Utilization Of Machine-Learning-Based Model Hybridized With Meta-Heuristic Frameworks For Estimation Of Unconfined Compressive Strength

Aug 5, 2026

Unconfined compressive strength (UCS) is a key rock property for building reliable geo-mechanical models used in many engineering applications. Traditional UCS determination relies on laboratory core testing or well-log analysis, which can be time-consuming. The study evaluates a radial basis function (RBF) machine-learning model whose parameters are optimized by two improved meta-heuristic frameworks, Improved Arithmetic Optimization Algorithm (IAOA) and Flying Foxes Optimization (FFO). Performance is assessed with RMSE, R², MAE, U95, and MNB on an existing dataset. Results show the hybrid RBFF approach substantially outperforms standalone RBF and RBF-IA, achieving R²=0.998 and lower error for improved UCS prediction.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Introduction

Background on UCS importance and conventional determination

Limitations of statistical and index-based methods

Machine learning approaches for UCS prediction

Journal of Applied Science and Engineering, Vol. 28, No 8, Page 1779-1794 1779 Utilization Of Machine-Learning-Based Model Hybridized With Meta-Heuristic Frameworks For Estimation Of Unconfined Compressive Strength She Wang1* and Qi Zhang2 1 School of Computer and Electronic Information Engineering, Wuhan City Polytechnic, Wuhan 430000, Hubei, China 2 Department of Commerce and Trade, Wuhan Instrument and Electronic Technical School, Wuhan 430205, Hubei, China *Corresponding author. E-mail: [wangshe20121207@126.com](wangshe20121207@126.com) [Received: Jul. 07](Received: Jul. 07), [2024](2024); [Accepted: Sep](Accepted: Sep). 30, 2024 Unconfined compressive strength (UCS) is one of the rocks’ most valuable mechanical properties in constructing an accurate geo-mechanical model. It has traditionally been determined through laboratory core sample testing or by analysis of well-log data. After a great deal of effort and growing investment in time, the proper adoption of machine learning methods, especially the radial basis function (RBF), opens a route to promising alternatives against empirical methods for better real-time prediction of UCS. The current study considers the RBF-based machine learning model, whose parameters have been optimized using two enhanced metaheuristic frameworks: Improved Arithmetic Optimization Algorithm (IAOA) and Flying Foxes Optimization (FFO) . Based on an extensive dataset already used in previous studies and applying some soft computing techniques, vigorous performance metrics such as RMSE, R², MAE, U95, and MNB were used to test the developed frameworks. The outcomes indicate a significant outperformance of the hybrid RBFF technique over the solo RBF and RBF-IA frameworks. Specifically, the RBFF model resulted in an R2 of 0 .998, an RMSE of 1 .313, and an MNB of-0.003, reflecting its better performance in UCS prediction. This study indicates the efficiency of integrating RBF with meta-heuristic optimization to enhance UCS predictions in geotechnical studies. Keywords: Unconfined compressive strength; Radial Basis Function; Improved Arithmetic optimization algorithm; Flying Foxes Optimization. © The Author(’s) . This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited. [http://dx.doi.org/10.6180/jase.202508_28](http://dx.doi.org/10.6180/jase.202508_28)(8).0015 1. Introduction The long-term stability and reliability of many land and subsurface critical infrastructures are substantially dependent on their underlying solid foundation consisting of undisturbed and fine-grained rock masses. Maintaining these rock masses’ integrity over a long period is essential to sustaining these infrastructures’ operational function. This need encompasses a wide range of applications. Still, it is not limited to slopes, roofing, and flooring, extending to creating load-carrying columns in operations like mining, tunneling, and other civil engineering activities [1] . In rock engineering projects, the stability of the structures is directly related to a host of geo-mechanical properties of the rock mass and the natural rock. These critical factors include vital characteristics such as UCS and tensile strength. Narrower but equally significantly, the UCS of rock material plays a vital role as a principal metric in structures’ design and performance appraisal, whether above or underground. The UCS always heads the list when mining engineers consider several properties related to the rock [2] . This increased focus on UCS can be attributed to the fact that the geological material plays a vital role in conducting stability studies for various underground projects. Such 1780 She Wang et al. projects include but are not limited to excavation, development of stook support given longwall mining, access tunnel dev

FAQs

How does the proposed hybrid RBFF method perform compared with other variants?

It significantly outperforms solo RBF and RBF-IA, reaching an R² of 0.998 with an RMSE around 1.313 and improved error metrics for UCS prediction.

Aug 5, 2026

Upload by Theodore