGGP, MLP, RF
  • Erciyes University Affiliated: Yes

Abstract

Landslides refer to a wide range of processes that result in the downward and outward movement of slope-forming materials, which may spread. Estimating lateral spreading of soil is essential because of the complexities associated with the lateral spreading behavior. Existing empirical models for predicting liquefaction-induced lateral spread displacement are developed using a dataset that varied in terms of earthquake magnitude, source distance, ground slope, layer thickness, fines content, and grain size. The aim of this study is to increase the accuracy of earthquake-induced lateral spreading prediction using multigene genetic programming (MGGP), multilayer perceptron (MLP), and random forest (RF) model. MGGP, MLP, and RF model predictions of lateral spreading are compared with the results anticipated using machine learning techniques and conventional approaches. Results showed that the MGGP outperforms the Hamada, Youd, MLP, and RF equations for estimating maximum lateral displacement under free-face and gently sloping ground conditions according to the comparisons. The MGGP, which is proved to be better, was also utilized to estimate total lateral displacement for Adapazari data, along with machine learning techniques and conventional approaches.