Scribd
Upload
12 views9 pages

LITRATURE

The document reviews various studies correlating California Bearing Ratio (CBR) values with soil index properties to improve pavement design efficiency. It highlights the challenges of traditional CBR testing and proposes alternative predictive models using Artificial Neural Networks and Multiple Linear Regression for quicker and more accurate estimations. The findings indicate that regression models generally outperform neural networks in predicting CBR values based on soil properties.

Uploaded by

vijaysureshbhosale4949
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as XLSX, PDF, TXT or read online on Scribd
12 views9 pages

LITRATURE

The document reviews various studies correlating California Bearing Ratio (CBR) values with soil index properties to improve pavement design efficiency. It highlights the challenges of traditional CBR testing and proposes alternative predictive models using Artificial Neural Networks and Multiple Linear Regression for quicker and more accurate estimations. The findings indicate that regression models generally outperform neural networks in predicting CBR values based on soil properties.

Uploaded by

vijaysureshbhosale4949
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as XLSX, PDF, TXT or read online on Scribd
You are on page 1/ 9

SL no TITLE AUTHOR YEAR

1 ANN model for Venkatasubramanian C., Dhinakaran G. 2011


predicting CBR from
index properties of soils

2 CORRELATION OF CBR Zohib S. Janjua, Jagdish Chand 2016


WITH INDEX
PROPERTIES OF SOIL

3 Prediction of CBR Value Sandhya Rani, Nagaraj 2017


with Soil Index
Properties; Case Study
on Yadadri Region
4 Experimental Study to Pravin K. Kushwaha, Hemant K. Mahiyar 2017
Correlate Soaked CBR
with Index and
Engineering Properties
of Black
Cotton Soil of Madhya
Pradesh

5 CORRELATION BETWEEN Krishnan Pal, Koustav Pal 2019


CBR VALUES AND
PLASTICITY INDEX OF
SOIL FOR KOLKATA
REGION

6 Correlation of CBR with Anant Mishra, J.P. Tegar 2019


Index Properties of Soil
of Kolar Region (Bhopal)
using Regression
Analysis

7 CORRELATION BETWEEN Krishnan Pal, S.J. Kurumban, Koustav Pal 2020


CBR VALUES AND
PLASTICITY INDEX OF
SOIL FOR BIRBHUM
REGION
CBR Characteristics and Akinola J. Olarewaju, A.S. Adebesin 2022
Behaviors of Black
Cotton Soil on
Sedimentary Formation

8
ABSTRACT

The subgrade supports the pavement and must be well compacted to maximize
strength and reduce pavement thickness. Its strength is measured by the CBR test
according to IS: 2720 (Part 16), using molded soil samples in the lab. The CBR test is
time-consuming and sometimes inaccurate due to poor lab conditions. When soil
quality is low, additives are used, complicating testing.

This study proposes predicting CBR values by correlating them with soil properties—
liquid limit, plastic limit, plasticity index, OMC, maximum dry density, and UCC—from
soils in three Tamil Nadu districts using Artificial Neural Networks. Results show that
Multiple Linear Regression models predict CBR better than Neural Networks.

The California Bearing Ratio (CBR) test measures the stiffness and shear strength of
subgrade soil for flexible pavement design. The subgrade must be well compacted to
provide a strong foundation. CBR testing, done per IS: 2720 using static penetration
on molded samples, is time-consuming and often challenging to perform accurately .
This study
aims to predict CBR values quickly by correlating them with soil properties using
Artificial Neural Networks (ANN) and Multiple Linear Regression (MLR). Tests on 11
samples of SW-SM soil from two Punjab districts showed that soaked CBR strongly
correlates with percent fines, liquid limit, maximum dry density, and optimum
moisture content. Regression analysis yielded an equation with an R² of 0.844,
indicating good predictive accuracy.

Geotechnical engineers develop empirical equations for specific areas and soils since
the CBR test is tedious and laborious. This study correlates CBR values with soil index
properties—such as sieve analysis, Atterberg limits, and compaction tests—from 18
samples collected in the Yadadri region. Laboratory tests were conducted, and single
and multiple linear regression analyses were performed using SPSS to predict CBR
values based on these correlations.
California Bearing Ratio (CBR) is a key input for pavement design, measuring the shear
strength and stiffness of subgrade materials. Due to the time and cost involved in
conventional CBR testing, predicting CBR from other soil properties is beneficial.

This study correlates soaked CBR values with index and engineering properties of
black cotton soil. Using 35 samples from eight districts in Madhya Pradesh, India,
simple and multiple linear regression analyses were performed. Soil properties
examined include specific gravity, liquid limit, plastic limit, free swell index,
unconfined compressive strength, and swelling pressure. Additionally, regression
models were developed to predict unconfined compressive strength and swelling
pressure from index properties.

The California Bearing Ratio (CBR) test is essential for flexible pavement design, but
it's time-consuming, labor-intensive, and often inaccurate due to poor field lab
conditions and unskilled technicians. As a result, only a limited number of tests are
done per kilometer of road, which may not capture soil variability, leading to poor
construction quality—especially in erratic alluvial soils common in the Indo-Gangetic
plain.
To address this, the study proposes a faster, low-cost method to estimate CBR values
using correlations with Percentage Finer and Plasticity Index. Using extensive data
from soil samples collected across Kolkata, a predictive equation was developed via
graphical analysis. This method helps geotechnical engineers better estimate
subgrade strength and verify lab results, improving design accuracy and construction
quality.

Soil properties like CBR, Liquid Limit, Plastic Limit, Maximum Dry Density, Optimum
Moisture Content, Plasticity Index, and % Fines are essential but time-consuming to
test due to soil variability.This study focuses on correlating CBR with index properties
for soils from the Bhopal Kolar region.

Used data from Birbhum to correlate CBR values with soil index properties through
statistical analysis. This hampers the ability of engineers to design safe and cost-
effective roads and to address weak zones along the alignment effectively.
Analyzed CBR behavior of black cotton soil over sedimentary formations for suitability
in road subgrades.Black cotton soils are problematic due to properties like swelling,
erosion, high compressibility, and low bearing capacity. California Bearing Ratio (CBR)
characteristics of black cotton soil from the basement complex were studied.
CONCLUSION
Experimental values of 15 samples on various properties are
given as input to develop model using neural network and
multiple linear regression analysis to predict CBR. From the
results like correlation coefficient, R2 value and mean square
error, it is concluded that CBR prediction model based on
multiple linear regression performs better than neural network
model and hence it is recommended to predict CBR values. But
this value may be improved
by modifying the parameters.

The tests performed on these soil samples were Grain size


analysis, Liquid limit, Plastic limit, Plasticity
index, Maximum dry density, Optimum moisture content and
California bearing ratio.
The model developed by neural networks is trained with 20
values obtained from literature review and 11
values from laboratory tests for soaked CBR.
A total of 3 models were developed by multiple linear
regression analysis (MLR), for soaked CBR, out of
that one models showed a good correlation of CBR with Soil
properties.

Single linear regression showed a moderate correlation


between CBR and liquid limit (LL). However, multiple regression
analysis using LL, plastic limit (PL), maximum dry density (MDD),
optimum moisture content (OMC), percent passing 200 mesh
(P200), and plasticity index (PI) gave better results. Overall,
combined soil properties predict CBR strength more accurately
than individual properties.
Reliable CBR prediction helps reduce cost and time in pavement
design and site evaluation.For preliminary design with black
cotton soil, these correlations are valid if predicted CBR is
between 1.6% and 4.5%. Otherwise, detailed lab tests per IS
standards are recommended.

The model enables quicker CBR estimation using simple soil


tests with good accuracy.Since CBR testing is laborious and
time-consuming, using easily obtained soil properties—like
Liquid Limit, Plastic Limit, and grain size analysis—can be a
practical alternative for estimating CBR. This study shows that
predicted soaked CBR values, based on grain size and Plasticity
Index (PI), closely match measured values. The model offers
geotechnical engineers a quick, cost-effective way to estimate
and verify CBR values, enhancing reliability in pavement design.

The study successfully established strong correlations between


CBR and basic soil index properties for the Bhopal Kolar region.

Since CBR testing is time-consuming and labor-intensive,


estimating CBR using simpler, quicker, and low-cost tests like
Liquid Limit, Plastic Limit, and grain size analysis is practical. This
study found that predicted CBR values closely match measured
ones. The developed model uses grain size and Plasticity Index
to estimate soaked CBR, helping geotechnical engineers both
predict CBR values and verify lab test accuracy.
Due to low CBR, these soils require stabilization before being
used in construction.he California Bearing Ratio (CBR)
characteristics of black cotton soil from the basement complex
were studied. The soil’s Maximum Dry Density (MDD) is
achieved at a lower moisture content compared to typical
basement complex soils. The results clearly indicate that the soil
has a low CBR value, classifying it as poor quality for
construction purposes.

You might also like