machine learning &

knowledge extraction

Communication
Using Machine Learning for Enhancing the
Understanding of Bullwhip Effect in the Oil and
Gas Industry
Ana L. Sousa 1, * , Tiago P. Ribeiro 2 , Susana Relvas 3 and Ana Barbosa-Póvoa 3
1 CERENA, Instituto Superior Técnico, Universidade de Lisboa, 1049-001 Lisboa, Portugal
2 Tal Projecto Lda., 1350-252 Lisboa, Portugal; [email protected]
3 CEG-IST, Instituto Superior Técnico, Universidade de Lisboa, 1049-001 Lisboa, Portugal;
[email protected] (S.R.); [email protected] (A.B.-P.)
* Correspondence: [email protected]; Tel.: +351-218-417-000




Received: 30 June 2019; Accepted: 4 September 2019; Published: 6 September 2019


Abstract: Several suppliers of oil and gas (O & G) equipment and services have reported the necessity
of making frequent resources planning adjustments due to the variability of demand, which originates
in unbalanced production levels. The occurrence of these specific problems for the suppliers and
operators is often related to the bullwhip effect. For studying such a problem, a research proposal
is herein presented. Studying the bullwhip effect in the O & G industry requires collecting data
from different levels of the supply chain, namely: services, upstream and midstream suppliers, and
downstream clients. The first phase of the proposed research consists of gathering the available
production and financial data. A second phase will be the statistical treatment of the data in order
to evaluate the importance of the bullwhip effect in the oil and gas industry. The third phase of the
program involves applying artificial neural networks (ANN) to forecast the demand. At this stage,
ANN based on different training methods will be used. Further on, the attained mathematical model
will be used to simulate the effects of demand fluctuations and assess the bullwhip effect in an oil and
gas supply chain.

Keywords: artificial neural networks; bullwhip effect; oil and gas industry; research proposal; supply
networks; machine learning

1. Introduction
The behavior problems experienced by the oil and gas (O & G) suppliers find one plausible
explanation on the bullwhip effect, which is a well-known distribution channel problem [1]. Considering
this, the scope of the research proposal is to develop an investigation of the bullwhip effect in the O &
G supply network, giving an insight overview of this phenomenon in this system.
Suppliers must adjust to the workload associated with the large demand peaks [2], as well as the
variation of cyclical periods of low and high demand [3]. Such variations lead both to challenging
industrial management and sustainability problems and to increased technical and maintenance
issues [4], fostered by over or underproduction. In fact, current industrial panorama is shaped by
ecological and social sustainability concerns [5,6], which have been ubiquitous in the fourth industrial
revolution, or Industry 4.0 [7,8], in which concepts such as the bullwhip effect have been articulated.
Production and financial data from different levels of the oil and gas supply chain shall be uniform,
embracing a wide market share throughout the supply chain and easy access.
It is significant to observe that ongoing and accelerating digitalization and the digital
transformation of energy and specifically O & G industries allow gathering and making available

Mach. Learn. Knowl. Extr. 2019, 1, 994–1012; doi:10.3390/make1030057 www.mdpi.com/journal/make

Mach. Learn. Knowl. Extr. 2019, 1 995

a profuse amount of multivalent data, including financial, production, and sustainability insights.
Thence, there is an immense potential for employing data analytics techniques and machine learning
in particular for enhancing the knowledge about well-known phenomena; however, some elements
require further studies, such as the bullwhip effect.
Within the oil and gas industry, it is a well-known fact that the demand variability increases from
downstream to upstream as in a usual supply chain [9,10]. This behavior is similar to the phenomenon
known as the bullwhip effect, which has been vastly studied in the retail industry [11–13]. However, in
the O & G industry, the analysis of the bullwhip effect considering the multiple-level supply network
has not been deeply studied.
Oil and gas supply networks can become complex, as they encompass multiple players and
intricate systems. As a result, a large amount of data shall be examined to evaluate the phenomenon.
Also, taking into consideration that too-simple models have been a major hindrance for past bullwhip
effect studies [14], this is clearly a case for using the current artificial intelligence capabilities. Cutting
edge techniques, such as the recent developments in artificial neural networks will provide the means
to infer the meaningful relations within an oil and gas supply network. Thus, the availability of
network simulation and forecasting tools will be pivotal for studying the bullwhip effect.
Beyond learning from data, the use of artificial neural networks (ANN) will allow attaining
mathematical solutions with a generalization capacity. This is yet another very significant advantage of
implementing ANN-based analysis methodologies, since the results of the proposed research program
will not cease in the bullwhip effect for one specific case, but rather are expected to provide tools for
the community that can be efficiently applied to similar problems. Developing industry-specific or
integrating broader [15] risk analysis tools is a possible outcome for the research output.

2. Literature Review

2.1. Bullwhip Effect

2.1.1. Overview
The industrial era brought the need for operations management and supply chain management
development. Over time, technology, the economy, and the social landscape have been responsible for
changing it.
By the industrial revolution, products were quite simple, markets were locally oriented,
communication was slower, as the transportation speed also was, and innovation cycles were
longer [16], allowing the supply chains to be simpler and more vertically integrated. After World War
II, the production paradigm changed. Seizing technological developments that enabled the production
of more innovative, complex, and customized products, companies adapted to provide the best service
to their clients. Productivity and efficiency were the main concepts that supported the production
paradigm. While compressing the speed of operations is the main method for increasing productivity,
efficiency is reached by reducing the use of resources and increasing asset utilization. To achieve a
large scale and reduce costs, it was necessary to standardize products and production systems.
During the 1950s and 1960s, Forrester focused his research on the operational behavior of
companies and published several studies about Industrial Dynamics [1,17], becoming known as the
first author who described the bullwhip effect. The effect of lack of intercoordination among supply
chains and the unawareness of managers to deal with the complex interactions results in an increased
variability of placed orders at suppliers, comparing to the received orders given to customers. To face
the challenges of manufacturing both more complex products and supplying them to global markets,
most companies focused on optimizing their internal processes. However, the bullwhip effect was
not perceived as a major issue. In the 1950s, supply chain structures were still quite simple and
vertically integrated [16]. In this period, the simulation of Industrial Dynamics was restricted due to
the technological limitations of analogue computers [18].
Mach. Learn. Knowl. Extr. 2019, 1 996

Due to the technological development and increasingly complex processes, it became impossible
to keep the full process inside a single company. More specialized and smaller companies have
grown skills to manufacture semi-finished products with greater cost efficiency. In globalized
markets, to cope with longer geographical distances, big companies generally choose to decentralize,
fragmentating or outsourcing some operations [19]. This decomposition changed the supply chains
into supply networks [14]. As the focus on intra-organizational operations management has shifted to
inter-organizational, the bullwhip effect has become one major target of analysis [16]. For instance, in
1989, Sterman proposed the Beer Game experiments, which demonstrated the severe impact of the
bullwhip effect [20].
In the same decade, the appearance of information technology systems transformed companies’
management operations, since it allowed much cheaper and faster communications, and more accurate
planning [21]. This technological development also enabled additional globalization, since it allowed an
improved interconnectivity between geographically isolated operations [22]. Offering more variability
and enhanced product customization were ongoing milestones in company strategies. During the
1990s, the bullwhip effect was comprehensively studied using empirical data. For instance, Pampers,
HP, and Barilla [10,23] were some of the studied companies.
During the 21st century, research on supply chain management has also focused on the
supply chain’s networked structure. Such a structure of modern supply chains forms a barrier
to collaboration [24], due to the players’ behavior [25,26]. Lee et al. concluded that it is challenging
to reach agreements between the parties in the supply chain to implement the bullwhip mitigation
solutions in order to improve the overall performance [10,21,27]. Although companies should focus
their strategies on cooperation and supply chain synchronization [28,29], since information sharing
is regarded as the main principle to prevent this effect [12,30–35], the managers tend to be hesitant
regarding collaborative programs [36]. Thus, the quantity and quality of shared information remains
low. Nevertheless, Tsay and Lovejoy concluded that some mitigation solutions developed under
such principle have resulted in counterproductive results [19]. The main causes for such a lack of
cooperation may be due to a lack of knowledge, lack of competence, or lack of trust.
Recently, some studies have been developed to understand the bullwhip phenomenon in the
oil and gas industry. However, while most studies consider a simple supply chain, the oil and gas
industry systems are a multilevel supply network. One example can be found in Shizheng et al., who
developed their study to quantify the bullwhip effect in a supply chain based on a single petroleum
distribution system [37]. Within the same study field, Huang et al. implemented a method to control
the bullwhip effect in two Chinese petrochemical companies: Liaoyang Petrochemical and Shanghai
Baosteel Yichang Steelstrip [38]. Jacoby simulated the cost of the bullwhip effect in the industry [39].
Zhang and Zhang analyzed the bullwhip effect in China’s processed oil supply chain, using a dynamic
analysis system and programming a model to mitigate this effect, and the authors concluded that the
delay time is the main reason for the existence of the bullwhip effect [40]. Lastly, Sherhart assessed a
specific bullwhip effect problem at British Petroleum [41].
However, few articles offer strategies that specifically mitigate the bullwhip effect in the oil and
gas industry. The aforementioned studies (Huang et al. [38], Jacoby [39], Zhang and Zhang [40],
Sherhart [41]) may be only regarded as useful developments for modeling the bullwhip effect in specific
oil and gas companies.

2.1.2. Performance Measures, Models, and Methods

The bullwhip effect refers to a demand distortion in a supply chain, which propagates and
amplifies as one moves up in the supply chain, from the buyer to the supplier ([10,28,29,35,42–45]).
Sucky (2009) divided the research on bullwhip effect into six categories [14], which are described
in the Table 1. Based on the research developed in this paper, the collected references were categorized
according to Sucky’s classification.
Mach. Learn. Knowl. Extr. 2019, 1 997

Table 1. Categorization of the bullwhip effect studies.

Categories References
Quantification of the bullwhip effect [10,46–52]
Identifying the causes of the bullwhip effect [44,53]
Observation studies in some industries [54–56]
Methods to reduce the bullwhip effect [57–61]
Simulation of the system behavior [62–65]
Experimental validation of the bullwhip effect [30,41,66–68]

To perform a statistical survey about the metrics applied to study the bullwhip effect, a wide
set of publications was gathered, and the information was treated in order to provide an overview
about the relevant supply chain contributions. Criteria for selecting the reference works included its
Mach. Learn. Knowl.
publication Extr.the
during 2019, 1 20 years on Web of Science (WoS) indexed journals with “bullwhip effect”
past 997
and “supply chain” keywords. After an initial group of articles was gathered, only the ones with explicit
and extensive performance Table 1. Categorization
metrics descriptionsofwere
the bullwhip
includedeffect studies.
in the collection herein presented in
this section. Considering the information provided
Categories in the set synthetized in Table 2, it was possible to
References
summarize the main performance metrics used to study the bullwhip effect (Figure 1).
Quantification of the bullwhip effect [10,46–52]
Identifying the2.causes
Table of the bullwhip
Performance metrics pereffect
reference. [44,53]
Observation studies in some industries [54–56]
Performance Metrics References
Methods to reduce the bullwhip effect [57–61]
Order rate variance ratio [43,48,61,64,69–79]
Simulation
Amplitudeofratethecost
system
ratio behavior [62–65]
[80]
Amplification ratio [81]
Experimental validation of the bullwhip effect [30,41,66–68]
Ratio inventory [71,75]
Ratio backlog inventory [64]
To perform a statistical survey about
Variance ratiothe
fill metrics
rate applied to study the bullwhip effect, a wide set
[75,77]
Ratio inventory
of publications was gathered, and theintegrated
information squared
waserror
treated in order[69]to provide an overview about
the relevant supply chainRatio root mean square
contributions. Criteriacostsfor selecting the [61]reference works included its
Fill rate [82]
publication during the past 20 years on Web of Science (WoS) indexed journals with “bullwhip effect”
Costs order rate variance ratio [43,73,83]
and “supply chain” keywords. After Costs an initial group of articles was [84]gathered, only the ones with
explicit and extensive performance metricsstock
Ratio inventory descriptions were included
[70,74] in the collection herein
Out size stock
presented in this section. Considering theout number
information provided in [70]
the set synthetized in Table 2, it
was possible to summarize the main performance metrics used to study the bullwhip effect (Figure 1).

Figure 1.
Figure Performance metrics
1. Performance metrics to
to study
study the
the bullwhip
bullwhip effect.
effect.

Table 2. Performance metrics per reference.

Performance Metrics References

Order rate variance ratio [43,48,61,64,69–79]
Amplitude rate cost ratio [80]
Amplification ratio [81]
Ratio inventory [71,75]
Ratio backlog inventory [64]
Mach. Learn. Knowl. Extr. 2019, 1 998

As Figure 1 illustrates, the most common performance metric suggested in the literature is the
order rate variance ratio [85], which was proposed by Chen et al. [48]. This indicator is given by the
following Equation (1) [29]:
σ2o /µo
Order rate variance ratio = 2 (1)
σd /µd

where σ2o is the orders’ variance, σ2d the variance of the market demand, and µo and µd are the respective
mean values.

2.2. Artificial Neural Networks

2.2.1. Background
Artificial intelligence (AI), also referred as machine intelligence, is a wide scope, cross-disciplinary
term deemed to encompass all machine behavior that is inspired by human cognitive functions.
AI is based on algorithms, but cannot be accurately defined as mathematical reasoning, since its
most notorious ability is the capability for adaptation with insufficient knowledge and resources [86].
Such ability is not provided by Mathematics, and can only be found in the current definition
of “intelligence”.
The roots of AI go back to the World War II [87,88], from which it has grown consolidated upon
several disciplines. While many classifications and clustering have been proposed for such techniques,
the most common taxonomy distinguishes between, at least, seven different disciplines [89–93]. Those,
which can also be referred as sub-disciplines, sub-fields, or branches, are machine learning (ML), expert
systems, fuzzy logic, robotics, computer vision, natural language processing, and speech recognition.
Furthermore, several techniques have achieved notability through wide employment. That is the case
of deep learning techniques, artificial neural networks, decision trees and random forests, among many
others, applied within the machine learning framework, but also evolutionary and genetic algorithms,
support vector machines, or case-based reasoning, to name a few.
Among the aforementioned disciplines, machine learning is the one with the widest employment
for engineering problems, and especially for the oil and gas industry [89,94]. ML aims to conceive
algorithms that improve automatically through experience [93]. Yet, for doing so, machine learning
techniques typically sort from a comprehensive set of possible algorithms, initially finding the best
suited for the data, given certain performance criteria. ML can be classified as supervised learning,
unsupervised learning, and reinforcement learning.
Machine learning is devoted to foster humanly behavior in machines. Specifically, it allows
“teaching” or “training” computers how to compute tasks by giving examples on how it should be
done [95]. Thus, it can be employed when there is profuse data (used as patterns or examples) on
a certain phenomenon, but not necessarily an analytical relation or perfectly described explanation
among variables.
Artificial neural networks (also referred as ANN) are machine learning techniques. In fact, those
are the most powerful and oldest techniques [96]. Therefore, it is not a surprise that ANN also leads
the number of known practical applications, and are applied to most fields of knowledge ([97,98]).
An ANN can be regarded as a mathematical model designed to perform a specific task, simulating
the manner by which the brain computes information by employing processing units (the neurons).
Despite being mathematically simple, ANN-based solutions combine different network topology and
network algorithms together in order to provide more accurate results than those given by classical
approaches, as the nonlinear multivariate regression. Furthermore, ANN-based solutions can be
employed even without an extensive knowledge of the modeled function or numerical relations [99].
Nonetheless, prior data analysis is critical for planning optimal ANN-based solutions. As a matter
of fact, the dataset shape has an indelible impact upon the solutions and approaches [100], making
topological studies of paramount importance, especially when datasets are gathered form profuse
information, as suggested in this research proposal.
Mach. Learn. Knowl. Extr. 2019, 1 999
Mach. Learn. Knowl. Extr. 2019, 1 999

The
The general
general ANN
ANN structure
structure integrates
integrates several
several nodes
nodes disposed
disposed inin connected
connected vertical
vertical layers
layers (an
(an
input
input layer, one or more hidden layers, and an output layer). Associated to each node, or neuron, in
layer, one or more hidden layers, and an output layer). Associated to each node, or neuron, in
the hidden and output layers is a transfer (or activation) function, which can be
the hidden and output layers is a transfer (or activation) function, which can be linear or nonlinear.linear or nonlinear.
Such
Such function
function receives
receivesaanet netinput
inputand andtransmits
transmitsan anoutput.
output.
Most
Most ANN are multilayered and feedforward. The
ANN are multilayered and feedforward. The latter
latter means
means that
that data
data inserted
inserted in in the
the input
input
layer
layer flows in the vertical direction only, so that each node is only connected to nodes belonging to
flows in the vertical direction only, so that each node is only connected to nodes belonging to
layers
layers disposed
disposed atat the
the following
following level.
level. Among
Among multilayer
multilayer feedforward
feedforward networks,
networks, one one common
common and and
useful
useful learning
learning algorithm
algorithm is is the
the back
back propagation
propagation (BP),(BP), proposed
proposed by by Rumelhart
Rumelhart et et al.,
al., in
in 1986
1986 [101].
[101].
BP
BP allows constantly adjusting the network weight and threshold values back and forth in order to
allows constantly adjusting the network weight and threshold values back and forth in order to
attain
attain the
the minimum
minimum sum sum of of aa squared
squared error.
error. Further
Further developments
developments brought
brought the the back
back propagation
propagation
adaptive
adaptive algorithm
algorithm (BPA)
(BPA) or or enhanced
enhanced convergence
convergence and and precision
precision even
even for
for few
few data,
data, asas attained
attained by by
implementing the Marquardt algorithm [102]. However, there are many
implementing the Marquardt algorithm [102]. However, there are many other learning algorithmsother learning algorithms
suited
suited for
for multilayer
multilayer feedforward
feedforward networks.
networks. The The recent
recent and
and non-iterative
non-iterative extreme
extreme learning
learning machine
machine
(ELM) algorithm is an example, including its developments in convex incremental
(ELM) algorithm is an example, including its developments in convex incremental ELM (CI-ELM), ELM (CI-ELM),
incremental
incremental ELM
ELM (I-ELM),
(I-ELM), and and mini-batch
mini-batch ELM ELM (MB-ELM)
(MB-ELM)[32,103–105].
[32,103–105]. Figure
Figure 22 shows
shows aa schematic
schematic
representation of a multilayered
representation of a multilayered ANN. ANN.

Figure 2. Example of a multilayer artificial neural networks (ANN).

Figure 2. Example of a multilayer artificial neural networks (ANN).
Using ANN enables a very efficient use of computing power, providing a rather scarce, and
Using ANN
surprisingly enables[106],
foreseeable a very efficient use
computing timeoftocomputing
yield results power,
fromproviding a rather This
complex models. scarce, and
is due
surprisingly foreseeable [106], computing time to yield results from complex models.
to its massively parallel distributed structure, making them suitable to efficiently solve simple and This is due to
its massively parallel distributed structure, making them suitable to efficiently
complex problems. In fact, the ANN structure owes its power and versatility to its core characteristics solve simple and
complex
of problems.
self-learning In fact, the ANNThe
and generalization. structure owesthat
first means its power and versatility
the algorithm to its core
development basedcharacteristics
on external
of self-learning and generalization. The first means that the algorithm
conditions enables ANN to adapt to several, and even evolving, data contexts, without thedevelopment based on external
need to
conditions enables ANN to adapt to several, and even evolving, data contexts,
be “reshaped” or “reprogrammed”. The latter is deemed to express that once the learning process without the need to
be “reshaped”
has been performed,or “reprogrammed”. The latter
the resulting algorithm canisbedeemed to express
less sensitive thatquality,
to input once the learning
being able toprocess
cover
has been performed, the resulting algorithm can be less sensitive to input quality,
a profuse amount of possible inputs. This also provides a greater tolerance to data lacking quality being able to cover
a profuse
or quantity. amount of possible inputs. This also provides a greater tolerance to data lacking quality or
quantity.
On the other hand, one cannot disregard some shortcomings in ANN. Among the most relevant
are theOninability
the otherforhand, one cannot
necessarily disregard
reaching some
globally shortcomings
optimized in ANN.
algorithms, theAmong the most relevant
slow convergence speed,
are the inability for necessarily reaching globally optimized algorithms,
and the lack of theoretical guidance for algorithm building (the number of hidden layers andthe slow convergence speed,
the
and the lack of theoretical guidance for algorithm building (the number of hidden
unit layer choice, for example), which leads to the need for profuse experiments [107]. Furthermore, layers and the unit
layer choice,
practical for example),
hindrances whichdealing
emerge when leads to thevery
with need for profuse
profuse raw data.experiments
Finding and [107]. Furthermore,
filtering the most
practical hindrances emerge when dealing with very profuse
significant subdata is a critical task that precedes ANN application [108]. raw data. Finding and filtering the most
significant subdata is a critical task that precedes ANN application [108].
Mach. Learn. Knowl. Extr. 2019, 1 1000

2.2.2. Developing ANN Solutions

Many practical issues arise when conceiving and developing ANN solutions. A brief summary
of important steps and solutions that an ANN developer must take into account is given in the
following lines.
Strategies must be implemented in order to assure that the input variables are independent and
relevant, considering its effect upon the ANN accuracy [109,110]. Thus, input dimensionality reduction
is a required step for which two approaches can be used, namely the feature selection and the feature
extraction. Several algorithms allow fulfilling the input reduction. Among the most common are the
linear correlation, the autoencoder or the orthogonal, and sparse random projections [110].
Once the input datasets handling has been defined, training, validation, and testing procedures
must also be set. This means deciding the amount of data to be shared between those tasks. Usual
relations for the former procedures are 50–25–25, 60–20–20, 70–15–15, or 80–10–10 [94].
Training efficiency can be impaired if the concentration of training data points is heterogeneous.
Thus, several techniques can be employed for input normalization, such as the ones proposed by
Lachtermacher and Fuller [111]; Pu and Mesbahi [112]; and Tohidi and Sharifi [113].
Some transfer functions can be employed for output and hidden layers. Among the most
common, one can count the logistic (a sigmoid type), the hyperbolic tangent (also sigmoid), bilinear,
identity, identity–logistic, bipolar, positive saturating linear, sinusoid and radial basis functions.
Afterwards, output normalization may be considered in order to avoid numerical problems in the
training phase [114,115].
Network arrangement must also be defined. Even if limiting the option to one type of ANN, such
as the multilayer feedforward, important decisions about the behavior of nodes in the hidden layers
will define the arrangement type. Common choices span from the multilayer perceptron network to
radial-basis function network [94]. While no arrangement can be concluded better than any other,
there is extensive literature about the adequacy of each of these networks for certain problems. Besides
the previous choices, the connectivity criteria must be defined. Even if considering a feedforward
network, only adjacent or distant layers may be set for connecting to each other.

2.2.3. Application to Supply Chain Management

A supply chain can be regarded as a complex network spanning several companies and sectors.
As materials and products are acquired, processed, and sold among the network players, a flow of
information is generated. Additionally, the companies involved keep extensive records of almost every
aspect of its activities. Cumulatively, this is an ever-growing amount of information which, when
wisely used, can provide analysts and managers with excellent tools for supply chain management.
However, as data becomes profuse, the difficulty of handling it increases. Furthermore,
mathematical descriptions for the relations that link the network have not been widely established
yet [116]. In this context, information can be used for extracting performance indicators.
Artificial neural networks are a possible answer for this problem [117], such that its use in supply
chain management has boomed in the past years [107]. The ability of ANN to learn from abundant
data allows modeling the behavior of supply chain networks [118–120]. Such mathematical output
is extremely important, since it offers the possibility of forecasting the network behavior both for
previous and new conditions as well as testing the effects of network manipulation and modifications.
Currently, ANN are employed in supply chain management mostly in three areas [107]. Those are
optimization [121,122], forecasting [123], and decision support [124–128]. In the first case, ANN are
used for performing optimization tasks within models deemed to solve logistic problems such as
optimizing transport routes, schedules, equipment, and shop scheduling, or for enhancing warehouse
management, especially in a context of real-time information updates. Forecasting plays a major role
in uncertainty and risk mitigation. Unfortunately, management decisions are often made in a context
where data is insufficient or unreliable; thus, other forecasting methods face significant inadequacy
problems. That is the case for expert systems, statistical methods, and time series [107], but not for
Mach. Learn. Knowl. Extr. 2019, 1 1001

well-trained ANN. Finally, decision support has benefited from ANN capacities for identification,
classification, and self-organization tasks. Furthermore, ANN versatility allows the development of
case-specific algorithms, which can support particular decision problems.

2.2.4. Application to Oil and Gas Industry

Considering the recent increase in using artificial intelligence techniques for studying virtually
all types of engineering problems, with a special emphasis on artificial neural networks, it is with no
surprise that recent developments both in oil and gas research and practice make extensive use of
ANN. In fact, the size of databases generated either in geological analyses or in production control and
optimization require the use of the most advanced and powerful big data analysis tools that science
can provide.
Recently, many scientific journals devoted to O & G research published special issues for AI
applications. Furthermore, many of the recently most cited articles within O & G research employ ANN.
That is the case of Ahmadi [129] (production performance), Mahdiani and Khamehchi [130] (crude
oil price), Mirzaei-Paiaman et al. [131] (production flow rate), Torabi et al. [132] (physical properties),
and Arjun and Aneesh [133] (physical properties), to name a few. However, using ANN to study the
bullwhip effect on the oil and gas industry is still not common. Benefiting from such resource is one
further reason why this research program proposal suggests the development of ANN.

3. Research Objectives

3.1. Main Goals Delineation

The problem the proposed work aims to address can be reduced to the following specific
research questions:

• Does the bullwhip effect occur in the petroleum industry?

• Does the suppliers’ level exhibit the highest demand variability in the supply network?
• Are smaller companies more susceptible to higher variability than larger companies?
• Is it possible to forecast the bullwhip effect using artificial neural network techniques?
• Is it possible to create mathematical models for O & G supply networks’ behavior, so that possible
remedial measures for the bullwhip effect can be assessed prior to testing in real conditions?

Answering these questions will allow the researcher to comprehensively characterize the effects
of demand variability onto a typical oil and gas supply network, draw conclusions about the existence
of the bullwhip effect upon the interest group, distinguish such effects based on the company profile of
that group, and ultimately, be able to model, forecast, and manipulate the network behavior, with a
special focus on the equipment and service suppliers.
Considering the former, the main goal of the research proposed can be summarized as to
characterize the bullwhip effect on the oil and gas industry, and to develop a tool to model it.

3.2. Scientific and Social Relevance

As highlighted in the historical brief regarding the state of the art of the bullwhip effect, its study
has gained an increasing relevance as supply chains have evolved to supply networks, as this has led to
the processes, stakeholders, and their connections becoming much more complex and the information
flow achieving unprecedented dimensions.
Therefore, understanding the bullwhip effect in depth and being able to promote validated
remedial measures is a stepping stone toward securing industry efficiency, keeping the consumer
prices as low and stable as possible, and avoiding product unavailability or excessive stocks, which
frequently come with high economic and environmental costs.
Mach. Learn. Knowl. Extr. 2019, 1 1002
Mach. Learn. Knowl. Extr. 2019, 1 1002

Considering the
Considering the reach
reach and
and impact
impact ofof oil
oil and
and hydrocarbons’
hydrocarbons’ sub-products
sub-products in in our
oursociety,
society, stability
stability
and cost effectiveness are paramount for our main goals, spanning from families’ economic
and cost effectiveness are paramount for our main goals, spanning from families’ economic stability stability to
global
to economic
global economic growth,
growth, energetic efficiency,
energetic andand
efficiency, environmental
environmental andand
sustainability goals.
sustainability goals.
From a scientific perspective, the ability to model the bullwhip effect in complexnetworks,
From a scientific perspective, the ability to model the bullwhip effect in complex supply supply
based on collected
networks, based ondata, is a remarkable
collected advance, since
data, is a remarkable it will
advance, additto
since theadd
will currently limited available
to the currently limited
models. Furthermore,
available proposingproposing
models. Furthermore, and validating remedial measures
and validating for this industry
remedial measures for thiscan bring real
industry can
innovation in supply networks management and foster further scientific advances.
bring real innovation in supply networks management and foster further scientific advances.

4. Methodology
4. Methodology and
and Planning
Planning

4.1. Methodology
4.1. Methodology
A data-driven work methodology is proposed for the current research idea. It shall encompass
A data-driven work methodology is proposed for the current research idea. It shall encompass
a problem definition, solution development, and solution testing, as extensively described by Sage
a problem definition, solution development, and solution testing, as extensively described by Sage
and Armstrong [134]. In a broader sense, the suggested methodology can be regarded as a pragmatist
and Armstrong [134]. In a broader sense, the suggested methodology can be regarded as a pragmatist
epistemology, since it will be based on real data collection—a form of empiricism—to perform advanced
epistemology, since it will be based on real data collection—a form of empiricism—to perform
analyses and achieve logical inferences, using the principles or rationalism.
advanced analyses and achieve logical inferences, using the principles or rationalism.
Among those stages, the problem definition has been briefly addressed in this proposal, and shall
Among those stages, the problem definition has been briefly addressed in this proposal, and shall
be enhanced with the research program’s initial literature review. Solution development and testing will
be enhanced with the research program’s initial literature review. Solution development and testing will
be nested under the ANN sub-program. In accordance, the following major specific steps for work
be nested under the ANN sub-program. In accordance, the following major specific steps for work
methodology are defined in Figure 3.
methodology are defined in Figure 3.

Figure 3. Methodology to perform this research program.

Figure 3. Methodology to perform this research program.
Despite the previous suggestions, which were deliberately kept broad in scope and far-reaching
Despite the previous suggestions, which were deliberately kept broad in scope and far-reaching
in order to foster researchers’ freedom, it is important to notice that the work methodology must be
in order to foster researchers’ freedom, it is important to notice that the work methodology must be
continuously reassessed throughout the program’s duration, taking into consideration the attained
continuously reassessed throughout the program’s duration, taking into consideration the attained
partial results as well as the literature review that will be undertaken.
partial results as well as the literature review that will be undertaken.
4.2. Work Phases
4.2. Work Phases
Table 3 provides a time schedule for the completion of the proposed research program. The tasks
haveTable
been 3clustered
provides in a time
five schedule
functional forgroups,
the completion
namelyof the proposed
performing theresearch program.
literature review The
andtasks
data
have been clustered in five functional groups, namely performing the literature
gathering planning; data gathering and analysis; artificial neural network; the writing and submission review and data
gathering
of articles; andplanning; data gathering
other dissemination and analysis; artificial neural network; the writing and
activities.
submission
While the of articles; andisother
first group dissemination
self-explanatory activities.
and accounts for the continuous research and planning
While the first group is self-explanatory
activities that must take place throughout the program, and accounts
thefor the continuous
remaining steps mustresearch andinplanning
proceed order.
activities that must take place throughout the program, the remaining steps
Therefore, data gathering and its critical analysis must commence from the first moment,must proceed in order.
since it
Therefore, data gathering and its critical analysis must commence from
is the basis for the remaining steps. Soon after, the ANN model development must take place, butthe first moment, since
it
it is the basis for the remaining steps. Soon after, the ANN model development must
may extend for four semesters, given its complexity. Also, at the beginning of the second year of work, take place, but
it
themayfirstextend
paperfor four
shall besemesters, givenboth
written, using its complexity. Also,
the literature at the
review andbeginning of the second
the preliminary year of
data analysis.
work, the first paper shall be written, using both the literature review and the preliminary data
analysis. From this stage on, ANN model development, validation, and implementation run
continuously, while the results analysis with papers and thesis writing shall occur whenever relevant
results are achieved.
Mach. Learn. Knowl. Extr. 2019, 1 1003

The last group of tasks related to disseminating the work through conferences may be left for
the scheduled second half of the research program.
Mach. Learn. Knowl. Extr. 2019, 1 1003
The following sections will detail the major aspects that should be developed at each phase.
Mach. Learn. Knowl. Extr. 2019, 1 1003
The last group of tasks related to disseminating the work through conferences may be left for
Table 3. Research program plan: Time schedule.
the scheduled second half of the research program.
From
Thethis stage on,sections
following ANNYear
model development,
will detail the majorvalidation, and
aspects that implementation
should
Year 1 run at
be developed
Year 2 Year continuously,
3 each 4 while
phase.
Year
the results analysis with papers and thesis writing shall occur whenever relevant results are achieved.
Semester 1 2 1 2 1 2 1 2
Table 3. Research program plan: Time schedule.
Literature Review
Table and Data
3. Research Gathering
program Planning
plan: Time schedule.
Year
Literature review and data gathering planning x Year 1 x Year
x Year 2 x Year
xYear 3x Year
xYear 4
xYear
Year 1 2 3 4
Semester
Data Gathering and Analysis1 2 1 2 1 2 1 2
Semester 1 2 1 2 1 2 1 2
Collecting the data Literature Review and Data Gathering
x x Planning
x
Literature Review and Data Gathering Planning
Literature review
Literature
Statistical and data gathering planning
review and data gathering planning
analysis x x x x x x xx x x x x x x x x x x
x
Data Gathering
Data
Artificial and
Gathering andAnalysis
Neural NetworkAnalysis
Collecting the data Mach. Learn. Knowl. Extr. 2019, 1 x
Collecting
Developing the the data
model x x x x x xx x x
Statistical
Statistical analysis
analysis x xx x x x
Validating the model The last group of tasks related to disseminating the work x through conferences may be left
Artificial
Artificial Neural
Neural Network
Implementing the modeltheto scheduled second half
the case study of theNetwork
research program. x x
Developing the model The
Developing the model following sections will detail the major x x xthat
aspects
x x xshould
x x be developed at each phase.
Writing and Submission of Articles
Validating the model x
Validating
Paper Implementing
the model
I—Quantitative analysis
the model on case
to the the bullwhip
study in 3. Research program plan: Time xschedule.
effectTable x x
Implementing the model to the case study x x x
a supply chain network Writing and Submission of Articles
PaperPaper
II—The bullwhip effect: a Writing
case
I—Quantitative analysis on the and
study
bullwhipthe oilYear
in Submission
effect and of Articles
in a supply
Year 1 Year 2 Year 3 Year 4
x x
gasPaper
industry
chain network
I—Quantitative analysis on the bullwhipSemester
effect in 1 2 1 2 1 2 1 2
Paper II—The bullwhip effect: a case study in the oil and gas x
a supply
Paper III—Oil chain
andnetwork
gas supply chain managementLiteraturebased Review and Data Gatheringx Planning
industry x
on Paper
artificial
Paper neural
II—The network
III—Oilbullwhip effect:chain
and gas Literature
supply a case study
review and
managementindata
the oil and
gathering
based on planning x x x x x x x x
x x
gasartificial
Writing industry
up the neural network
thesis x x x
Data Gathering and Analysis x x
Writing up the thesis x x x x x
Paper III—Oil and gas supply Collecting chain
Other the management
dissemination
datadisseminationbased
activities x xx x
on artificial neural network Other activities
Oral presentation at a conference Statistical analysis x x x x x
Oral presentation at a conference
Writing
PosterPoster up
presentationthe thesis
at aat conference x
x Network x x x x
presentation a conference Artificial Neural x
Other
Legend: dissemination
Critical activities
tasks.
Developing the model
Legend: Critical tasks. x x x x
Oral presentation at a conference x
4.2.1. Poster
Literature Review and Validating the model x
The presentation
last group of aData
attasks Gathering
conference Planning
related to disseminating the work through conferences x may be left for the
Implementing the model to the case study x x
scheduled
The first stepsecond
is to half of thea research
perform program.review,
deeper Legend:
literature which will support the identification of
Critical tasks.
systemic and The behavioral
following sections will have
factors that detailan the majoron
impact theWriting
aspects andeffect
that should
bullwhip Submission
bein the oilof
developed Articles
andat gas
eachsupply
phase.
4.2.1. Literature Review Paper
and DataI—Quantitative
Gathering analysis
Planning on the bullwhip
chain. Also, a clear definition of the oil and gas supply chain and all the stakeholders involved will effect in
x
be set4.2.1.
up,
The
Literature
in order
first steptoReview
understand
is
aand
supply
to performhow
Datachain
Gathering
eachnetwork
a deeper entityPlanning
interacts
literature review, with the others
which and thethe
will support impact of each of
identification
entity on the
systemic The bullwhip
andfirst stepeffect.
behavioral Paper
is tofactors
perform II—The
a deeper
that have bullwhip effect:
anliterature
impact on a case
review,
the study will
which
bullwhip in the
effect oil
supportandoil
in the theandidentification
gas supplyof x
In this
systemic stage,
and the definition
behavioral gas of
factorswhich
industrythat parameters
have an should
impact on be
thegathered
bullwhip
chain. Also, a clear definition of the oil and gas supply chain and all the stakeholders involved in order
effect in to characterize
the oil and gasthe
supply
will
oil be
and gas
chain.
set up, supply
Also,
in ordernetwork
a clear should
todefinition
Paper of
understand betheestablished.
III—Oil
how oileach
andgas
and gassupply
supply
entity chainwith
chain
interacts and the
all the
management stakeholders
others based
and the impactinvolved of will
eachbe
set up, x
entity on intheorder to understand
bullwhip how each
on artificial
effect. neural entity interacts with the others and the impact of each entity
network
4.2.2.onData Gathering
thethis
bullwhip and
effect.Analysis
In stage, the definition
Writingofup which parameters should be gathered in order to characterize thex
the thesis x x x x
andIn
oilDuring gasthis
this stage,
phase,
supply the
thedefinition
network shouldof
available which
financial
be parameters
and production
established. should databe will
gathered in orderfrom
be collected to characterize
the New the
Other dissemination activities
York oil andExchange
Stock gas supply network
database should be
[135,136] established.
as well as in companies [137–143] other stock markets [144],
4.2.2. Data
projects, Gathering
and national andOral presentation
Analysis
authorities’ reports. at a conference x
4.2.2. Data Gathering and Analysis
Poster
It isDuring
important thisto stressthe
phase, that all presentation
the significant
available
at a conference
financial companies
and production involveddatainwill
the be
oil collected
and gas industry
from theare New
x
either During
listed this markets
inExchange
stock phase, the or available
are publicly financial and with
detained, production
thorough
Legend:data will
public be collected
reporting.
Critical tasks. from the New
Therefore,
York Stock database [135,136] as well as in companies [137–143] other stock markets [144],
data York StockisExchange
gathering a feasible database
task using[135,136] as wellavailable
information as in companies
on the [137–143]
internet. other stock
Beyond markets
financial data,[144],
projects, and national authorities’ reports.
which projects,
allows and national
assessing 4.2.1. Literature
authorities’ Review
reports. and Data Gathering Planning
It is important towhether
stress that orders
all the and production
significant were
companies matched
involved byinsales,
the oilcompanies’ reportsare
and gas industry
It is important to stress that all the significant companies involved in the oil and gas industry are
either listed in stock markets The orfirst
are step
publicly
is to detained,
perform awith deeperthorough
literaturepublic reporting.
review, which Therefore,
will support the identification
either listed in stock markets or are publicly detained, with thorough public reporting. Therefore, data
data gathering is a feasible task using information available on the internet.
systemic and behavioral factors that have an impact on the bullwhip effect Beyond financial data,
in the oil and gas sup
gathering is a feasible task using information available on the internet. Beyond financial data, which
which allows assessingchain. whether Also,orders
a clear and production
definition were
of the oil matched by sales,
and gas supply companies’
chain and all thereports
stakeholders involved w
allows assessing whether orders and production were matched by sales, companies’ reports provide
be set up, in order to understand how each entity interacts with the others and the impact of e
entity on the bullwhip effect.
In this stage, the definition of which parameters should be gathered in order to characterize
oil and gas supply network should be established.

4.2.2. Data Gathering and Analysis

Mach. Learn. Knowl. Extr. 2019, 1 1004
Mach. Learn. Knowl. Extr. 2019, 1 1004

other important
provide information
other important on its business
information context, directly
on its business context, focusing on production
directly focusing and stocking
on production and
when significant.
stocking when significant.
Furthermore, most
Furthermore, most companies,
companies, joint
joint ventures,
ventures, and and large
large projects
projects voluntarily
voluntarily provide,
provide, in in their
their
websites and to the media, information about their activity, which can
websites and to the media, information about their activity, which can be used for redundancy and be used for redundancy
and checking.
checking.
Very significant
Very significant andand growing
growing digitalization
digitalization efforts,
efforts, undertaken
undertaken by by oil
oil and
and gasgas companies
companies
throughout recent times [145,146], assure not only a very wide availability
throughout recent times [145,146], assure not only a very wide availability of data, but also of data, but also data quality.
data
That is mostly
quality. That is due to the due
mostly employment of data analytics
to the employment techniques
of data analyticsfor techniques
data filtering, forstructuring,
data filtering,and
consolidation
structuring, and[93,147,148]. Furthermore,
consolidation [93,147,148].data analytics, including
Furthermore, severalincluding
data analytics, tools for big data tools
several analysis,
for
can be successfully used with ease by the researchers to extract high-quality
big data analysis, can be successfully used with ease by the researchers to extract high-quality information from the
aforementioned
information fromdatathe sources.
aforementioned data sources.
It is also noteworthy that aggregation on each level of the supply network could be paramount
for simulating it with available data. This This means
means that each each network level is considered as a collection
of nodes, each one with similar data stucture. At At every
every level,
level, financial
financial flow is gathered, using the
revenue for demand estimation and production costs costs for assessing offer.
for assessing offer.
In order to select which companies should be studied, it will be necessary to verify which ones
are directly involved in the oil and gas industry, and then verify whether they have a set of public
data within
within aa common
commonperiodperiodand
andcommon
commonnature.
nature.Another
Another selection
selection criterion
criterion is toisensure
to ensurethatthat
the
the companies
companies have have
notnot been
been merged,
merged, separated,acquired,
separated, acquired,ororundergone
undergone significant
significant organizational
changes in the period under study. This principle is paramount to guarantee that information about
production or demand data is not biased. One
production Onefurther
furtherrelevant
relevant need
needisisclosely
closelyanalyzing
analyzingthe thesuppliers’
suppliers’
level. In
chain level. Infact,
fact,equipment
equipmentand andservice
servicesuppliers
suppliers maymay interact
interact with
with several
several other
other chain
chain levels,
levels, as
as depicted in Figure 4; however, their most profuse activity
depicted in Figure 4; however, their most profuse activity is likely to be providedis likely to be provided to upstream
enterprises, leading
enterprises, leading to a higher vulnerability to the bullwhip effect. Therefore, companies must be
analyzed to tofigure,
figure,through
throughtheir client
their base
client andand
base financial data,
financial whether
data, they they
whether can be fitted
can on theon
be fitted upper
the
level of this supply chain. The companies should be arranged at the respective
upper level of this supply chain. The companies should be arranged at the respective level of the level of the supply
chain tochain
supply which tothey
whichbelong, as defined
they belong, in Figure
as defined in4.Figure 4.

Figure 4. Oil and gas supply chain.

Figure 4. Oil and gas supply chain.
Table 4. describes and summarizes the companies on each supply chain level.
Table 4. describes and summarizes the companies on each supply chain level.
Table 4. Description of the oil and gas supply chain levels.
Table 4. Description of the oil and gas supply chain levels.
Supply Chain Levels Description
Supply Chain The population includes companies that provide services and equipment to
Equipment and services Description
Levels support all the oil and gas activities, such as drilling, exploration, gathering,
suppliers
storing, and processing the oil and gas.
The population includes companies that provide services and equipment
Upstreamand
Equipment Crude oil and natural gas production
Midstream to support all the oil and gas activities,
Transportation such
of oil andasgas
drilling, exploration,
services suppliers
Downstream Refinegathering,
and processstoring,
crude oiland
andprocessing
natural gas tothebeoil and
sold to gas.
the consumers
Upstream Crude oil and natural gas production
Midstream
This Transportation
phase characterizes each company, using the number ofofemployees
oil and gas
and revenue, to understand
its size. This
Downstreaminformation
Refine and process crude oil and natural gas to be soldfor
will allow selecting the significant companies to the study, further
the consumers
investigating whether the bullwhip effect has a different level of importance for different company
sizes,This
as well as assuring
phase a consistent
characterizes business size
each company, along
using thethenumber
chain nodes.
of employees and revenue, to
understand its size. This information will allow selecting the significant companies for the study,
further investigating whether the bullwhip effect has a different level of importance for different
company sizes, as well as assuring a consistent business size along the chain nodes.
Mach. Learn. Knowl. Extr. 2019, 1 1005
Mach. Learn. Knowl. Extr. 2019, 1 1005
In fact, the oil and gas companies belong to a supply network, as shown in Figure 5. As noted in
significant research [14,24], it would be interesting to study the correlations between them, and to
In fact,
evaluate if ittheis oil and gas
possible tocompanies
study this belong to a supply
phenomenon network,
as a supply as shown
network, in Figure
instead of in5.relation
As noted to
in significant research [14,24], it would be interesting to study the correlations
supply chain behavior, in case significant correlations among companies are found. between them, and to
evaluate if it5ishas
Figure possible
been to study thiswith
populated phenomenon as a supply
some significant network,toinstead
companies better of in relation
illustrate thistoresearch
supply
chain behavior, in case significant correlations among companies are found.
idea with real cases and provide concrete suggestions for a preliminary investigation.

Figure 5. Oil
Figure5. Oil and
and gas
gas supply
supply network
network (example
(example with
with some
some companies).
companies).

Figure 5 has been populated with some significant companies to better illustrate this research idea
Based on the collected series, clients’ demands and orders submitted to suppliers shall be
with real cases and provide concrete suggestions for a preliminary investigation.
analyzed in order to determine the coefficient of variation in demand and supply, and evaluate the
Based on the collected series, clients’ demands and orders submitted to suppliers shall be analyzed
cyclical patterns of production and demand.
in order to determine the coefficient of variation in demand and supply, and evaluate the cyclical
The statistical treatment of the gathered data will allow developing a bullwhip effect model that
patterns of production and demand.
can be applied to the oil and gas industry. The aim of such analysis is to determine whether the higher
The statistical treatment of the gathered data will allow developing a bullwhip effect model that
level in the supply network is more variable than the lower level. Further on, the attained
can be applied to the oil and gas industry. The aim of such analysis is to determine whether the higher
mathematical model will be used to simulate the effects of demand fluctuations and assess the
level in the supply network is more variable than the lower level. Further on, the attained mathematical
bullwhip effect within an oil and gas supply network.
model will be used to simulate the effects of demand fluctuations and assess the bullwhip effect within
an oil and
4.2.3. gas supply
Artificial Neuralnetwork.
Network
4.2.3. The
Artificial NeuralofNetwork
next phase the research program consists of developing an ANN model to forecast the
demand. Artificial neural networks can accommodate nonlinear data to determine functional
The next phase of the research program consists of developing an ANN model to forecast
relationships between empirical data, even when the underlying relationships are unknown or
the demand. Artificial neural networks can accommodate nonlinear data to determine functional
difficult to describe.
relationships between empirical data, even when the underlying relationships are unknown or difficult
The choice of ANN methods will depend not only on the researchers’ proficiency, experience,
to describe.
and resources, but also on its suitability for a specific problem. Model conception and ANN
The choice of ANN methods will depend not only on the researchers’ proficiency, experience, and
adaptation will be shaped by the dimensions of the input and output variables, and continuously
resources, but also on its suitability for a specific problem. Model conception and ANN adaptation
reassessed as preliminary results are inspected.
will be shaped by the dimensions of the input and output variables, and continuously reassessed as
At this stage, different artificial neural networks based on different methods shall be used, and
preliminary results are inspected.
a comparative study shall be carried out in order to minimize the prediction error.
At this stage, different artificial neural networks based on different methods shall be used, and a
A brief of suggested ANN algorithms can be found in Section 2.2.1, while several practical issues
comparative study shall be carried out in order to minimize the prediction error.
are dealt with in Section 2.2.2, spanning some of the most common decisions that developers must
A brief of suggested ANN algorithms can be found in Section 2.2.1, while several practical issues
take.
are dealt with in Section 2.2.2, spanning some of the most common decisions that developers must take.
The last steps of the program will be devoted to generalized results and recommendations for
The last steps of the program will be devoted to generalized results and recommendations for
suitable supply networks. This will also allow suggesting interesting and relevant future works to
suitable supply networks. This will also allow suggesting interesting and relevant future works to the
the scientific community.
scientific community.
Mach. Learn. Knowl. Extr. 2019, 1 1006

5. Conclusions
Oil equipment and services suppliers are at the end of a complex supply network, and experience
a bullwhip effect with oil demand fluctuations. However, no extended and comprehensive studies
have been found yet on this issue, embodying a significant research gap. The relevancy of the
research work herein proposed is further highlighted by the emerging need to mitigate uncertainty
and inefficiency in an industry recently affected by a downturn and facing a fierce pressure to meet
sustainability requirements.
Considering the latest developments and increasing computing power of ANN, there are now
sufficiently mature resources to conduct the proposed research. Furthermore, data availability in the oil
and gas industry has increased significantly in recent years due to higher accountability requirements
and an industry wide effort toward digitalization.
The work outcome is deemed to be innovative and generalizable for similar problems, and
is expected to offer solutions with scientific and social impact. Both theoretical and managerial
applications are expect to be attained. Among the former, a better understanding and clarification
of the bullwhip effect through real data is, necessarily, an advance in supply chain theory. Secondly,
conclusions about the ability of ANN for forecasting the bullwhip effect could provide a sound scientific
advance, as would conclusions about the oil and gas supplier’s susceptibility to the bullwhip effect.
On the other hand, managerial applications can include developing an ANN-based tool for
modeling the oil and gas supply network and drawing validated proposals for mitigation measures.
Concerning the proposed study limitations, one shall highlight that, as a research idea, the current
proposal is necessarily limited by the researcher freedom it is deemed to cultivate, as well as by
the multitude of investigation paths it is supposed to offer. Yet, other important limitations lie in
the inherent complexity and uncertainty on bullwhip effect analysis based on real data. Data pool
idiosyncrasies will, undoubtedly, limit research outcome scope. Furthermore, attaining a pertinent,
reliable, and robust analysis tool for modeling through machine learning cannot be regarded as a
guaranteed result for the research program.
Data gathering and ANN development tasks shall be closely controlled, their planning shall be
prioritized, and redundancy shall be pursued.

Author Contributions: Conceptualization, A.L.S.; methodology, A.L.S.; investigation, A.L.S.; resources, A.L.S.;
writing—original draft preparation, A.L.S.; writing—review and editing, A.L.S. and T.P.R.; supervision, S.R.
and A.B.-P.
Funding: This research was funded by Fundação para a Ciência e a Tecnologia, I. P, grant number
SFRH/BD/131005/2017.
Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the
study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to
publish the results.

References
1. Forrester, J.W. Industrial Dynamics—A major breakthrought for decision makers. Harv. Bus. Rev.
1958, 36, 37–66.
2. Kadivar, M.; Shirazi, M.A. Analyzing the behavior of the bullwhip effect considering different distribution
systems. Appl. Math. Model. 2018, 59, 319–340. [CrossRef]
3. De Pacheco, E.O.; Cannella, S.; Lüders, R.; Barbosa-Povoa, A.P. Order-up-to-level policy update procedure
for a supply chain subject to market demand uncertainty. Comput. Ind. Eng. 2017, 113, 347–355. [CrossRef]
4. Sousa, A.L.; Matos, H.A.; Guerreiro, L.P. Preventing and removing wax deposition inside vertical wells:
A review. J. Pet. Explor. Prod. Technol. 2019, 1–17. [CrossRef]
5. Stock, T.; Obenaus, M.; Kunz, S.; Kohl, H. Industry 4.0 as enabler for a sustainable development: A qualitative
assessment of its ecological and social potential. Process. Saf. Environ. Prot. 2018, 118, 254–267. [CrossRef]
6. Müller, J.M.; Kiel, D.; Voigt, K.I. What drives the implementation of Industry 4.0? The role of opportunities
and challenges in the context of sustainability. Sustainability 2018, 10, 247. [CrossRef]
Mach. Learn. Knowl. Extr. 2019, 1 1007

7. Stock, T.; Seliger, G. Opportunities of Sustainable Manufacturing in Industry 4.0. Procedia CIRP
2016, 40, 536–541. [CrossRef]
8. Müller, J.M.; Däschle, S. Business Model Innovation of Industry 4.0 Solution Providers Towards Customer
Process Innovation. Processes 2018, 6, 260. [CrossRef]
9. Sodhi, M.S.; Tang, C.S. The incremental bullwhip effect of operational deviations in an arborescent supply
chain with requirements planning. Eur. J. Oper. Res. 2011, 215, 374–382. [CrossRef]
10. Lee, H.L.; Padmanabhan, V.; Whang, S. Information Distortion in a Supply Chain: The Bullwhip Effect.
Manag. Sci. 1997, 43, 546–558. [CrossRef]
11. Wang, X.; Disney, S.M. The bullwhip effect: Progress, trends and directions. Eur. J. Oper. Res.
2016, 250, 691–701. [CrossRef]
12. Dai, J.; Peng, S.; Li, S. Mitigation of Bullwhip Effect in Supply Chain Inventory Management Model.
Procedia Eng. 2017, 174, 1229–1234. [CrossRef]
13. Tanweer, A.; Li, Y.-Z.; Duan, G.; Song, J.-Y. An Optimization Model for Mitigating Bullwhip-effect in a
Two-echelon Supply Chain. Procedia -Soc. Behav. Sci. 2014, 138, 289–297. [CrossRef]
14. Sucky, E. The bullwhip effect in supply chains-An overestimated problem? Int. J. Prod. Econ.
2009, 118, 311–322. [CrossRef]
15. Birkel, H.S.; Veile, J.W.; Müller, J.M.; Hartmann, E.; Voigt, K.I. Development of a risk framework for Industry
4.0 in the context of sustainability for established manufacturers. Sustainability 2019, 11, 384. [CrossRef]
16. Mentzer, J.T.; DeWitt, W.; Keebler, J.S.; Nix, N.W.; Smith, C.D.; Zacharia, Z.G. Defining Supply Chain
Management. J. Bus. Logist. 2001, 22, 1–25. [CrossRef]
17. Forrester, J.W. Industrial dynamics—After the first decade. Manag. Sci. 1968, 14, 398–415. [CrossRef]
18. Zymelman, M. A Stabilization Policy for the Cotton Textile Cycle. Manag. Sci. 1965, 11, 572–580. [CrossRef]
19. Tsay, A.A.; Lovejoy, W.S. Quantity Flexibility Contracts and Supply Chain Performance. Manuf. Serv.
Oper. Manag. 1999, 1, 89–111. [CrossRef]
20. Sterman, J.D. Modeling Managerial Behavior: Misperceptions of Feedback in a Dynamic Decision Making
Experiment. Manag. Sci. 1989, 35, 321–339. [CrossRef]
21. Lee, H.L. The triple—A supply chain. Harv. Bus. Rev. 2004, 82, 102–113. [PubMed]
22. Yu, Z.; Yan, H.; Cheng, T.C.E. Benefits of information sharing with supply chain partnerships. Ind. Manag.
Data Syst. 2001, 101, 114–119. [CrossRef]
23. Simchi-Levi, P.; Kaminsky, P.; Simchi-Levi, E. Barilla Case Study. Available online: http://www2.newpaltz.
edu/~{}liush/MOIT/Barilla%20Case%20Study.pdf (accessed on 6 September 2019).
24. Toh, A.K.; Zhang, Q. A conceptual analysis of virtual supply chain collaborative network. Int. J. Soc. Syst. Sci.
2014, 6, 308–330. [CrossRef]
25. Fu, D.; Ionescu, C.; Aghezzaf, E.-H.; De Keyser, R. Quantifying and mitigating the bullwhip effect
in a benchmark supply chain system by an extended prediction self-adaptive control ordering policy.
Comput. Ind. Eng. 2015, 81, 46–57. [CrossRef]
26. Liu, P.; Yi, S.P. Investment Decision-Making and Coordination of Supply Chain: A New Research in the Big
Data Era. Discret. Dyn. Nat. Soc. 2016, 2016, 2026715. [CrossRef]
27. Cannella, S.; Ciancimino, E. On the bullwhip avoidance phase: Supply chain collaboration and order
smoothing. Int. J. Prod. Res. 2010, 48, 6739–6776. [CrossRef]
28. Croson, R.; Donohue, K. Behavioral Causes of the Bullwhip Effect and the Observed Value of Inventory
Information. Manag. Sci. 2006, 52, 323–336. [CrossRef]
29. Ciancimino, E.; Cannella, S.; Bruccoleri, M.; Framinan, J.M. On the Bullwhip Avoidance Phase:
The Synchronised Supply Chain. Eur. J. Oper. Res. 2012, 221, 49–63. [CrossRef]
30. Moyaux, T.; Brahim, C.; D’Amours, S. Information Sharing as a Coordination Mechanism for Reducing the
Bullwhip Effect in a Supply Chain. IEEE Trans. Syst. Man Cybern. Part. C 2007, 37, 396–409. [CrossRef]
31. Lotfi, Z.; Mukhtar, M.; Sahran, S.; Zadeh, A.T. Information Sharing in Supply Chain Management.
Procedia Technol. 2013, 11, 298–304. [CrossRef]
32. Liang, N.Y.; Huang, G.B.; Saratchandran, P.; Sundararajan, N. A fast and accurate online sequential learning
algorithm for feedforward networks. IEEE Trans. Neural Netw. 2006, 17, 1411–1423. [CrossRef] [PubMed]
33. AlSudairi, M.; Vasista, T.G.; Zamil, A.M.; Algharabat, R.S. Mitigating the Bullwhip Effect with eWord of
Mouth: eBusiness Intelligence Perspective. Int. J. Manag. Value Supply Chain 2012, 3, 27–41. [CrossRef]
Mach. Learn. Knowl. Extr. 2019, 1 1008

34. Lampret, T.; Potočan, V. Bullwhip Effect in the Information Flow of a Supply Chain: A Role of Culture.
Logist. Sustain. Transp. 2014, 5, 34–45. [CrossRef]
35. Agrawal, S.; Sengupta, R.N.; Shanker, K. Impact of information sharing and lead time on bullwhip effect and
on-hand inventory. Eur. J. Oper. Res. 2009, 192, 576–593. [CrossRef]
36. de Almeida, M.M.K.; Marins, F.A.S.; Salgado, A.M.P.; Santos, F.C.A.; da Silva, S.L. Mitigation of the bullwhip
effect considering trust and collaboration in supply chain management: A literature review. Int. J. Adv.
Manuf. Technol. 2015, 77, 495–513. [CrossRef]
37. Shizhengt, G.; Zhen, L.; Xiaoyuan, H. Stochastic Control of Bullwhip Effect in Supply Chain. In Proceedings
of the 2006 IEEE International Conference on Service Systems and Service Management, Troyes, France,
25–27 October 2006; Volume 2, pp. 1455–1459.
38. Huang, X.Y.; Yan, N.N.; Guo, H.F. An H∞ control method of the bullwhip effect for a class of supply chain
system. Int. J. Prod. Res. 2007, 45, 207–226. [CrossRef]
39. Jacoby, D. The oil price ‘bullwhip’: Problem, cost, response. Oil Gas. J. 2010, 108, 20.
40. Zhang, J.H.; Zhang, Q.Q. The System Dynamics Analyses of Bullwhip Effect in China Processed Oil Supply
Chain. Appl. Mech. Mater. 2013, 295, 3310–3317. [CrossRef]
41. Sherhart, E. The Bullwhip Effect: Recognizing the Phenomenon and Mitigating It Using the Theory of Constraints
Illustrated by a Case Study from British Petroleum; Baylor University: Waco, TX, USA, 2013.
42. Croson, R.; Donohue, K. Impact of Pos Data Sharing on Supply Chain Management: An Experimental Study.
Prod. Oper. Manag. 2003, 12, 1–11. [CrossRef]
43. Caloiero, G.; Strozzi, F.; Comenges, J.M.Z. A supply chain as a series of filters or amplifiers of the bullwhip
effect. Int. J. Prod. Econ. 2008, 114, 631–645. [CrossRef]
44. Geary, S.; Disney, S.M.; Towill, D.R. On bullwhip in supply chains - Historical review, present practice and
expected future impact. Int. J. Prod. Econ. 2006, 101, 2–18. [CrossRef]
45. Springer, M.; Kim, I. Managing the order pipeline to reduce supply chain volatility. Eur. J. Oper. Res.
2010, 203, 380–392. [CrossRef]
46. Kahn, J.A. Inventories and the Volatility of Production. Am. Econ. Rev. 1987, 77, 667–679.
47. Carlsson, C.; Fullér, R. Reducing the bullwhip effect by means of intelligent, soft computing methods.
In Proceedings of the 34th Annual Hawaii International Conference on System Sciences, Maui, HI, USA,
6 January 2001.
48. Chen, F.; Drezner, Z.; Ryan, J.K.; Simchi-Levi, D. Quantifying the Bullwhip Effect in a Simple Supply Chain:
The Impact of Forecasting, Lead Times, and Information. Manag. Sci. 2000, 46, 436–443. [CrossRef]
49. Dejonckheere, J.; Disney, S.M.; Lambrecht, M.R.; Towill, D.R. Measuring and avoiding the bullwhip effect:
A control theoretic approach. Eur. J. Oper. Res. 2003, 147, 567–590. [CrossRef]
50. Metters, R. Quantifying the bullwhip effect in supply chains. J. Oper. Manag. 1997, 15, 89–100. [CrossRef]
51. Zhou, L.; Disney, S.M. Bullwhip and inventory variance in a closed loop supply chain. OR Spectr.
2006, 28, 127–149. [CrossRef]
52. Nagaraja, C.H.; McElroy, T. The multivariate bullwhip effect. Eur. J. Oper. Res. 2017, 267, 96–106. [CrossRef]
53. Nienhaus, J.; Ziegenbein, A.; Schoensleben, P. How human behaviour amplifies the bullwhip effect. A study
based on the beer distribution game online. Prod. Plan. Control. 2006, 17, 547–557. [CrossRef]
54. Sarkar, S.; Kumar, S. Demonstrating the Effect of Supply Chain Disruptions through an Online Beer
Distribution Game. Decis. Sci. J. Innov. Educ. 2016, 14, 25–35. [CrossRef]
55. Cachon, G.P.; Randall, T.; Schmidt, G.M. In Search of the Bullwhip Effect. Manuf. Serv. Oper. Manag.
2007, 9, 457–479. [CrossRef]
56. Chen, F.; Drezner, Z.; Ryan, J.K.; Simchi-Levi, D. The bullwhip effect: Managerial insights on the impact
of forecasting and information on variability in a supply chain. Quant. Model. Supply Chain Manag.
1999, 17, 419–439.
57. Barlas, Y.; Gunduz, B. Demand forecasting and sharing strategies to reduce fluctuations and the bullwhip
effect in supply chains. J. Oper. Res. Soc. 2011, 62, 458–473. [CrossRef]
58. Fu, D.; Ionescu, C.M.; Aghezzaf, E.H.; de Keyser, R. Decentralized and centralized model predictive control
to reduce the bullwhip effect in supply chain management. Comput. Ind. Eng. 2014, 73, 21–31. [CrossRef]
59. Feng, J.; Wang, H. Controlling the bullwhip effect of supply chain system based on the stability analysis.
2010 Int. Conf. Logist. Eng. Manag. Logist. Sustain. Econ. Dev. -Infrastruct. Inf. Integr. ICLEM 2010
2010, 387, 4336–4345.
Mach. Learn. Knowl. Extr. 2019, 1 1009

60. Chao, Y. The Bullwhip effect in supply chain and countermeasures. Adv. Mater. Res. 2013, 711, 799–804.
[CrossRef]
61. Wright, D.; Yuan, X. Mitigating the bullwhip effect by ordering policies and forecasting methods. Int. J.
Prod. Econ. 2008, 113, 587–597. [CrossRef]
62. Coppini, M.; Rossignoli, C.; Rossi, T.; Strozzi, F. Bullwhip effect and inventory oscillations analysis using the
beer game model. Int. J. Prod. Res. 2010, 48, 3943–3956. [CrossRef]
63. Disney, S.M.; Towill, D.R. Vendor-managed inventory and bullwhip reduction in a two-level supply chain.
Int. J. Oper. Prod. Manag. 2003, 23, 625–651. [CrossRef]
64. Ingalls, R.G.; Foote, B.L.; Krishnamoorthy, A. Reducing the bullwhip effect in supply chains with control-based
forecasting. Int. J. Simul. Process. Model. 2005, 1, 90–110. [CrossRef]
65. Makajić-Nikolić, D.; Panić, B.; Vujošević, M. Bullwhip effect and supply chain modelling and analysis using
CPN Tools. In Proceedings of the Fifth Workshop and Tutorial on Practical Use of Coloured Petri Nets and
the CPN Tools, Aarhus, Denmark, 8–11 October 2004; pp. 219–234.
66. Lockwood, A. The Modulation of Wound Healing for Glaucoma Filtration Surgery. Ph.D. Thesis, University
College London, London, UK, 2014.
67. Hahn, C.K.; Duplaga, E.A.; Hartley, J.L. Supply-Chain Synchronization: Lessons from Hyundai Motor
Company. Interfaces 2000, 30, 32–45. [CrossRef]
68. Coleman, J. Extending supply chain synchronisation to upstream tiers: A collaborative approach illustrated
with an automotive case study. Int. J. Netw. Virtual Organ. 2010, 7, 257–271. [CrossRef]
69. Aggelogiannaki, E.; Doganis, P.; Sarimveis, H. An adaptive model predictive control configuration for
production-inventory systems. Int. J. Prod. Econ. 2008, 114, 165–178. [CrossRef]
70. Chaharsooghi, S.K.; Heydari, J. LT variance or LT mean reduction in supply chain management: Which one
has a higher impact on SC performance? Int. J. Prod. Econ. 2010, 124, 475–481. [CrossRef]
71. Chandra, C.; Grabis, J. Application of multi-steps forecasting for restraining the bullwhip effect and improving
inventory performance under autoregressive demand. Eur. J. Oper. Res. 2005, 166, 337–350. [CrossRef]
72. Chatfield, D.C.; Kim, J.G.; Harrison, T.P.; Hayya, J.C. The Bullwhip Effect-Impact of Stochastic Lead Time,
Information Quality, and Information Sharing: A Simulation Study. Prod. Oper. Manag. 2009, 13, 340–353.
[CrossRef]
73. Chen, L.; Lee, H.L. Information Sharing and Order Variability Control Under a Generalized Demand Model.
Manag. Sci. 2009, 55, 781–797. [CrossRef]
74. Disney, S.M.; Towill, D.R.; van de Velde, W. Variance amplification and the golden ratio in production and
inventory control. Int. J. Prod. Econ. 2004, 90, 295–309. [CrossRef]
75. Disney, S.M.; Farasyn, I.; Lambrecht, M.; Towill, D.R.; van de Velde, W. Taming the bullwhip effect whilst
watching customer service in a single supply chain echelon. Eur. J. Oper. Res. 2006, 173, 151–172. [CrossRef]
76. Jakšič, M.; Rusjan, B. The effect of replenishment policies on the bullwhip effect: A transfer function approach.
Eur. J. Oper. Res. 2008, 184, 946–961. [CrossRef]
77. Kelepouris, T.; Miliotis, P.; Pramatari, K. The impact of replenishment parameters and information sharing
on the bullwhip effect: A computational study. Comput. Oper. Res. 2008, 35, 3657–3670. [CrossRef]
78. Kim, J.G.; Chatfield, D.; Harrison, T.P.; Hayya, J.C. Quantifying the bullwhip effect in a supply chain with
stochastic lead time. Eur. J. Oper. Res. 2006, 173, 617–636. [CrossRef]
79. Zhang, X. The impact of forecasting methods on the bullwhip effect. Int. J. Prod. Econ. 2004, 88, 15–27.
[CrossRef]
80. Dejonckheere, J.; Disney, S.M.; Lambrecht, M.R.; Towill, D.R. Transfer function analysis of forecasting induced
bullwhip in supply chains. Int. J. Prod. Econ. 2002, 78, 133–144. [CrossRef]
81. Kim, I.; Springer, M. Measuring endogenous supply chain volatility: Beyond the bullwhip effect. Eur. J.
Oper. Res. 2008, 189, 172–193. [CrossRef]
82. Boute, R.N.; Disney, S.M.; Lambrecht, M.R.; van Houdt, B. An integrated production and inventory model to
dampen upstream demand variability in the supply chain. Eur. J. Oper. Res. 2007, 178, 121–142. [CrossRef]
83. Wang, J.L.; Kuo, J.H.; Chou, S.Y.; Wang, S.Z. A comparison of bullwhip effect in a single-stage supply
chain for autocorrelated demands when using Correct, MA, and EWMA methods. Expert Syst. Appl.
2010, 37, 4726–4736. [CrossRef]
84. Cachon, G.P.; Fisher, M. Supply Chain Inventory Management and the Value of Shared Information.
Manag. Sci. 2000, 46, 1032–1048. [CrossRef]
Mach. Learn. Knowl. Extr. 2019, 1 1010

85. Disney, S.M.; Farasyn, I.; Lambrecht, M.R.; Towill, D.R.; van de Velde, W. Controlling bullwhip and inventory
variability with the golden smoothing rule. Eur. J. Ind. Eng. 2007, 1, 241. [CrossRef]
86. Wang, P. Three fundamental misconceptions of Artificial Intelligence. J. Exp. Theor. Artif. Intell.
2007, 19, 249–268. [CrossRef]
87. Brown, D.J.H. Artificial Intelligence—A New Discipline in the Computer Sciences (A Viewpoint). Kybernetes
1977, 6, 245–256. [CrossRef]
88. Cucker, F.; Smale, S. On the mathematical foundations of learning. Bull. New. Ser. Am. Math. Soc.
2001, 39, 1–49. [CrossRef]
89. Bravo, C.E.; Saputelli, L.; Rivas, F.; Pérez, A.G.; Nickolaou, M.; Zangl, G.; De Guzmán, N.; Mohaghegh, S.D.;
Nunez, G. State of the Art of Artificial Intelligence and Predictive Analytics in the E & P Industry: A Technology
Survey. SPE J. Pet. Technol. 2014, 19, 547–563.
90. Bello, O.; Oppelt, J.; Holzmann, J.; Obiwanne, A. Application of Artificial Intelligence Techniques in
Drilling System Design and Operations: A State of the Art Review and Future Research Pathways. SPE J.
Pet. Technol. 2016. [CrossRef]
91. Litjens, G.; Sánchez, C.I.; Timofeeva, N.; Hermsen, M.; Nagtegaal, I.; Kovacs, I.; Hulsbergen-Van De Kaa, C.;
Bult, P.; Van Ginneken, B.; Van Der Laak, J. Deep learning as a tool for increased accuracy and efficiency of
histopathological diagnosis. Sci. Rep. 2016, 6, 26286. [CrossRef] [PubMed]
92. Edwards, R.E.; New, J.; Parker, L.E.; Cui, B.; Dong, J. Constructing large scale surrogate models from big data
and artificial intelligence. Appl. Energy 2017, 202, 685–699. [CrossRef]
93. Carvajal, G.; Maucec, M.; Cullick, S. Intelligent Digital Oil and Gas. Fields. Concepts, Collaboration, and
Right-Time Decisions; Gulf Professional Publishing: Houston, TX, USA, 2018.
94. Abambres, M.; Rajana, K.; Tsavdaridis, K.D.; Ribeiro, T.P. Neural Network-Based Formula for the Buckling
Load Prediction of I-Section Cellular Steel Beams. Computers 2018, 8, 2. [CrossRef]
95. Hertzmann, A.; Fleet, D. Machine Learning and Data Mining Lecture Notes; University of Toronto:
Toronto, ON, Canada, 2012.
96. McCulloch, W.S.; Pitts, W. A logical calculus of the ideas immanent in nervous activity. Bull. Math. Biophys.
1943, 5, 115–133. [CrossRef]
97. Prieto, A.; Prieto, B.; Ortigosa, E.M.; Ros, E.; Pelayo, F.; Ortega, J.; Rojas, I. Neural networks: An overview of
early research, current frameworks and new challenges. Neurocomputing 2016, 214, 242–268. [CrossRef]
98. Wilamowski, B.M.; Irwin, J.D. The Industrial Electronics Handbook: Intelligent Systems 1; Taylor and Francis
Group: Didcot, UK, 2011.
99. Flood, I. Towards the next generation of artificial neural networks for civil engineering. Adv. Eng. Inform.
2008, 22, 4–14. [CrossRef]
100. Ferri, M. Why Topology for Machine Learning and Knowledge Extraction? Mach. Learn. Knowl. Extr.
2019, 1, 115–120. [CrossRef]
101. Rumelhart, D.E.; Hinton, G.E.; Williams, R.J. Learning representations by back-propagating errors. Nature
1986, 324, 698–699. [CrossRef]
102. Hagan, M.T.; Menhaj, M.B. Training Feedforward Networks with the Marquardt Algorithm. IEEE Trans.
Neural Netw. 1994, 5, 2–6. [CrossRef] [PubMed]
103. Huang, G.-B.; Zhu, Q.-Y.; Siew, C.-K. Extreme learning machine: Algorithm, theory and applications.
Neurocomputing 2006, 70, 489–501. [CrossRef]
104. Huang, G.-B.; Chen, L.; Siew, C.-K. Universial approximation using incremental constructive feedforward
neural networks with random hidden nodes. IEEE Trans. Neural Netw. 2006, 17, 879–892. [CrossRef]
[PubMed]
105. Huang, G.B.; Chen, L. Convex incremental extreme learning machine. Neurocomputing 2007, 70, 16–18, 3056–3062.
[CrossRef]
106. Hutter, F.; Xu, L.; Hoos, H.H.; Leyton-brown, K. Algorithm runtime prediction: Methods & evaluation.
Artif. Intell. 2014, 206, 79–111.
107. Liu, H. Forecasting Model of Supply Chain Management Based on Neural Network. In Proceedings of
the 2015 International Conference on Automation, Mechanical Control and Computational Engineering,
Changsha, China, 24–26 April 2015; Volume 124, pp. 177–181.
108. Musa, A.; Dehmer, M.; Yli-harja, O.; Emmert-streib, F. Exploiting Genomic Relations in Big Data Repositories
by Graph-Based Search Methods. Mach. Learn. Knowl. Extr. 2018, 1, 205–210. [CrossRef]
Mach. Learn. Knowl. Extr. 2019, 1 1011

109. Gholizadeh, S.; Pirmoz, A.; Attarnejad, R. Assessment of load carrying capacity of castellated steel beams by
neural networks. J. Constr. Steel Res. 2011, 67, 770–779. [CrossRef]
110. Kasun, L.L.C.; Yang, Y.; Huang, G.B.; Zhang, Z. Dimension Reduction with Extreme Learning Machine.
IEEE Trans. Image Process. 2016, 25, 3906–3918. [CrossRef] [PubMed]
111. Lachtermacher, G.; Fuller, J.D. Backpropagation in Time-series Forecasting. J. Forecast. 1995, 14, 381–393.
[CrossRef]
112. Pu, Y.; Mesbahi, E. Application of artificial neural networks to evaluation of ultimate strength of steel panels.
Eng. Struct. 2006, 28, 1190–1196. [CrossRef]
113. Tohidi, S. Inelastic lateral-torsional buckling capacity of corroded web opening steel beams using artificial
neural networks. IES J. Part. A Civ. Struct. Eng. 2015, 8, 24–40. [CrossRef]
114. Flood, I.; Kartam, N. Neural Networks in Civil Engineering I: Principles and Understanding. J. Comput.
Civ. Eng. 1994, 8, 131–148. [CrossRef]
115. Mukherjee, A.; Deshpande, J.M.; Anmala, J. Prediction of Buckling Load of Columns using Artificial Neural
Networks. J. Struct. Eng. 2006, 122, 1385–1387. [CrossRef]
116. Dyckhoff, H.; Lackes, R.; Reese, J. Supply Chain Management and Reverse Logistics; Springer:
Berlin/Heidelberg, Germany, 2004.
117. Carbonneau, R.; Laframboise, K.; Vahidov, R. Application of machine learning techniques for supply chain
demand forecasting. Eur. J. Oper. Res. 2008, 184, 1140–1154. [CrossRef]
118. Fradinata, E.; Suthummanon, S.; Suntiamorntut, W. Reducing the bullwhip effect from signal demand of
hybrid artificial neural network models of supply chain in Indonesia. Int. J. Adv. Appl. Sci. 2017, 4, 64–75.
[CrossRef]
119. Slimani, I.; El Farissi, I.; Achchab, S. Artificial neural networks for demand forecasting: Application using
Moroccan supermarket data. In Proceedings of the 2015 15th International Conference on Intelligent Systems
Design and Applications (ISDA), Marrakech, Morocco, 14–16 December 2016; pp. 266–271.
120. Chawla, A.; Singh, A.; Lamba, A.; Gangwani, N. Demand Forecasting Using Artificial Neural Networks—A
Case Study of American Retail Corporation. In Applications of Artificial Intelligence Techniques in Engineering;
Springer: Singapore, 2019; Volume 697, pp. 79–89.
121. Prakash, A.; Chan, F.T.S.; Liao, H.; Deshmukh, S.G. Network optimization in supply chain: A KBGA
approach. Decis. Support. Syst. 2012, 52, 528–538. [CrossRef]
122. Wang, F.; Lai, X.; Shi, N. A multi-objective optimization for green supply chain network design.
Decis. Support. Syst. 2011, 51, 262–269. [CrossRef]
123. Lau, H.C.W.; Ho, G.T.S.; Zhao, Y. A demand forecast model using a combination of surrogate data analysis
and optimal neural network approach. Decis. Support. Syst. 2013, 54, 1404–1416. [CrossRef]
124. Park, H.; Bellamy, M.A.; Basole, R.C. Visual Analytics for Supply Network Management: System Design and
Evaluation. Decis. Support. Syst. 2016, 91, 89–102. [CrossRef]
125. Beheshti, H.M. A decision support system for improving performance of inventory management in a supply
chain network. Int. J. Product. Perform. Manag. 2010, 59, 452–467. [CrossRef]
126. Bhattacharya, S.; Xu, D.; Kumar, K. An ANN-based auditor decision support system using Benford’s law.
Decis. Support. Syst. 2011, 50, 576–584. [CrossRef]
127. Ilie-Zudor, E.; Ekárt, A.; Kemeny, Z.; Buckingham, C.; Welch, P.; Monostori, L. Advanced
predictive-analysis-based decision support for collaborative logistics networks. Supply Chain Manag.
An. Int. J. 2015, 20, 369–388. [CrossRef]
128. Kuo, R.J.; Chen, C.H.; Hwang, Y.C. An intelligent stock trading decision support system through integration of
genetic algorithm based fuzzy neural network and artificial neural network. Fuzzy Sets Syst. 2001, 118, 21–45.
[CrossRef]
129. Ahmadi, M.A.; Soleimani, R.; Lee, M.; Kashiwao, T.; Bahadori, A. Determination of oil well production
performance using artificial neural network (ANN) linked to the particle swarm optimization (PSO) tool.
Petroleum 2015, 1, 118–132. [CrossRef]
130. Mahdiani, M.R.; Khamehchi, E. A modified neural network model for predicting the crude oil price.
Intellect. Econ. 2016, 10, 71–77. [CrossRef]
131. Mirzaei-Paiaman, A.; Salavati, S. The application of artificial neural networks for the prediction of oil
production flow rate. Energy Sources Part. A Recover. Util. Environ. Eff. 2012, 34, 1834–1843. [CrossRef]
Mach. Learn. Knowl. Extr. 2019, 1 1012

132. Torabi, F.; Abedini, A.; Abedini, R. The development of an artificial neural network model for prediction of
crude oil viscosities. Pet. Sci. Technol. 2011, 29, 804–816. [CrossRef]
133. Arjun, K.S.; Aneesh, K. Modelling studies by application of artificial neural network using matlab. J. Eng.
Sci. Technol. 2015, 10, 1477–1486.
134. Sage, A.P.; Armstrong, J.E., Jr. Introduction to Systems Engineering; Wiley: Hoboken, NJ, USA, 2000.
135. Azhar, M. The Study of the Bullwhip Effect in the Oil and Gas. Industry; Oklahoma State University:
Stillwater, OK, USA, 2013.
136. NYSE. New York Stock Exchange Database. Available online: https://www.nyse.com/index (accessed on
30 July 2019).
137. Schlumberger. Schlumberger Newsroom. Available online: https://www.slb.com/newsroom/ (accessed on
30 July 2019).
138. Encana Corporation. Encana Quarterly Reports. Available online: https://www.encana.com/investors/
financial/quarterly-reports/ (accessed on 30 July 2019).
139. Sempra Energy. Sempra Energy Quarterly Reports. Available online: https://www.sempra.com/investors
(accessed on 30 July 2019).
140. Valero Energy Company. Valero Quarterly Earnings. Available online: http://www.investorvalero.com/
financial-information/quarterly-results (accessed on 30 July 2019).
141. Varco Inc. National Oilwell Varco Quarterly Results. Available online: https://investors.nov.com/financial-
information/quarterly-results/ (accessed on 30 July 2019).
142. Halliburton. Halliburton Quarterly Results. Available online: http://ir.halliburton.com/financial-information/
quarterly-results (accessed on 30 July 2019).
143. Baker Hughes. Baker Hughes Investor Center. Available online: https://investors.bhge.com/financial-
information/quarterly-results (accessed on 30 July 2019).
144. DataHub. Stock Market Data. Available online: https://datahub.io/collections/stock-market-data (accessed
on 30 July 2019).
145. Equinor. Digitalisation Driving Value Creation. Available online: https://www.equinor.com/en/news/
digitalisation-driving-value-creation.html (accessed on 6 September 2017).
146. World Economic Forum and Accenture. Digital Transformation Initiative: Oil and Gas Industry. Available
online: http://reports.weforum.org/digital-transformation/wp-content/blogs.dir/94/mp/files/pages/files/dti-
oil-and-gas-industry-white-paper.pdf (accessed on 6 September 2017).
147. Kaisler, S.; Armour, F.; Espinosa, J.A.; Money, W. Big data: Issues and challenges moving forward.
In Proceedings of the Annual Hawaii International Conference on System Sciences, Hawaii, HA, USA,
7–10 January 2013; pp. 995–1004.
148. Yoo, Y. It is not about Size: A Further thought on Big Data. J. Inf. Technol. 2015, 30, 63–65. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).