Model tree approach for predicting uniaxial compressive strength and Young’s modulus of carbonate rocks

Ebrahim Ghasemi, Hamid Kalhori, Raheb Bagherpour, Saffet Yagiz

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

The uniaxial compressive strength (UCS) and Young’s modulus (E) of rock are important parameters for evaluating the strength, deformation, and stability of rock engineering structures. Direct measurement of these parameters is expensive, time-consuming, and even infeasible in some circumstances due to the difficulty involved in obtaining core samples. Recently, soft computing tools have been used to predict UCS and E based on index tests. Most of these tools are not as transparent and easy to use as empirical regression-based models. This study presents another soft computing approach—model trees—for predicting the UCS and E of carbonate rocks. The main advantages of model trees are that they are easier to use than other data learning tools and, more importantly, they represent understandable mathematical rules. In this study, the M5P algorithm was employed to build and evaluate model trees (UCS and E model trees). First, the models were developed in an unpruned form, and then they were pruned to avoid overfitting. The data used to train and test the model trees were collected from quarries in southwestern Turkey. Model trees included Schmidt hammer, effective porosity, dry unit weight, P‐wave velocity, and slake durability index as input variables. When the models were assessed using a number of statistical indices (RMSE, MAE, VAF, and R2), it was found that unpruned and pruned model trees provide acceptable predictions of UCS and E, although the pruned models are simpler and easier to understand.

Original languageEnglish
Pages (from-to)331-343
Number of pages13
JournalBulletin of Engineering Geology and the Environment
Volume77
Issue number1
DOIs
Publication statusPublished - Feb 1 2018
Externally publishedYes

Fingerprint

Young modulus
compressive strength
carbonate rock
Compressive strength
Carbonates
Elastic moduli
Rocks
Soft computing
Core samples
Quarries
Hammers
durability
quarry
rock
train
Durability
Porosity
learning
porosity
engineering

Keywords

  • Carbonate rocks
  • Index tests
  • M5P algorithm
  • Model tree
  • Uniaxial compressive strength
  • Young’s modulus

ASJC Scopus subject areas

  • Geotechnical Engineering and Engineering Geology
  • Geology

Cite this

Model tree approach for predicting uniaxial compressive strength and Young’s modulus of carbonate rocks. / Ghasemi, Ebrahim; Kalhori, Hamid; Bagherpour, Raheb; Yagiz, Saffet.

In: Bulletin of Engineering Geology and the Environment, Vol. 77, No. 1, 01.02.2018, p. 331-343.

Research output: Contribution to journalArticle

@article{90eea1cca8ea445caaaee43f2b6561fd,
title = "Model tree approach for predicting uniaxial compressive strength and Young’s modulus of carbonate rocks",
abstract = "The uniaxial compressive strength (UCS) and Young’s modulus (E) of rock are important parameters for evaluating the strength, deformation, and stability of rock engineering structures. Direct measurement of these parameters is expensive, time-consuming, and even infeasible in some circumstances due to the difficulty involved in obtaining core samples. Recently, soft computing tools have been used to predict UCS and E based on index tests. Most of these tools are not as transparent and easy to use as empirical regression-based models. This study presents another soft computing approach—model trees—for predicting the UCS and E of carbonate rocks. The main advantages of model trees are that they are easier to use than other data learning tools and, more importantly, they represent understandable mathematical rules. In this study, the M5P algorithm was employed to build and evaluate model trees (UCS and E model trees). First, the models were developed in an unpruned form, and then they were pruned to avoid overfitting. The data used to train and test the model trees were collected from quarries in southwestern Turkey. Model trees included Schmidt hammer, effective porosity, dry unit weight, P‐wave velocity, and slake durability index as input variables. When the models were assessed using a number of statistical indices (RMSE, MAE, VAF, and R2), it was found that unpruned and pruned model trees provide acceptable predictions of UCS and E, although the pruned models are simpler and easier to understand.",
keywords = "Carbonate rocks, Index tests, M5P algorithm, Model tree, Uniaxial compressive strength, Young’s modulus",
author = "Ebrahim Ghasemi and Hamid Kalhori and Raheb Bagherpour and Saffet Yagiz",
year = "2018",
month = "2",
day = "1",
doi = "10.1007/s10064-016-0931-1",
language = "English",
volume = "77",
pages = "331--343",
journal = "Bulletin of the International Association of Engineering Geology - Bulletin de l'Association Internationale de G{\'e}ologie de l'Ing{\'e}nieur",
issn = "0074-1612",
publisher = "Springer Verlag",
number = "1",

}

TY - JOUR

T1 - Model tree approach for predicting uniaxial compressive strength and Young’s modulus of carbonate rocks

AU - Ghasemi, Ebrahim

AU - Kalhori, Hamid

AU - Bagherpour, Raheb

AU - Yagiz, Saffet

PY - 2018/2/1

Y1 - 2018/2/1

N2 - The uniaxial compressive strength (UCS) and Young’s modulus (E) of rock are important parameters for evaluating the strength, deformation, and stability of rock engineering structures. Direct measurement of these parameters is expensive, time-consuming, and even infeasible in some circumstances due to the difficulty involved in obtaining core samples. Recently, soft computing tools have been used to predict UCS and E based on index tests. Most of these tools are not as transparent and easy to use as empirical regression-based models. This study presents another soft computing approach—model trees—for predicting the UCS and E of carbonate rocks. The main advantages of model trees are that they are easier to use than other data learning tools and, more importantly, they represent understandable mathematical rules. In this study, the M5P algorithm was employed to build and evaluate model trees (UCS and E model trees). First, the models were developed in an unpruned form, and then they were pruned to avoid overfitting. The data used to train and test the model trees were collected from quarries in southwestern Turkey. Model trees included Schmidt hammer, effective porosity, dry unit weight, P‐wave velocity, and slake durability index as input variables. When the models were assessed using a number of statistical indices (RMSE, MAE, VAF, and R2), it was found that unpruned and pruned model trees provide acceptable predictions of UCS and E, although the pruned models are simpler and easier to understand.

AB - The uniaxial compressive strength (UCS) and Young’s modulus (E) of rock are important parameters for evaluating the strength, deformation, and stability of rock engineering structures. Direct measurement of these parameters is expensive, time-consuming, and even infeasible in some circumstances due to the difficulty involved in obtaining core samples. Recently, soft computing tools have been used to predict UCS and E based on index tests. Most of these tools are not as transparent and easy to use as empirical regression-based models. This study presents another soft computing approach—model trees—for predicting the UCS and E of carbonate rocks. The main advantages of model trees are that they are easier to use than other data learning tools and, more importantly, they represent understandable mathematical rules. In this study, the M5P algorithm was employed to build and evaluate model trees (UCS and E model trees). First, the models were developed in an unpruned form, and then they were pruned to avoid overfitting. The data used to train and test the model trees were collected from quarries in southwestern Turkey. Model trees included Schmidt hammer, effective porosity, dry unit weight, P‐wave velocity, and slake durability index as input variables. When the models were assessed using a number of statistical indices (RMSE, MAE, VAF, and R2), it was found that unpruned and pruned model trees provide acceptable predictions of UCS and E, although the pruned models are simpler and easier to understand.

KW - Carbonate rocks

KW - Index tests

KW - M5P algorithm

KW - Model tree

KW - Uniaxial compressive strength

KW - Young’s modulus

UR - http://www.scopus.com/inward/record.url?scp=84983438099&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84983438099&partnerID=8YFLogxK

U2 - 10.1007/s10064-016-0931-1

DO - 10.1007/s10064-016-0931-1

M3 - Article

VL - 77

SP - 331

EP - 343

JO - Bulletin of the International Association of Engineering Geology - Bulletin de l'Association Internationale de Géologie de l'Ingénieur

JF - Bulletin of the International Association of Engineering Geology - Bulletin de l'Association Internationale de Géologie de l'Ingénieur

SN - 0074-1612

IS - 1

ER -