www.nature.

com/scientificreports

OPEN Utilizing machine learning

for flow zone indicators
prediction and hydraulic flow unit
classification
Tengku Astsauri 1, Muhammad Habiburrahman 1, Ahmed Farid Ibrahim 1,2* & Yuzhu Wang 1,2
Reservoir characterization, essential for understanding subsurface heterogeneity, often faces
challenges due to scale-dependent variations. This study addresses this issue by utilizing hydraulic
flow unit (HFU) zonation to group rocks with similar petrophysical and flow characteristics. Flow Zone
Indicator (FZI), a crucial measure derived from pore throat size, permeability, and porosity, serves as
a key parameter, but its determination is time-consuming and expensive. The objective is to employ
supervised and unsupervised machine learning to predict FZI and classify the reservoir into distinct
HFUs. Unsupervised learning using K-means clustering and supervised algorithms including Random
Forest (RF), Extreme Gradient Boosting (XGB), Support Vector Machines (SVM), and Artificial Neural
Networks (ANN) were employed. FZI values from RCAL data formed the basis for model training and
testing, then the developed models were used to predict FZI in unsampled locations. A methodical
approach involves 3 k-fold cross-validation and hyper-parameter tuning, utilizing the random search
cross-validation technique over 50 iterations was applied to optimize each model. The four applied
algorithms indicate high performance with coefficients determination ­(R2) of 0.89 and 0.91 in training
and testing datasets, respectively. RF showed the heist performance with training and testing ­R2
values of 0.957 and 0.908, respectively. Elbow analysis guided the successful clustering of 212 data
points into 10 HFUs using k-means clustering and Gaussian mixture techniques. The high-quality
reservoir zone was successfully unlocked using the unsupervised technique. It has been discovered
that the areas between 2370–2380 feet and 2463–2466 feet are predicted to be high-quality reservoir
potential areas, with average FZI values of 500 and 800, consecutively. The application of machine
learning in reservoir characterization is deemed highly valuable, offering rapid, cost-effective, and
precise results, revolutionizing decision-making in field development compared to conventional
methods.

Keywords Machine learning, Flow zone indicators, Hydraulic flow unit, Reservoir characterization

List of symbols
k Permeability in ­m2
φe Effective porosity
KT Pore-level effective zoning factor
Svgr Specific surface area per unit grain volume
FZI Flow zone indicator
HFU Hydraulic flow unit
RF Random forest
XGB Extreme gradient boosting
SVM Support vector machines
ANN Artificial neural networks
Fs Shape factor
τ Tortuosity

1
Department of Petroleum Engineering and Geosciences, King Fahd University of Petroleum & Minerals,
31261 Dhahran, Saudi Arabia. 2Center for Integrative Petroleum Research, King Fahd University of Petroleum &
Minerals, 31261 Dhahran, Saudi Arabia. *email: ahmed.ibrahim@kfupm.edu.sa

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 1

Vol.:(0123456789)
 www.nature.com/scientificreports/

Sgv Surface area per unit grain in μm

j
xi The ith pattern belonging to the jth cluster
cj Centroid of the jth cluster
CGR​ Corrected gamma ray
DRHO Bulk density correction
DTC Delta-T compressional
GR Gamma ray
HNPO High-resolution enhanced thermal neutron
LLD Laterolog deep resistivity
LLHR High-resolution laterolog resistivity
LLS Laterolog shallow resistivity
MRES Mud resistivity
MSFC Micro spherically focused resistivity
NPHI Thermal neutron porosity
NPOR Enhanced thermal neutron porosity
POTA Potassium concentration
RHOB Bulk density
SP Spontaneous potential
THOR Thorium concentration
URAN Uranium concentration
xnorm Normalized value with a range of values 0–1
x Variable on the dataset while max and min refers to the maximum and minimum value of the
variable

Reservoir characterization is a fundamental part of petroleum engineering that involves gathering and evaluat-
ing data to understand the properties of a subsurface reservoir. This process is necessary for making informed
decisions involving the production and recovery of hydrocarbons from the ­reservoir1. The information gathered
during reservoir characterization is critical for accurate hydrocarbon reserve estimation, optimization of produc-
tion techniques, risk reduction, and improved recovery, which is vital to financial analysis and decision-making in
the oil and gas ­industry2. In addition, reservoir characterization provides valuable information on the reservoir’s
properties and behavior, which contributes to the development of an optimum field development plan, includ-
ing the determination of the number and placement of wells, production rates, and field infrastructure design.
Reservoir characterization is a challenging task due to the uncertainty imposed by reservoir heterogeneity,
which refers to the variability of reservoir properties across various geological scales. To address this uncertainty,
the hydraulic flow unit (HFU) zonation is used to cluster rocks with identical petrophysical and flow charac-
teristics into the same u­ nit3. This allows for the prediction of unknown reservoir properties and eliminates
unnecessary coring expenses. HFUs are based on geological and physical flow principles and provide a more
accurate representation of reservoir heterogeneity compared to traditional lithological or depositional facies-
based approaches. The Hydraulic Flow Unit method is related to the Flow Zone Indicator (FZI), a commonly
used measure in reservoir characterization. The FZI provides a quantitative method for analyzing the relationship
between microscopic characteristics like pore throat size and distribution and macroscopic ones like perme-
ability and porosity. This consequently suggests that rock properties derived from depositional and diagenetic
processes play a significant role in determining the surface area, shape factor, and tortuosity of carbonates, and
thus the FZI v­ alue4.
Conventional methods for reservoir characterization primarily focus on directly measuring or estimating
permeability and porosity, which are crucial for understanding reservoir potential. The primary tools for this
purpose are core measurements and well logs. Core measurements involve physically extracting a sample from
the reservoir and analyzing it to determine properties like permeability and porosity. Well logs, on the other
hand, are continuous recordings of various physical parameters along the wellbore, providing indirect estimates
of these reservoir properties.
While core measurements offer high accuracy, they are often expensive, time-consuming, and only provide
data for a limited section of the reservoir. Well logs, including tools like bulk density, neutron porosity, sonic,
and nuclear magnetic resonance logs, are more extensive but can sometimes yield less satisfactory results. This
is due to uncertainties in the empirical parameters used for interpretation and the adaptability issues of response
equations to different reservoir conditions. These limitations of conventional methods highlight the need for
more efficient and comprehensive approaches in reservoir characterization (Rock Typing)5,6. Therefore, there
is a need to identify advanced methods capable of overcoming the limitations inherent in traditional reservoir
characterization techniques.
AI and Machine Learning (ML) offer solutions to these challenges by efficiently processing vast quantities
of data, surpassing the limitations of human analysis in both speed and complexity. These advanced technolo-
gies can interpret intricate datasets from logs more effectively, identifying patterns and correlations that might
be missed by traditional methods. Furthermore, AI-driven methods are not confined to the data from cored
intervals, enabling a more comprehensive analysis of the reservoir. This holistic approach can integrate diverse
data sources, including seismic, geological, and production data, offering a more nuanced understanding of
reservoir characteristics. Studies have utilized a range of supervised machine learning algorithms, including
Random Forest (RF)7, Support Vector Machines (SVM)8, Artificial Neural Networks (ANN)9, adaptive network
fuzzy inference system (ANFIS)10, and Extreme Gradient Boosting (XGB)6, to accurately predict permeability

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 2

Vol:.(1234567890)
www.nature.com/scientificreports/

values. Additionally, unsupervised machine learning algorithms such as K-Means have been studied to classify
the reservoir based on the hydraulic flow units (HFUs)11,12.
The main objective of this study is to create a supervised machine-learning model that directly estimates the
flow zone indicator (FZI) at unsampled locations using well-logging data during the initial exploration phase.
This approach is highly valuable as it allows for the direct determination of FZI at specific depths of interest,
leveraging the power of supervised machine learning. Additionally, an unsupervised machine-learning model
will be developed to cluster hydraulic flow unit numbers in the target zone. This clustering approach is also valu-
able as it enables the assessment of distinct petrophysical properties associated with flow units, which greatly
influences reservoir characterization.
To accomplish the study’s objective, the implementation of popular machine learning algorithms like K-Means
for the unsupervised machine learning model, and Random Forest, Extreme Gradient Boosting, Support Vector
Machines, and Artificial Neural Network for the supervised machine learning model is planned. Additionally, the
credibility of the results will be ensured by evaluating the physics-based approach in conjunction with the data-
driven approach of supervised machine learning. This combination of approaches will enhance the classification
of rock reservoir types, resulting in improved accuracy and efficiency. Therefore, this research aims to introduce
a dependable and data-driven approach for predicting flow zone indicators in unsampled locations, utilizing
advanced machine learning techniques. This innovative methodology is poised to significantly contribute to the
advancement of rock reservoir type classification within the petroleum industry, marking a shift towards more
sophisticated, analytics-based strategies.

The flow zone indicator (FZI)

In reservoir characterization, predicting permeability is crucial for understanding hydrocarbon production.
The Hydraulic Flow Unit (HFU) approach was first introduced ­by3 which is based on the modification of the
Kozeny-Carman equation:
   
1 φe3
k= × (1)
KT Sv2gr (1 − φe )2

­ 2, φe is effective porosity, KT is the pore-level effective zoning factor and Svgr is the
where k is permeability in m
specific surface area per unit grain volume. The KT parameter is a function of pore size and shape, grain size
and shape, pore and grain distribution, tortuosity, cementation, and pore system (intergranular, intracrystalline,
vuggy, or fractured)13.
The HFU approach uses the normalized porosity index or the void ratio ( φz ) and reservoir quality index
(RQI) to predict permeability. The method involves plotting φz against RQI on a log–log scale and fitting a unit
slope trend line. The Flow Zone Indicator (FZI), which characterizes the geological and petrophysical attributes
of a given HFU, is determined by the intercept value of the trend line at ( φz) = 1. The previous parameters are
calculated using the following equations.

k
RQI = 0.0314 (2)
φ

φe
φz = (3)
1 − φe

1 RQI
FZI = √ = (4)
FS τ Sgv φz

where k is permeability in mD, φe is effective porosity in fractions, Fs is the shape factor, τ is the tortuosity, Sgv is
the surface area per unit grain in μm. The permeability can be recalculated based on the flow unit of a sample,
considering the FZI and effective porosity, using the following equation.
φe3
k = 1014.24(FZI)2 (5)
(1 − φe )2
When the samples for a given HFU are closely aligned with the trend line, the FZI value is equal to or close
to the FZI arithmetic average of these samples, and the predicted permeability is identical to the measured one.
However, if the samples are scattered around the trend line, the FZI value differs greatly from the FZI arithmetic
average, and the predicted permeability is far from the measured one, with a significant e­ rror14. Fine-grained
rocks, poorly sorted sands, rocks with authigenic pore filling, pore filling, and pore bridging clays are more
­ y3. The shape factors and tortuosity
likely to have a large surface area and a high degree of tortuosity, as stated b
of coarse-grained, well-sorted sands are much lower. Integrating FZI with other well logs and core data enables
the classification of HFUs, leading to more accurate reservoir characterization and better reservoir management.

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 3

Vol.:(0123456789)
 www.nature.com/scientificreports/

Machine learning description

Supervised machine learning
Random forest (RF)
Random forest is implemented via bootstrap a­ ggregation15. The bagging is based on the concept of building
multiple decision trees independently from one another using a subset of the input predictor parameters and a
bootstrap sample of the training d­ ata16. It randomly selects the training dataset Tb (b = 1, …, B) from the whole
training set T with replacement (bootstrapping sampling) and randomly selects M features or input variables
from P input variables or (M < P)17,18. By following these steps, the proxy model’s bias, excess variance, and overfit-
ting will be reduced to acceptable levels. Like decision trees, random forests are effective at resolving non-linear
patterns within data while also being scalable and resistant to outliers in imbalanced ­datasets19–23.
For each tree within the Classification and Regression Tree (CART) framework, the ideal division is calcu-
lated using a random selection of both Tb and P features. The collective set of these trees can be expressed as
an ensemble.
 
φTb ,m | b = 1, · · · , B (6)
In the regression approach used by the Random Forest algorithm, the final prediction is derived through an
averaging process rather than majority voting. The prediction Y  for a given input X is calculated as the average
of the predictions from all the individual trees in the ensemble:
B

 = φT,P (X) = 1
Y φTb ,m (X) (7)
B
b=1

This equation suggests that the collective prediction is the meaning of the outcomes from each of the B
individual Classification and Regression Trees (CART) that constitute the forest. By averaging, Random Forest
harnesses the diversity of the ensemble, effectively reducing the overall prediction error. This method capitalizes
on the ensemble’s ability to minimize the average squared error across the predictions, often resulting in a more
­ utput18,23.
accurate prediction than any single tree’s o

Extreme gradient boosting (XGB)

The gradient boosting approach is a robust ensemble training algorithm designed for both non-linear classifi-
cation and regression applications by upgrading a weak learning model into a strong l­earner24–26. The primary
objective of the gradient boosting approach is to identify a new sub-model with a lower error rate than the pre-
vious model. Hence, this method relies on the use of multiple models (bagging) which are trained to minimize
errors from the previous m­ ethod17,27.
One of the most well-known gradient-boosting enhancements is Extreme Gradient Boosting (XGB), which
employs Gradient Boosting Decision Trees (GDBT)28. This method avoids overfitting because it considers more
regularization terms than standard gradient tree boosting. Furthermore, it enhances model robustness by employ-
ing sampling techniques across both rows and columns, effectively diminishing the model’s v­ ariance29. A key
factor in XGBoost’s effectiveness is its ability to scale efficiently across various configurations. The ensemble
model of XGBoost is formulated in an additive manner.
K

yi = fK (xi ), fK ∈ F (8)
K=1

where f symbolizes a specific tree within the space F , which encompasses the entire set of regression trees. Here
xi , ignifies the i -th eigenvector, and K is the total count of trees in the model. The expression of cost function
presented as follows:
     
L(j) = l yi , 
yi + � fK (9)
i K
 
where The sum of the loss function l yi , 
yi , measuring the difference between the observed yi and the predicted
yi values and  denotes the regular punishment. The regularization term  itself is further defined as a combina-

tion of two components.
1
�(fK ) = γ T + ω2 (10)
2
γ T , where γ is the coefficient penalizing the complexity of the model by the number of leaf nodes T , enforcing
the 1 norm, while 12 ω2, with  as the coefficient for the 2 norm and ω as the leaf weight.

Support vector machines (SVM)

Support vector machines are based on the inductive concept of structural risk minimization (SRM), which allows
for reasonable generalizations to be made from a limited set of training ­examples30–33. This method utilizes a
margin-based loss function to control the input space dimensions and a kernel function to project the prediction
model onto a higher-dimensional space.
A support vector regressor (SVR) is a member of the Support Vector Machine, which has extremely potent
and flexible performance, is not confined to linear models, and is resistant to outliers. This method utilizes the

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 4

Vol:.(1234567890)
www.nature.com/scientificreports/

kernel trick to translate the original data into a higher-dimensional space without explicitly declaring the higher
­dimension34,35. This method’s compatibility with linear models (using linear kernels) or non-linear models (using
polynomial or radial kernels) makes it extremely v­ ersatile17. The effectiveness of the SVR relies heavily on the
model selection and kernel function settings (C, Gamma, and Epsilon)36.
The introduction of Vapnik’s epsilon-insensitive loss function has enabled Support Vector Regression (SVR)
to effectively address nonlinear regression estimation challenges. This approach involves the approximation of
given datasets using this specialized loss function.
   
D = x 1 , y 1 , . . . , x l , y l , x ∈ Rn , y ∈ R (10)

With a linear function

f (x) = �ω, x� + b, ω ∈ X, b ∈ R (11)
where the dot product in X is denoted by , . SVR aims to find a function f (x) that approximates output values
within a deviation of ∈ from the actual training data. The choice of ε is crucial, as smaller values lead to tighter
models that penalize a larger portion of the training data, while larger values result in looser models with less
penalization. The ideal regression function is identified by addressing an optimization problem, which is designed
to calculate the values of ω and b:
 l
1 
minimize � ω�2 + C ξi + ξi∗ (12)
2
i=1


yi − �ω, xi � − b ≤ ε + ξi
subject to �ω, xi � + b − yi ≤ ε + ξi∗ (13)
ξi , ξi∗ ≥ 0
where ξi and ξi∗ are the slack variables, and the model parameters ω and b. This approach balances minimizing
training error and penalizing model complexity, thus controlling the generalization error. The regularization
constant C in the optimization formulation helps to trade off between these two aspects. The epsilon-insensitive
loss function further adds to this balance by penalizing errors only when they exceed ε. This methodology allows
SVR to achieve better generalization performance compared to some other models like neural n ­ etworks37.

Artificial neural network (multi‑layer perceptron)

Artificial Neural Network (ANN) or multi-layer perceptron is one of the most effective machine learning
approaches. Its mathematical design is inspired by biological neural networks. This technique consists of 3 main
layers, the input layer is aimed for receiving input information from X variable. This data will be received and
learned by the hidden layer. This information will be generated by the output layer as a consequence of t­ esting38–41.
This study will concentrate on feed-forward back-propagation neural networks, one of the numerous forms
of neural networks. In this method, the input information flows in a forward manner from the input layer to the
hidden layer and ends up in the output layer. The errors that arise during this procedure will be calculated and
backpropagated by resetting the network’s weight and bias. It is an iterative procedure until the finest inaccuracy
is ­discovered34,41,42.
Hagan and c­ olleagues41 stated that single cycle of the process is described by the following equation.
Zk+1 = Zk − αk gk (14)
where gk represents the current gradiend, Zk denotes the current set of weights and biases, and αk is the learn-
ing rate. To adjust the connection weights for a specific neuron i during a particular iteration p, the following
equation outlines the ­process43.
wi (p + 1) = wi (p) + �wi (p) (15)
This equation updates the weight of the i -th neuron for the next iteration (p + 1) by adding a weight correc-
tion factor �wi (p) to the current weight wi (p).
The weight correction factor �wi (p) is calculated based on the equation.
�wi (p) = αxi (p)e(p) (16)
For the j-th neuron in a hidden layer γi, alternate expression for the weight correction factor �wi (p) is defined
as ­follows43,44.
wi (p) = αγi (p)δk (p) (17)
where δk (p) denotes the error gradient at neuron k in the output layer during the iteration p. This equation is
commonly referred to as the delta rule.

Unsupervised machine learning

K‑means clustering
In this study, the K-Means algorithm is exclusively used as an unsupervised machine learning technique. It is
selected for its simplicity and widespread application in clustering tasks. The algorithm minimizes a performance

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 5

Vol.:(0123456789)
 www.nature.com/scientificreports/

criterion called P, which is calculated as the sum of squared error distances between data points and their cor-
responding cluster c­ enters45. The algorithm begins with a random initial partition, and patterns are then reas-
signed to clusters based on their similarity to the cluster centers until a convergence criterion is satisfied, such
as no further reassignments or a significant reduction in squared error after a certain number of i­ terations46. The
squared error for a clustering L of a pattern set H containing K clusters is as follows.
nj
K 
 j 2
e2 (H, L) = �xi − cj � (18)
j=1 i=1
j
where xi is the ith pattern belonging to the jth cluster and cj is the centroid of the jth cluster.
In this study, along with the K-means algorithm, the Gaussian Mixture Model will also be implemented to
reinforce the confidence in the outcomes derived from the K-means clustering. The Gaussian Mixture Model
(GMM) offers a probabilistic approach to clustering, presenting the advantage of accommodating clusters of
different sizes and orientations due to its use of covariance matrices. This capability enables the GMM to identify
and adapt to elliptical or anisotropic clusters, unlike simpler algorithms like k-means which assume isotropic
clusters. Additionally, GMM provides a soft-clustering approach, assigning probabilities of membership to each
point for all clusters, rather than forcing a hard assignment. This results in a more nuanced understanding of the
data’s structure, particularly useful when the relationship between variables is complex and not easily separable
into distinct g­ roupings47. Hence, incorporating both K-means and Gaussian Mixture Model (GMM) methods
in a single study leverages the strengths of both clustering techniques.

Methodology
Data acquisition
This study analyzes open-source data consisting of thousands of well logs from the Halibut Oil Field, which are
supplemented with routine core analysis studies. A total of 212 data sets are chosen for analysis, based on the
specific formation depth and the availability of porosity and permeability data at that depth. These data sets
encompass 17 different types of well-log information, including Corrected Gamma Ray (CGR), Bulk Density
Correction (DRHO), Delta-T Compressional (DT5), Gamma Ray (GR), High-Resolution Enhanced Thermal
Neutron (HNPO), Laterolog Deep Resistivity (LLD), High-Resolution Laterolog Resistivity (LLHR), Laterolog
Shallow Resistivity (LLS), Mud Resistivity (MRES), Micro Spherically Focused Resistivity (MSFC), Thermal
Neutron Porosity (NPHI), Enhanced Thermal Neutron Porosity (NPOR), Potassium Concentration (POTA),
Bulk Density (RHOB), Spontaneous Potential (SP), Thorium Concentration (THOR), and Uranium Concentra-
tion (URAN), alongside porosity and permeability data. The focus of this study is the FZI parameter, which is
directly influenced by permeability.
It is acknowledged that the FZI exhibits a non-normal distribution, as evident from Fig. 1. Consequently,
predicting the FZI directly could potentially lead to misleading results due to its extremely non-normal distribu-
tion. To address this issue, an approach is taken to transform the FZI values using a logarithmic scale, aiming
to approximate a normal distribution, as illustrated in Fig. 1. To provide an initial understanding of the data,
Table 1 presents the data statistics, while Fig. 1 showcases the distribution of each parameter considered in the
study and a pair chart for the input versus the output parameter. The cross plot between the input and the output
parameters in Fig. 1b shows a linear (in orange) and nonlinear (in black) relationship between the output and
input parameters.
Eighteen parameters (including the LOGFZI) were chosen at the initial phase of this study, as shown in
Table 1. It is necessary to reduce the number of parameters to optimize the model’s dimensionality and improve
its processing ­time48. However, initially applying all 18 input factors will allow for a more comprehensive under-
standing of how these parameters affect the precision with which the machine learning model predicts the flow
zone indicator. When the connection between input factors and model accuracy is better understood, it’s possible
to reduce the number of parameters and thereby boost model efficiency. The correlation coefficient analysis of
each input parameter to the output parameter of LOGFZI is presented in Fig. 2. The heat map was generated
using seaborn.heatmap python library. Figure 2a presents Pearson’s correlation coefficients, that highlight the
linear relationship between the parameters with each other, while Fig. 2b presents the Spearman’s correlation
coefficients, that was used to exclude the nonlinearity and outliers’ effect. The correlation coefficients for most
parameters remained consistent, except in a few instances where the correlation either increased or decreased
when Spearman’s coefficient was calculated compared to Pearson’s coefficients. This variation can be attributed
to the presence of outliers or nonlinear relationships. For instance, the correlation for DTS slightly increased
from -0.1 to -0.3, indicating a more negative relation with LogFZI. Similarly, the LLD coefficient increased from
− 0.7 to − 0.8 due to the nonlinear relation between LogFZI and LLD. Conversely, the correlation for RHOB
decreased from − 0.7 to − 0.4.
Data normalization needs to be done to improve integrity and reduce data redundancy especially for the
algorithm that basically relies on the distance technique (KNN and SVR). This is normally done because the
input and output data used in the study have very large unit and range differences. The normalization technique
employed in this study is the MinMaxScaler. A significant benefit of this scaler is its capability to preserve the
original shape of the dataset’s distribution. This preservation is critical as it ensures the integral information
within the data remains unaffected during scaling. Unlike several other scaling methods, MinMaxScaler does
not alter the core characteristics of the original data, thus maintaining the crucial details and patterns necessary
for accurate analysis. The normalization formula applied is:

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 6

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 1.  Histogram of 17 well-log parameters, illustrating the diverse distribution types for each parameter.
The LOGFZI distribution demonstrates a closer resemblance to a log-normal distribution compared to the
original FZI distribution.

x − xmin
xnorm =
xmax − xmin (19)

where xnorm is a normalized value with a range of values 0–1, x is the variable on the dataset while max and min
refers to the maximum and minimum value of the v­ ariable42,49.

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 7

Vol.:(0123456789)
 www.nature.com/scientificreports/

No. Parameter Count Mean Standard deviation Minimal Maximal

1 LOGFZI 212 2.31 0.71 0.65 3.19
2 FZI 212 456.23 400.79 4.50 1536.15
3 CGR​ 212 45.03 24.70 14.16 123.35
4 DRHO 212 0.03 0.01 0.00 0.09
5 DT5 212 80.31 3.97 73.31 95.14
6 GR 212 55.62 26.47 27.68 138.58
7 HNPO 212 0.15 0.04 0.09 0.34
8 LLD 212 2.20 1.69 1.08 10.04
9 LLHR 212 2.91 2.73 1.15 17.12
10 LLS 212 2.08 1.67 0.88 9.04
11 MRES 212 0.07 0.00 0.07 0.07
12 MSFC 212 2.45 2.72 0.87 16.75
13 NPHI 212 0.15 0.04 0.08 0.35
14 NPOR 212 0.15 0.04 0.09 0.34
15 POTA 212 1.55 0.80 0.19 3.86
16 RHOB 212 2.43 0.08 2.18 2.67
17 SP 212 2.18 1.51 − 1.56 4.92
18 THOR 212 4.83 3.15 1.75 15.29
19 URAN 212 1.66 1.20 0.12 5.32

Table 1.  Statistical summary of 17 well-log parameters, including FZI and LOGFZI.

Machine learning design

Supervised machine learning design
In the supervised machine learning section of this study, 212 datasets will be split into two groups: 65% for train-
ing purposes and 35% for testing. As an effort to prevent overfitting and leakage on the testing data both holdout
and k-fold cross-validation method are adopted in this study to serves a dual purpose: ensuring an unbiased
evaluation of the model and a thorough assessment of its generalizability. The holdout method provides a clean
dataset for final model evaluation, free from any influence of the training process. Meanwhile, k-fold cross-
validation is applied to the training data to reduce the potential variance in model performance that could result
from a single train-test split, particularly important in datasets of limited size. This nested approach is a robust
strategy for hyperparameter tuning, enabling the model to demonstrate consistent performance across multiple
subsets of the data, thus reinforcing its ability to generalize beyond the training sample. In this scenario, the
model will continue to be trained until all folds have been used for testing once. The average score of the testing
fold will recognize as validation score (Fig. 3)50.
The algorithm’s hyperparameters will also be tuned to determine the optimal model for each m ­ ethod18. The
set of each individual’s input hyper-parameters is displayed in Tables 2, 3, 4 and 5.

Unsupervised machine learning design

In the unsupervised machine learning section, the distribution of the log FZI data will be examined through a
histogram and a normal probability plot to make initial judgments regarding data clustering. A statistical method
incorporating the normal probability plot will be employed, where a straight line in the plot signifies a normal
distribution. If multiple straight lines with varying slopes are present, it indicates the existence of different data-
sets that share the same normal distribution, implying the presence of distinct clusters.
To determine the optimal number of clusters in the K-Means algorithm, the elbow criterion is utilized. The
elbow criterion suggests selecting the number of clusters where the addition of another cluster does not signifi-
cantly contribute new ­information51. In this study, the elbow method incorporates the Root Mean Square Error
(RMSE) and R-squared as measures to evaluate the clustering of flow u ­ nits52. These metrics quantify the devia-
tion between observed and estimated values, providing insights into the optimum cluster numbers for reservoir
characterization of hydraulic flow units. Several previous studies have utilized the elbow method in conjunction
with the RMSE and R-squared metrics to determine the optimal number of clusters for hydraulic flow units in
reservoir characterization ­efforts6,53,54.
It is crucial to follow an organized process to obtain precise and trustworthy outcomes. Figure 4 displays the
precise study methodology in detail.

Result and discussion

Supervised machine learning for flow zone indicator prediction
Pre‑processing machine learning model
In this work, a predictive model is constructed using three-fold cross-validation and hyperparameter optimiza-
tion. Randomized search cross-validation is used as a solution to Grid Search Cross-Validation (Exhausted Cross
Validation) to reduce computation time when performing ­hyperparameters55. Using this approach, 50 iterations

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 8

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 2.  Heatmap of correlation coefficients between each parameter, illustrating the strength of correlation
for all parameters; (A) Pearson’s Coefficients, (B) Spearman’s coefficients. The LOGFZI exhibits noticeable
strong correlations (mostly negative) with several parameters, (Heat map was generated using seaborn. heatmap
python library).

of hyper-parameters are paired with 3 folds of cross-validation to generate 150 training models, which are then
assessed using coefficient determination metric evaluation (­ R2).

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 9

Vol.:(0123456789)
 www.nature.com/scientificreports/

Figure 3.  Threefold cross-validation scenario.

Hyper parameter Code/symbol Minimum Maximum Best value

Maximum tree algo__n_estimators 100 200 151
Maximum depth algo__max_depth 20 80 70
Maximum features algo__max_features 0.1 1 0.12828
Minimum sample leaf algo__min_samples_leaf 1 20 7

Table 2.  Hyper-parameter tuning for random forest (RF).

Hyper parameter Code/symbol Minimum Maximum Best value

Maximum tree algo__n_estimators 100 200 185
Maximum depth algo__max_depth 1 20 3
Learning rate algo__learning_rate −2 0 0.12164
Maximum features algo__max_features 0.1 1 0.7865
Gamma algo__gamma 1 10 1
Alpha algo__reg_alpha −3 1 0.0115
Lambda algo__reg_lambda −3 1 0.0094

Table 3.  Hyper-parameter tuning for extreme gradient boosting (XGB).

Hyper parameter Code/symbol Minimum Maximum Best value

Gamma algo__gamma −3 3 0.4374
Regularization parameter algo__C −3 3 12.746

Table 4.  Hyper-parameter tuning for SVM (support vector machine).

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 10

Vol:.(1234567890)
www.nature.com/scientificreports/

Hyper parameter Code/symbol Minimum Maximum Best values

Hidden layer size mlp__hidden_layer_sizes (16, 8, 4) (8, 4) (16, 8, 4)
Learning rate mlp__learning_rate_init 0.001 0.01 0.01
Alpha mlp__alpha 0.0001 0.003 0.0003

Table 5.  Hyper-parameter tuning ANN (artificial neural network).

Figure 4.  Workflow of the study for supervised and unsupervised machine learning models.

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 11

Vol.:(0123456789)
 www.nature.com/scientificreports/

The initial investigation will compare models that have undergone scaling to those that have not. Evalua-
tion metrics such as R-squared (­ R2), Mean Squared Error (MSE), Mean Absolute Error (MAE), and Root Mean
Squared Error (RMSE) for each algorithm are compiled in Tables 6, 7, 8, and 9 and illustrated in Figs. 5, 6, 7,
and 8.
The outcomes presented in Figs. 5 and 7 indicate that applying scaling techniques improves the performance
of both the SVM and ANN models, with notable enhancement observed in the ANN. The SVM is recognized for
its robustness, employing a margin-based loss function that effectively manages the dimensionality of the input
space. However, SVM may underperform with skewed datasets as finding the optimal separating hyperplane
becomes challenging with imbalanced d ­ ata56. A similar challenge is observed with neural network algorithms,
which, at their core, rely on linear regression principles. Extreme skewness in the data can substantially impact the
performance of neural networks. Meanwhile the stability of the scores for both the Random Forest and XGBoost
models, even after dataset standardization, can be attributed to their foundational decision tree structure. These
models utilize bootstrapping sampling methods and an aggregation technique known as bagging to produce the
final score. This approach equips the models with resilience against imbalanced or skewed datasets, ensuring
consistent performance irrespective of data ­standardization19–22.

Data processing and features reduction

To enhance machine learning model accuracy, various data processing techniques are used. Parameter reduc-
tion is achieved by analyzing the impact of excluding each variable using the feature importance method. The
Feature Importance Analysis is performed using the random forest model as the benchmark to identify the most
important parameters in the dataset. The random forest model is selected for its high accuracy, as indicated by
the high R-squared values observed during the pre-processing stage of the data in both the training and testing

Model Training Validation Testing

Random forest 0.9106 0.8310 0.8850
XGB 0.886 0.796 0.8780
SVM 0.912 0.775 0.8800
ANN 0.469 0.3805 0.4458

Table 6.  Coefficient determination ­(R2) summary (before scaling).

Model MSE MAE RMSE

Random forest 0.06 0.19 0.25
XGB 0.06 0.21 0.25
SVM 0.06 0.19 0.25
ANN 0.29 0.44 0.54

Table 7.  Error summary (before scaling).

Model Training Validation Testing

Random forest 0.91 0.83 0.89
XGB 0.89 0.80 0.88
SVM 0.92 0.84 0.88
ANN 0.81 0.78 0.82

Table 8.  Coefficient determination ­(R2) summary (before scaling).

Model MSE MAE RMSE

Random forest 0.06 0.19 0.25
XGB 0.06 0.21 0.25
SVM 0.06 0.18 0.25
ANN 0.09 0.24 0.31

Table 9.  Error summary (after scaling).

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 12

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 5.  Coefficient determination (­ R2) summary for the different ML methods using unscaled datasets.

sets. Figure 9 presents the results of the analysis, showing the relative importance of each input parameter in
the dataset.
Figure 9 displays the relative importance of each parameter in the output model, as determined by a feature
importance analysis using a random forest model. This analysis calculates the decrease in the Mean Squared
Error (MSE) of the prediction, where a higher importance score indicates a greater role of the parameter in
reducing the M ­ SE19–22. LLHR emerges as a notable parameter with an importance score of 27%. It is important
to understand that this score does not imply that excluding LLHR would directly result in a 27% change in
the model’s performance. Instead, it signifies LLHR’s relative contribution to enhancing the model’s predictive
accuracy by reducing the MSE. The parameter selection process in this study was guided by the aim to include
input parameters that collectively have a substantial impact on the model’s effectiveness. The cumulative rela-
tive importance from LLHS to HNPO is 51%, indicating their combined significance in the model. Therefore,
the final set of selected input parameters, comprising LLHR, LLS, MSFC, LLD, CGR, NPHI, THOR, NPOR,
and HNPO, was chosen based on their collective ability to decrease the MSE and improve the model’s overall
predictive performance, rather than solely on their individual importance scores.
The feature importance analysis results are in line with the existing literature, as these parameters demon-
strate a strong relationship with the calculation of FZI using a physics-based approach. The LLHR (Laterolog
High-Resistivity), LLS (Laterolog Shallow), and LLD (Laterolog Deep) logs are crucial resistivity logs utilized in
formation ­evaluation57 explored the relationship between resistivity and permeability using known water satu-
ration and the apparent formation factor. The results of the study demonstrated a strong relationship between
resistivity and permeability. The MSFC log provides quantitative resistivity data at a micro-scale and can be
converted into visual images, allowing for detailed core permeability description through visual examination.
­Bourke58 observed a strong visual correlation between micro resistivity and permeability images, indicating their
potential for capturing porosity–permeability variations. Micro resistivity data offer high-resolution permeability
transformation, surpassing traditional logs, and have been used in various studies for permeability assessment
and characterization. These findings highlight the significance of the MSFC log in permeability prediction.
Yao and ­Holditch59 established a correlation between core permeability and open-hole well-log data, high-
lighting the significance of the relationship between gamma-ray and permeability estimation, which ultimately
contributes to the estimation of FZI. Thus, CGR is an important parameter in this model. NPHI, NPOR, and
HNPO are different versions of thermal neutron porosity logs widely used for characterizing reservoir poros-
ity. These logs have been extensively studied in combination with other parameters to determine lithology and

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 13

Vol.:(0123456789)
 www.nature.com/scientificreports/

Figure 6.  Error summary for the different ML methods using unscaled datasets.

estimate clay volume that reflects its importance for FZI prediction. The THOR (Thorium Concentration) log
measures the thorium concentration in parts per million (ppm) using energy emissions from radioactive miner-
als, which are detected by the spectral gamma ray log. According ­to60, high concentrations of thorium are indica-
tive of dark, organic-rich shale, as well as calcareous, brittle, and fractured shale. Hence, Thorium concentrations
directly influence permeability and porosity and the rock type.
In addition to parameter reduction, data transformation using the Yeo-Johnson method was applied to the
dataset. This transformation technique is employed to address the issue of non-normality in the data distribution.
By employing this transformation, the data distribution becomes more symmetrical, thus meeting the assump-
tions of certain statistical models and improving the accuracy of subsequent analyses.

Post‑data‑processing machine learning model

In this step, the machine learning model proceeds to apply the same characteristic (hyper-parameter combina-
tion) as the previous model. The following Figs. 10, 11, 12 and 13 represent the results of the machine learning
algorithm that was applied following data processing. These cross-plot figures showed the capabilities of the
different machine learning to predict the flow zone index, where most of the data are aligned with the 45-degree
line. Additionally, Figs. 14 and 15 and Tables 10 and 11 summarize the comparison models’ performance post
scaling and transformation process.
Model evaluations demonstrate steady efficacy throughout the training, validation, and testing stages. The
Random Forest model stands out with the highest accuracy in training and testing, at 0.9566 and 0.9081, respec-
tively. Table 11 and Fig. 14 collectively suggest that the models retain high accuracy post-data processing. In
Table 12 it can be seen the comparison of the final model to the initial model which did not undergo data process-
ing, the final model that incorporated scaling and transformation exhibited enhancements. This is particularly
noticeable in the case of the ANN model, which, as previously discussed, showed significant improvement. Due
to the highest model performances resulting from the final model, it is recommended to use the post-processed
models for future research, as they offer a well-tuned blend of dimensionality reduction and predictive capability.

Unsupervised machine learning for hydraulic flow unit classification

Initial observation
Figure 16 displays the histogram plot of FZI value, showing a non-normal distribution. Despite attempts to
transform the heavily non-normal FZI data to log FZI (Fig. 17), the resulting distribution remains non-normal
due to FZI being influenced by the direction of fluid flow (permeability) and requiring further averaging or

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 14

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 7.  Coefficient determination (­ R2) summary for the different ML methods using scaled datasets.

upscaling methods. Consequently, determining the number of hydraulic flow units (HFUs) solely from this plot
is challenging. The histogram represents overlapping individual normal distributions, necessitating the isolation
and identification of these individual distributions to accurately estimate the number of ­HFUs54. Therefore, while
the histogram provides insights into the variation of HFU distribution across the formation, it offers a qualitative
analysis rather than a precise count of HFUs.
The normal probability plot is used as a statistical technique to assess the normality of a dataset. The presence
of multiple straight-line segments in the plot indicates the presence of different hydraulic flow units (HFUs),
each with its distinct normal distribution. Figure 18 displays nine distinct straight lines, suggesting the existence
of nine HFUs in the formation. However, it’s important to note that this approach relies on statistical analysis
and visual interpretation, which can be subjective. Caution should be exercised when interpreting the results.
Despite its limitations, this method is a valuable tool in data analytics and provides insights into the properties
and behavior of a reservoir.

The optimum cluster number

In the initial stage of K-Means clustering, the elbow method is utilized to determine the optimal number of flow
units (clusters). In this study, the elbow method uses RMSE and R-squared evaluations to determine the optimal
number of flow u ­ nits61. The results of the elbow method plot are displayed in Fig. 19.
Both the RMSE and R-squared approach may provide a different interpretation of the optimal HFU value.
Considering the previously assessed heterogeneity, for the RMSE method, the optimum HFU value is taken
as the number that has a minimum difference of 10% from the previous HFU value. Thus, the optimum HFU
number for RMSE is 10, as, at 11, the value drops below 10%. In contrast, the R-squared method shows very
small differences between R-squared values for each HFU number. Therefore, the interpretation of the R-squared
method relies on visually observing the plot itself. By examining the plot, it is evident that an HFU of 10 exhibits
the most horizontal straight line among all the previous ones. Consequently, the HFU value of 10 is considered
the optimum value based on both the RMSE and R-squared approaches.

The K‑means clustering

After determining the optimum cluster number, the K-Means clustering algorithm is utilized. The selected opti-
mum HFU value is 10, and to ensure consistent and reproducible results, the random state parameter is set to
42 during the initialization of the K-Means clustering model. This parameter controls the random initialization
of cluster ­centroids62. By using a specific random seed, the same initial centroids are used each time the code

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 15

Vol.:(0123456789)
 www.nature.com/scientificreports/

Figure 8.  Error summary for the different ML methods using scaled datasets.

Figure 9.  Feature importance analysis quantifying the significance of each parameter in the model
construction.

runs, ensuring consistent results and facilitating debugging and result comparison. Once the K-Means model is
initialized with 10 clusters and the random state is set, the model is fitted to the data, allowing for further analysis
and interpretation based on the resulting cluster assignments. The fitted data is then plotted in a log–log plot of
RQI vs PHIZ, with each FZI value corresponding to its respective HFU unit.
Figures 20 and 21 display the results of dataset clustering using the K-means algorithm and the Gaussian
Mixture Model (GMM), respectively. Upon examination of the figures, it is apparent that both algorithms have

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 16

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 10.  Coefficient determination (­ R2) random forest model.

Figure 11.  Coefficient determination (­ R2) XGB model.

delineated similar clusters within the dataset. Although the labeling of the clusters differs between the two
methods, the composition of the data points within corresponding clusters is largely analogous. This consistency
between the K-means and GMM clustering outcomes suggests that both methods are capturing the inherent
groupings within the dataset effectively. The parallelism in results reinforces the reliability of the clustering,
affirming that the dataset possesses a structure that is robust to the clustering technique applied. The congruence
of these clustering methods provides a validated foundation for further analysis.
The performance of the evaluation and clustering method was assessed by calculating the permeability using
the FZI values for each flow unit cluster by using Eq. (5). The calculated permeability was then compared to the
actual permeability. Figure 22 displays the comparison between predicted and actual permeability values. The
results indicate a high R-squared value of 0.93, demonstrating the effectiveness of the clustering method. This
outcome validates the evaluation of reservoir heterogeneity, the determination of optimum HFU numbers, and
the utilization of FZI for clustering. Table 13 provides the average permeability and porosity values for each
flow unit cluster. It is important to note that when addressing heterogeneity, the choice of averaging method
(arithmetic, harmonic, geometric) for permeability depends on the distribution of permeabilities within the rock

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 17

Vol.:(0123456789)
 www.nature.com/scientificreports/

Figure 12.  Coefficient determination (­ R2) SVM model.

Figure 13.  Coefficient determination (­ R2) ANN model.

during ­deposition63. By examining the FZI values alongside their respective average permeabilities, it is possible
to predict the permeability quality of a specific location, thus enabling an assessment of its potential for fluid flow.

Models validation and applications in unsampled formations

Figure 23 shows the results of applying four models to the additional unseen location. The trend reveals that the
ANN model performs the poorest, followed by the SVM, while the XGB and RF models exhibited the highest
performance. This result is consistent with the values presented in Tables 10 and 11, which showed that the RF
model is the optimal choice for predicting the FZI. This decision is based on the highest R-squared and lowest
error score values obtained from both the training and testing datasets, and with their proximity indicating
good generalization.
Several literatures further support this decision by acknowledging the acceptability of different models after
data processing, as each model possesses strengths based on the nature of the data. The superior performance of
the Random Forest model over others, particularly SVM and ANN, in this study is likely due to the characteristics

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 18

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 14.  Coefficient determination (­ R2) summary (final model).

of the dataset. Moreover, the dataset size used in this study is relatively small, comprising only 159 training and
53 testing data points. This condition is disadvantageous for algorithms like SVM, which depend on the spatial
dimensions of the data, and for ANN models, which are based on fundamental linear regression ­calculations56,64.
However, this limitation does not significantly impact the Random Forest algorithm, which employs bootstrap-
ping sampling and a technique called bagging for final score computation. This method allows the algorithm
to randomly select the training dataset from the whole training set with replacement and randomly selects M
features or input variables from input variables. Such as methodology makes the model robust against imbalanced
or skewed datasets, ensuring stable performance regardless of data ­standardization19–23.
It’s crucial to understand that the results of the machine learning models in this study are specific to the
dataset employed and should not be generalized. The effectiveness of each algorithm heavily depends on the
characteristics of the data used, meaning Random Forest may not always outperform other algorithms in dif-
ferent scenarios.
Figure 24 illustrates the application of the ML models in an unseen dataset, and the forecasting of FZI value
using a random forest model in an unsampled location. Figure 24a reveals a noticeable resemblance between
the predicted and observed trends in the FZI data. This finding holds significant value for reservoir modeling
scenarios. Upon careful examination of Fig. 24b, two distinct depth ranges emerge as potential reservoir develop-
ment zones. The first zone, represented by the red box, has an approximate thickness of 10 ft and an average FZI
value of 500. Based on the clustering analysis presented in Table 13, it is likely associated with HFU number 8,
which displays a harmonic average permeability value of 2806 millidarcy (mD). The second zone, indicated by
the blue box, spans approximately 15 ft, and exhibits an average FZI value of around 800. Referring to Table 13,
HFU number 9 is linked to an FZI value of around 800, suggesting the presence of a zone characterized by
remarkably high permeability, measuring approximately 6410 mD. These significant findings strongly indicate
the existence of favorable reservoir zones within the delineated areas. By combining the clustering analysis of
HFUs and employing machine learning models to predict FZI based on well-log data, it becomes possible to
estimate potential reservoir characterization zones. However, for an optimized approach to hydrocarbon recov-
ery, it is imperative to consider additional petrophysical properties such as water and hydrocarbon saturation.
Furthermore, accurate calculations of the initial hydrocarbon in place within these predicted potential zones
should be incorporated into the analysis.

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 19

Vol.:(0123456789)
 www.nature.com/scientificreports/

Figure 15.  Error summary (final model).

Model Training Validation Testing

RF 0.96 0.84 0.91
XGB 0.88 0.80 0.86
SVM 0.89 0.84 0.89
ANN 0.82 0.80 0.87

Table 10.  Coefficient determination ­(R2) summary (final model).

Model MSE MAE RMSE

Random forest 0.05 0.17 0.22
XGB 0.07 0.22 0.27
SVM 0.06 0.19 0.24
ANN 0.07 0.22 0.26

Table 11.  Error summary (final model).

Model Training (%) Validation (%) Testing (%)

RF 5.05 1.22 2.61
XGB − 0.27 − 2.13 0.72
SVM 2.25 8.45 1.66
ANN 74.43 109.15 94.57

Table 12.  Relative differences of initial and final model score ­(R2).

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 20

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 16.  The distribution of FZI value.

Figure 17.  The histogram for log FZI data.

Conclusion
This study utilized state-of-the-art machine learning methodologies to augment the efficacy of reservoir char-
acterization. The supervised learning algorithms, including Random Forest (RF), Extreme Gradient Boosting
(XGB), Support Vector Machines (SVM), and Artificial Neural Network (ANN), were used to predict Formation
Zone Indicator (FZI) values in unsampled locations, while unsupervised learning technique of K-Means and
Gaussian mixture clustering algorithm was employed to classify Hydrocarbon Flow Units (HFUs) in the reservoir.
The findings of this study are summarized as follows:

• The four implemented algorithms demonstrate robust performance in estimating the flow zone indicator
of the reservoir, yielding high coefficients of determination (­ R2) of 0.89 and 0.95 in the training and testing
datasets, respectively.
• The RF model emerged as the optimal choice for FZI prediction in unsampled locations, with R ­ 2 values of
0.957 for training and 0.908 for testing.

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 21

Vol.:(0123456789)
 www.nature.com/scientificreports/

Figure 18.  The normal probability plot for the log FZI data.

Figure 19.  The Elbow method plot illustrates the RMSE and R-squared values.

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 22

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 20.  HFU clusterization using K-means method.

• By combining the elbow method with K-Means clustering analysis, the study effectively delineated 10 unique
HFUs. Additionally, results from the Gaussian mixture clustering corroborate the observed clustering behav-
ior of HFUs.
• The RF model demonstrated strong performance in predicting FZI values in unsampled locations, revealing
two potential reservoir development zones:

– Zone 1 (2370 ft–2380 ft): Approximately 10 ft thick with an average FZI value of 500, associated with
HFU number 8.
– Zone 2 (2463–2477 ft): Spanning around 15 ft with an average FZI value of approximately 800, cor-
responding to HFU number 9 and indicating a zone characterized by remarkably high permeability.

The study’s findings hold significant implications for reservoir characterization practices in the petroleum
industry. The successful integration of machine learning, particularly Random Forest, into conventional meth-
ods allows for rapid and cost-effective reservoir assessments. This approach not only enhances decision-making
speed but also identifies specific zones with high-quality reservoir potential. The study showcases the robustness
of machine learning in petroleum engineering applications, marking a shift towards more efficient and accurate
reservoir characterization. To further advance the field, future research should explore additional machine
learning models and incorporate a broader set of features for a comprehensive analysis in addition to validating
the results on different datasets.

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 23

Vol.:(0123456789)
 www.nature.com/scientificreports/

Figure 21.  HFU clusterization using Gaussian mixture method.

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 24

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 22.  The predicted (calculated) vs actual permeability.

Permeability
HFU Number of samples FZI Mean porosity Arithmetic Harmonic Geometric
1 12 7 0.08 0.035 0.024 0.03
2 18 13 0.08 0.176 0.036 0.092
3 8 21 0.14 1.7 1.4 1.5
4 16 41 0.15 7.7 6.9 7.3
5 10 95 0.16 60 52 56
6 21 211 0.20 624 502 568
7 41 341 0.20 1444 1960 762
8 35 532 0.19 3794 2806 3334
9 40 887 0.18 8184 6410 7314
10 21 1200 0.18 14,299 12,511 13,446

Table 13.  Average porosity and permeability for each HFU.

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 25

Vol.:(0123456789)
 www.nature.com/scientificreports/

Figure 23.  Different algorithm performance for additional unseen data.

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 26

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 24.  (a) The comparison of random forest and other algorithm performance in unseen sample data, (b)
The prediction of FZI value using from unsampled data based on random forest model.

Data availability
Most of the data are presented in the manuscript, and a detailed sample will be provided upon request through
email “ahmed.ibrahim@kfupm.edu.sa”.

Received: 26 August 2023; Accepted: 18 February 2024

References
1. Ertekin, T. The efficacy and superiority of the expert systems in reservoir engineering decision making processes. Appl. Sci. 11,
14. https://​doi.​org/​10.​3390/​app11​146347 (2021).
2. Aminzadeh, F., Temizel, C. & Hajizadeh, Y. Applications in reservoir characterization and field development optimization. in
Artificial Intelligence and Data Analytics for Energy Exploration and Production, 271–311 (Wiley, 2022). https://​doi.​org/​10.​1002/​
97811​19879​893.​ch10.
3. Amaefule, J. O., Altunbay, M., Tiab, D., Kersey, D. G. & Keelan, D. K. Enhanced reservoir description: using core and log data to
identify hydraulic (flow) units and predict permeability in uncored intervals/wells. in Paper presented at the SPE Annual Technical
Conference and Exhibition, SPE (1993). https://​doi.​org/​10.​2118/​26436-​MS.
4. Rebelo, T. B., Batezelli, A., Mattos, N. H. S. & Leite, E. P. Flow units in complex carbonate reservoirs: A study case of the Brazilian
pre-salt. Mar. Pet. Geol. 140, 105639. https://​doi.​org/​10.​1016/j.​marpe​tgeo.​2022.​105639 (2022).
5. Zhang, Z., Zhang, H., Li, J. & Cai, Z. Permeability and porosity prediction using logging data in a heterogeneous dolomite reservoir:
An integrated approach. J. Nat. Gas Sci. Eng. https://​doi.​org/​10.​1016/j.​jngse.​2020.​103743 (2021).
6. Mohammadian, E., Kheirollahi, M., Liu, B., Ostadhassan, M. & Sabet, M. A case study of petrophysical rock typing and perme-
ability prediction using machine learning in a heterogenous carbonate reservoir in Iran. Sci. Rep. 12, 1. https://​doi.​org/​10.​1038/​
s41598-​022-​08575-5 (2022).
7. Man, H. Q. et al. Hydraulic flow unit classification and prediction using machine learning techniques: A case study from the Nam
Con Son Basin, Offshore Vietnam. Energies 14, 22. https://​doi.​org/​10.​3390/​en142​27714 (2021).

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 27

Vol.:(0123456789)
 www.nature.com/scientificreports/

8. Sebtosheikh, M. A., Motafakkerfard, R., Riahi, M. A., Moradi, S. & Sabety, N. Support vector machine method, a new technique
for lithology prediction in an Iranian heterogeneous carbonate reservoir using petrophysical well logs. Carbon. Evapor. 30(1),
59–68. https://​doi.​org/​10.​1007/​s13146-​014-​0199-0 (2015).
9. Salman, O., Hasan, O. F. & Al-Jawad, S. Permeability prediction in one of Iraqi carbonate reservoir using statistical, hydraulic flow
units, and ANN methods. Iraqi J. Chem. Pet. Eng. 23(4), 17–24. https://​doi.​org/​10.​31699/​ijcpe.​2022.4.3 (2022).
10. Abnavi, A. D., Torghabeh, A. K. & Qajar, J. Hydraulic flow units and ANFIS methods to predict permeability in heterogeneous
carbonate reservoir: Middle East gas reservoir. Arab. J. Geosci. 14, 754. https://​doi.​org/​10.​1007/​s12517-​021-​07084-5 (2021).
11. Adnan, M. & Abed, A. Hydraulic flow units and permeability prediction in a carbonate reservoir, southern Iraq from well log data
using non-parametric correlation. Sci. Technol. Eng. 3(1), 480–486 (2014).
12. Abbas, M. A. & Al Lawe, E. M. Clustering analysis and flow zone indicator for electrofacies characterization in the upper shale
member in Luhais Oil Field, Southern Iraq. in Paper Presented at the Abu Dhabi International Petroleum Exhibition & Conference
(2019). https://​doi.​org/​10.​2118/​197906-​MS.
13. Tiab, D. & Donaldson, E. C. Porosity and permeability. in Petrophysics, 188–189 (Elsevier, 2004). https://​doi.​org/​10.​1016/​B978-​
07506​7711-0/​50007-4.
14. Elnaggar, O. M. A new processing for improving permeability prediction of hydraulic flow units, Nubian Sandstone, Eastern Desert,
Egypt. J. Pet. Explor. Prod. Technol. 8(3), 677–683. https://​doi.​org/​10.​1007/​s13202-​017-​0418-z (2018).
15. Breiman, L. Bagging predictors. Mach. Learn. 24(2), 123–140. https://​doi.​org/​10.​1007/​BF000​58655 (1996).
16. Xue, L. et al. A data-driven shale gas production forecasting method based on the multi-objective random forest regression. J. Pet.
Sci. Eng. 196, 107801. https://​doi.​org/​10.​1016/j.​petrol.​2020.​107801 (2021).
17. Gupta, I., Devegowda, D., Jayaram, V., Rai, C. & Sondergeld, C. Machine learning regressors and their metrics to predict synthetic
sonic and brittle zones, 1–20 (2019). https://​doi.​org/​10.​15530/​urtec-​2019-​148.
18. Xue, L. et al. A data-driven shale gas production forecasting method based on the multi-objective random forest regression. J. Pet.
Sci. Eng. 196(2020), 107801. https://​doi.​org/​10.​1016/j.​petrol.​2020.​107801 (2021).
19. Hegde, C., Wallace, S. & Gray, K. Using trees, bagging, and random forests to predict rate of penetration during drilling. Soc. Pet.
Eng. https://​doi.​org/​10.​2118/​176792-​ms (2015).
20. Alhashem, M. Machine learning classification model for multiphase flow regimes in horizontal pipes. in International Petroleum
Technology Conference 2020, IPTC 2020 (2020). https://​doi.​org/​10.​2523/​iptc-​20058-​abstr​act.
21. Aulia, A., Rahman, A. & QuijanoVelasco, J. J. Strategic well test planning using random forest. Soc. Pet. Eng. 1, 140–162. https://​
doi.​org/​10.​2118/​167827-​ms (2014).
22. Hidayat, F. & Astsauri, T. M. S. Applied random forest for parameter sensitivity of low salinity water Injection (LSWI) implementa-
tion on carbonate reservoir. Alexandria Eng. J. 61(3), 2408–2417. https://​doi.​org/​10.​1016/j.​aej.​2021.​06.​096 (2022).
23. Breiman, L. Random Forests (2001).
24. Zhong, R., Johnson, R. & Chen, Z. Generating pseudo density log from drilling and logging-while-drilling data using extreme
gradient boosting (XGBoost). Int. J. Coal Geol. https://​doi.​org/​10.​1016/j.​coal.​2020.​103416 (2020).
25. Bikmukhametov, T. & Jäschke, J. Oil production monitoring using gradient boosting machine learning algorithm. IFAC-Pap.
OnLine 52(1), 514–519. https://​doi.​org/​10.​1016/j.​ifacol.​2019.​06.​114 (2019).
26. Song, M. & Zhou, X. A casing damage prediction method based on principal component analysis and gradient boosting decision
tree algorithm. SPE Middle East Oil Gas Show Conf. MEOS Proc. 2019, 1–11. https://​doi.​org/​10.​2118/​194956-​ms (2019).
27. Tang, J. et al. A new tool for searching sweet spots by using gradient boosting decision trees and generative adversarial networks.
in International Petroleum Technology Conference 2020, IPTC 2020, 1–19 (2020). https://​doi.​org/​10.​2523/​iptc-​19941-​abstr​act.
28. Chen, T. & Guestrin, C. XGBoost: A scalable tree boosting system. in Proceedings of the ACM SIGKDD International Conference
on Knowledge Discovery and Data Mining, Association for Computing Machinery, 785–794 (2016). https://​doi.​org/​10.​1145/​29396​
72.​29397​85.
29. Otchere, D. A., Ganat, T. O. A., Gholami, R. & Lawal, M. A novel custom ensemble learning model for an improved reservoir
permeability and water saturation prediction. J. Nat. Gas Sci. Eng. https://​doi.​org/​10.​1016/j.​jngse.​2021.​103962 (2021).
30. Al-Anazi, A. F. & Gates, I. D. Support vector regression to predict porosity and permeability: Effect of sample size. Comput. Geosci.
39, 64–76. https://​doi.​org/​10.​1016/j.​cageo.​2011.​06.​011 (2012).
31. Cranganu, C. & Breaban, M. Using support vector regression to estimate sonic log distributions: A case study from the Anadarko
Basin, Oklahoma. J. Pet. Sci. Eng. 103, 1–13. https://​doi.​org/​10.​1016/j.​petrol.​2013.​02.​011 (2013).
32. Naimi, S. R., Shadizadeh, S. R., Riahi, M. A. & Mirzakhanian, M. Estimation of reservoir porosity and water saturation based on
seismic attributes using support vector regression approach. J. Appl. Geophys. 107, 93–101. https://​doi.​org/​10.​1016/j.​jappg​eo.​2014.​
05.​011 (2014).
33. Vapnik, V. N. The nature of statistical learning. in Theory (1995). https://​ci.​nii.​ac.​jp/​naid/​10020​951890. Accessed 24 Aug 2021.
34. Yu, G. et al. A new method of predicting the saturation pressure of oil reservoir and its application. Int. J. Hydrogen Energy 45(55),
30244–30253. https://​doi.​org/​10.​1016/j.​ijhyd​ene.​2020.​08.​042 (2020).
35. Onwuchekwa, C. Application of machine learning ideas to reservoir fluid properties estimation. in Society of Petroleum Engineers:
SPE Nigeria Annual International Conference and Exhibition 2018, NAIC 2018 (2018). https://​doi.​org/​10.​2118/​193461-​ms.
36. Bagheripour, P., Gholami, A., Asoodeh, M. & Vaezzadeh-Asadi, M. Support vector regression based determination of shear wave
velocity. J. Pet. Sci. Eng. 125, 95–99. https://​doi.​org/​10.​1016/j.​petrol.​2014.​11.​025 (2015).
37. Al-Anazi, A. F. & Gates, I. D. Support vector regression for porosity prediction in a heterogeneous reservoir: A comparative study.
Comput. Geosci. 36(12), 1494–1503. https://​doi.​org/​10.​1016/j.​cageo.​2010.​03.​022 (2010).
38. Khamehchi, E. & Bemani, A. Prediction of pressure in different two-phase flow conditions: Machine learning applications. Meas‑
urement https://​doi.​org/​10.​1016/j.​measu​rement.​2020.​108665 (2021).
39. Negash, B. M. & Yaw, A. D. Artificial neural network based production forecasting for a hydrocarbon reservoir under water injec-
tion. Pet. Explor. Dev. 47(2), 383–392. https://​doi.​org/​10.​1016/​S1876-​3804(20)​60055-6 (2020).
40. Pérez-Zárate, D., Santoyo, E., Acevedo-Anicasio, A., Díaz-González, L. & García-López, C. Evaluation of artificial neural networks
for the prediction of deep reservoir temperatures using the gas-phase composition of geothermal fluids. Comput. Geosci. 129,
49–68. https://​doi.​org/​10.​1016/j.​cageo.​2019.​05.​004 (2019).
41. Urang, J. G., Ebong, E. D., Akpan, A. E. & Akaerue, E. I. A new approach for porosity and permeability prediction from well logs
using artificial neural network and curve fitting techniques: A case study of Niger Delta, Nigeria. J. Appl. Geophys. https://​doi.​org/​
10.​1016/j.​jappg​eo.​2020.​104207 (2020).
42. Ansari, A., Heras, M., Nones, J., Mohammadpoor, M. & Torabi, F. Predicting the performance of steam assisted gravity drainage
(SAGD) method utilizing artificial neural network (ANN). Petroleum 6(4), 368–374. https://​doi.​org/​10.​1016/j.​petlm.​2019.​04.​001
(2020).
43. Negnevitsky, M. N. Artificial Intelligence A Guide to Intelligent Systems Artificial Intelligence Second Edition. (2005). www.​pears​
oned.​co.​uk.
44. Akpan, A. E., Narayanan, M. & Harinarayana, T. Estimation of subsurface temperatures in the Tattapani geothermal field, central
India, from limited volume of Magnetotelluric data and borehole thermograms using a constructive back-propagation neural
network. Earth Interact. 18(6), 1–26. https://​doi.​org/​10.​1175/​2013E​I0005​39.1 (2014).
45. DeFilipps, P. J. Clustering Concepts in Automatic Pattern Recognition. (1975). http://​libra​r y.​ucf.​edu.
46. Jain, A. K., Murty, M. N. & Flynn, P. J. Data clustering: A review. Comput. Surv. 31(3), 278–279 (1999).

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 28

Vol:.(1234567890)
www.nature.com/scientificreports/

47. Otchere, D. A., Ganat, T. O. A., Nta, V., Brantson, E. T. & Sharma, T. Data analytics and Bayesian optimised extreme gradient
boosting approach to estimate cut-offs from wireline logs for net reservoir and pay classification. Appl. Soft Comput. https://​doi.​
org/​10.​1016/j.​asoc.​2022.​108680 (2022).
48. Yan, Y. et al. Harnessing the power of machine learning for carbon capture, utilization, and storage (CCUS): A state-of-the-art
review. Energy Environ. Sci. 14(12), 6122–6157. https://​doi.​org/​10.​1039/​D1EE0​2395K (2021).
49. Song, Y., Sung, W., Jang, Y. & Jung, W. Application of an artificial neural network in predicting the effectiveness of trapping mecha-
nisms on C ­ O2 sequestration in saline aquifers. Int. J. Greenhouse Gas Control https://​doi.​org/​10.​1016/j.​ijggc.​2020.​103042 (2020).
50. Kor, K. & Altun, G. Is support vector regression method suitable for predicting rate of penetration?. J. Pet. Sci. Eng. 194, 107542.
https://​doi.​org/​10.​1016/j.​petrol.​2020.​107542 (2020).
51. Azar, A. T., El-Said, S. A. & Hassanien, A. E. Fuzzy and hard clustering analysis for thyroid disease. Comput. Methods Programs
Biomed. 111(1), 1–16. https://​doi.​org/​10.​1016/j.​cmpb.​2013.​01.​002 (2013).
52. Nainggolan, R., Perangin-angin, R., Simarmata, E. & Tarigan, A. F. Improved the performance of the K-means cluster using the
sum of squared error (SSE) optimized by using the Elbow method. J. Phys. Conf. Ser. 1361(1), 012015. https://​doi.​org/​10.​1088/​
1742-​6596/​1361/1/​012015 (2019).
53. AlFattah, N. A., Ghaleb, M. A. & AlMahdy, O. M. M. Flow units delineation of multiple hydrocarbon reservoirs using hydraulic
zonation technique via cluster analysis algorithm Zeit Bay Field Gulf of Suez, Egypt. Arab. J. Geosci. 9(7), 504. https://​doi.​org/​10.​
1007/​s12517-​016-​2462-2 (2016).
54. Abdulelah, H., Mahmood, S. & Hamada, G. Hydraulic flow units for reservoir characterization: A successful application on arab-D
carbonate. in IOP Conference Series: Materials Science and Engineering, (Institute of Physics Publishing, 2018). https://​doi.​org/​10.​
1088/​1757-​899X/​380/1/​012020.
55. Kirori, Z. Hyper-parameter parameter optimization: Towards practical sentiment analysis using a Convolutional Neural Network
(CNN). Res. J. Comput. Inf. Technol. Sci. 7(2), 1–5 (2019).
56. Cervantes, J., Garcia-Lamont, F., Rodríguez-Mazahua, L. & Lopez, A. A comprehensive survey on support vector machine classifi-
cation: Applications, challenges and trends. Neurocomputing 408, 189–215. https://​doi.​org/​10.​1016/j.​neucom.​2019.​10.​118 (2020).
57. Saner, S., Kissami, M. & Al Nufaili, S. Estimation of permeability from well logs using resistivity and saturation data. SPE Format.
Eval. 12(01), 27–31. https://​doi.​org/​10.​2118/​26277-​PA (1997).
58. Bourke, L. T. Core permeability imaging: Its relevance to conventional core characterization and potential application to wireline
measurement. Mar. Pet. Geol. 10(4), 318–324. https://​doi.​org/​10.​1016/​0264-​8172(93)​90076-5 (1993).
59. Yao, C. Y. & Holditch, S. A. Estimating permeability profiles using core and log data. in SPE Eastern Regional Meeting, Society of
Petroleum Engineers (1993). https://​doi.​org/​10.​2118/​26921-​MS.
60. Fertl, W. H. & Rieke, H. H. Gamma ray spectral evaluation techniques identify fractured shale reservoirs and source-rock char-
acteristics. Journal of Pet. Technol. 32(11), 2053–2062. https://​doi.​org/​10.​2118/​8454-​PA (1980).
61. Bhattacharya, S. Summarized Applications of Machine Learning in Subsurface Geosciences, 123–165 (2021). https://d ​ oi.o
​ rg/1​ 0.1​ 007/​
978-3-​030-​71768-1_5.
62. Raykov, Y. P., Boukouvalas, A., Baig, F. & Little, M. A. What to do when k-means clustering fails: A simple yet principled alternative
algorithm. PLoS ONE 11(9), e0162259. https://​doi.​org/​10.​1371/​journ​al.​pone.​01622​59 (2016).
63. Ahmed, T. Fundamentals of rock properties. in Working Guide to Reservoir Rock Properties and Fluid Flow, 31–115 (Elsevier, 2010).
https://​doi.​org/​10.​1016/​B978-1-​85617-​825-9.​00002-8.
64. Witek-Krowiak, A., Chojnacka, K., Podstawczyk, D., Dawiec, A. & Bubala, K. Application of response surface methodology and
artificial neural network methods in modelling and optimization of biosorption process. Bioresour. Technol. 160, 150–160. https://​
doi.​org/​10.​1016/j.​biort​ech.​2014.​01.​021 (2014).

Author contributions
Conceptualization, T.A., M.H., and A.I.; methodology,T.A., M.H., and A.I.; formal analysis, T.A and M.H.;
investigation, T.A.. and M.H.; resources, M.H and T.A.; data curation, M.H and T.A.; writing—original draft
preparation, M.H., and T.A.; writing—review and editing, A.I. and A.M.; visualization, A.I.; supervision, A.I.
and Y.W..; All authors have read and agreed to the published version of the manuscript.

Competing interests
The authors declare no competing interests.

Additional information
Correspondence and requests for materials should be addressed to A.F.I.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the
Creative Commons licence, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/.

© The Author(s) 2024

Scientific Reports | (2024) 14:4223 | https://doi.org/10.1038/s41598-024-54893-1 29

Vol.:(0123456789)