See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/366645510

E-LOGISTICS FOR INDUSTRY 4.0: A DATA-DRIVEN ANALYSIS OF THE

CHALLENGES AND BARRIERS TO E- PROCUREMENT AND TIME LAG IN
TRANSPORTATION SCHEDULING

Research Proposal · December 2022

CITATIONS READS
0 718

1 author:

Olatunji Apampa
Tuni Knowledge Systems
21 PUBLICATIONS 44 CITATIONS

SEE PROFILE

All content following this page was uploaded by Olatunji Apampa on 28 December 2022.

The user has requested enhancement of the downloaded file.

 E-LOGISTICS FOR INDUSTRY 4.0: A DATA-DRIVEN
ANALYSIS OF THE CHALLENGES AND BARRIERS TO E-
PROCUREMENT AND TIME LAG IN TRANSPORTATION
SCHEDULING

PROPOSAL IN SUPPORT OF PH.D. ADMISSION INTO DIGITAL

HUMANITIES/COMMUNICATION & INFORMATION SCIENCE, DATA
SCIENCE, BIG DATA & BUSINESS ANALYTICS PROGRAMMES

BY

OLATUNJI APAMPA
[M.INF. SC. UNIVERSITY OF IBADAN, NIGERIA (2005): M.SC. COMMUNICATION
AND INFORMATION SCIENCE, TILBURG UNIVERSITY, THE NETHERLANDS (2016)]

DECEMBER 2022
1.0 Introduction
Big data analytics is concerned with sieving through enormous volumes of data, and intricate
databases in order to find obscure patterns, undiscovered correlations, and other valuable
insights and information that can aid businesses in making better decisions. Big data analytics
can be used to enhance a variety of business processes in the field of logistics and supply chain
management, including demand forecasting, inventory management, and transportation
planning. Businesses can obtain important insights into the effectiveness and efficiency of their
supply chain and logistics processes by gathering and analyzing vast amounts of data from several
sources. They can utilize big data analytics, for instance, to locate operational bottlenecks, plan
the best routes for transportation and the movement of goods and services, and more precisely
forecast client demand. Furthermore, big data analytics can assist companies in strengthening
their communication with consumers and other stakeholders.
The movement and storage of commodities, services, and information from the point of origin
to the site of consumption are referred to as logistics activities. These operations are planned,
carried out, and controlled. Logistics operations are essential in the context of supply chain
management in ensuring that the appropriate products are delivered to the right location at the
right time, in the right quantity, and for the right price. Transportation, warehousing, inventory
control, and information systems are only a few of the resources and activities that make up an
efficient logistics operation network. To adapt to changing consumer and stakeholder demands,
these activities and resources must be properly managed and optimized.
In today's globalized and highly competitive business environment, logistics operations are
becoming increasingly complex and challenging. To provide value to their consumers and keep a
competitive edge, companies must negotiate a variety of circumstances, such as fluctuating
demand, changing client preferences, shifting economic situations, and regulatory obligations.
Digital technology can assist businesses in managing and optimizing their logistics operations.
Examples include supply chain management software and e-logistics.

1.1 What is E-Logistics

Electronic logistics, or e-logistics, is the application of digital technologies to enhance and
optimize supply chain operations. E-logistics is concerned with managing the flow of products,
services, information, and financial transactions throughout the supply chain, this includes the
use of computer systems, the internet, and other electronic communication methods. The use of
e-logistics can significantly improve supply chain operations by boosting productivity, cutting
costs, and enhancing information quality and timeliness. For instance, using electronic data
interchange (EDI) and other digital communication tools can speed up and improve the accuracy
of information flow between trading partners, enabling the processing and delivery of orders.
Real-time tracking and monitoring systems can be used with e-logistics, which can increase

1
control and visibility of the supply chain and help to identify and resolve potential issues before
they become problems.
In addition to these advantages, e-logistics can also contribute to the sustainability of supply
chain operations by lowering the reliance on paper-based modes of communication and
transportation and promoting the use of more effective and environmentally friendly ones.
Overall, e-logistics adoption can significantly improve supply chain operations' performance and
support organizations' ability to compete in today's quick-paced and increasingly globalized
market.
According to Kanagavalli and Azeez (2019), Logistics Management is an effective system for
companies to deliver products to customers at the exact place where they demand at right time
with the right cost. Logistics management comprises the planning, implementation and control
of the flow of raw materials and finished goods from the suppliers to the customers.
Due to globalization, most large corporations are now required to cut back on their company
expenses, and these corporations are working to find ways to do so without sacrificing efficiency.
They began using new management techniques as a result, including supply chain management
and e-logistics. E-logistics totally transforms the idea of simple logistics into e-logistics by bringing
the logistics business into the current world via the internet. The most recent innovation in
logistics, known as e-logistics, condenses the whole corporate world into a single virtual space.
By exchanging data, delivering information, and precise statistics with the supply chain clients,
this revolutionary set of communication transforms the new logistics into a customer-oriented
technology. E-logistics aids in managing the industry's newest supply chain difficulties. Multi-
channel operations, cross-border tasks, warehouse planning, inventory, planning, estimating,
and performance management are the core elements of e-logistics (Kanagavalli and Azeez, 2019).

1.2 SMEs and the Adoption of E-Logistics in the era of Industry 4.0
Businesses classified as small and medium-sized enterprises (SMEs) typically have fewer than 250
employees and a yearly revenue of less than €50 million. Small and medium-sized businesses
(SMEs) play a substantial role in the economies of most nations, contributing significantly to both
employment and GDP (GDP). SMEs can be categorised as sole proprietorships, partnerships, or
corporations, among other configurations. They can work in many different sectors, including
manufacturing, retail, and services. Compared to huge multinational firms, SMEs are frequently
more adaptable and flexible, and they can frequently react swiftly to market developments.
However, SMEs also have particular difficulties, such as restricted access to resources and
finance, and may find it difficult to compete with bigger businesses.
The Fourth Industrial Revolution, or Industry 4.0, is the term used to describe the present trend
of automation and data sharing in the manufacturing and other sectors of the economy. By using
Industry 4.0 technologies like e-logistics and big data analytics, small and medium-sized

2
businesses (SMEs) can increase productivity, cut costs, and gain a competitive advantage. As
already espoused above, e-logistics, commonly referred to as digital logistics or online logistics,
is the application of digital technology to the supply chain to streamline the movement and
storage of commodities. Implementing e-logistics can help small and medium-sized businesses
(SMEs) in a number of ways:
(i) Reduced costs: By automating logistics processes and using data analytics to optimize routes
and schedules, SMEs can reduce transportation and storage costs.
(ii) Cost savings: SMEs can cut transportation and storage expenses by automating logistics
procedures and utilizing data analytics to optimize routes and schedules.
(iii) E-logistics systems can help SMEs provide real-time tracking and updates to customers,
improving their satisfaction and loyalty.
(iv) Increased competitiveness: By implementing e-logistics, SMEs can outperform their larger
counterparts, gain a competitive edge over their rivals and better serve the needs of their
customers.

1.3 Objectives of Study and Research Questions

The objective of the study is to determine empirically how the use of big data analytics tools (such
as data mining and machine learning) can be combined with electronic logistics in industry 4.0
for the optimisation of procurement activities, transportation scheduling, and performance
management in SMEs and/or multinational corporations.
Specifically, the study should provide cogent answers to the following research questions.
1. What are the insights and new knowledge derived from the application of predictive
and prescriptive data analysis to the use of an electronic procurement system (e-
procurement)?
2. What are the effects of E-Logistics on the reduction of time lags in the transportation
of goods and services to their destinations?
3. How can data from social media platforms (Twitter) be used to improve marketing
and transportation in the online retail industry?

1.4 Limitations of the Study

This study is limited by a number of factors. Most important of which is data from companies that are
involved in the logistics and supply chain sector of the economy. Access to records and data from the
Enterprise Resource Planning (ERP) database and social media platform(s) is critical to the overall
success of the project. The study is targeted primarily at SMEs.

3
2.0 Literature Review

Big data and its analysis have received a lot of interest since the advent of Web 2.0, Industry 4.0,
the Internet of Things (IoT), and other digital technologies. Globally, interest in this research area
is growing since it can both increase performance and provide new insights and knowledge of an
organization based on the kinds of data that they keep. Currently, enormous amounts of data are
gathered from a variety of sources, including enterprise resource planning (ERP) systems,
distributed manufacturing environments, orders, and shipment logistics, social media feeds,
consumer buying patterns, product lifecycle operations, and technology-driven data sources like
global positioning systems (GPS), radio frequency based identification (RFID) tracking, mobile
devices, surveillance videos, procurement and others.
Govindan et al (2018), emphasized that recent studies in the field of big data analytics have come
up with tools and techniques to make data-driven supply chain decisions. Analysing and
interpreting results in real time can assist enterprises in making better and faster decisions to
satisfy customer requirements. It will also help organisations to improve their supply chain design
and management by reducing costs and mitigating risks. Recently, various research studies have
indicated the benefits of using big data methods in logistics and supply chain management.
Ghalehkhondabi et al. (2020), reported that more companies are using big data analytics in their
supply chains. About 60% of the research on big data applications in supply chain management
was published after 2017. These publications have increasingly focused on big data applications
in predictive analysis, rather than on the other three types of data analysis: descriptive analysis,
diagnostic analysis, and prescriptive analysis. This view was corroborated by Govindan et al
(2018).
Yu et al. (2017) investigated how a Chinese manufacturing company's financial performance was
impacted by data-driven supply chain capabilities. In this study, a method of data analysis based
on structural equation modeling was suggested. According to the study's findings, an
organization's improved financial performance is favorably correlated with the supply chain
partners' cooperation and the supply chain's response to changes in market demand.
A 2017 research by Basole and Nowak examined the use of tracking devices in the supply chain.
They created an understanding of the elements influencing the assimilation of monitoring
technology on the basis of institutional theory and transaction costs. They used a dataset of 535
supply chain decision-makers and executives to test their approach. They concluded from their
investigation that each of the three assimilation phases is significantly influenced by three
contexts: the product, the supply network, and the environment (initiation, adoption, and
routinization). The findings of this study may help decision-makers who are implementing supply
chain tracking systems.
Big data analytics capabilities in the context of supply chains were the subject of a systematic
literature analysis by Arunachalam et al. (2017), who also created a capability maturity model. In
order to theorize big data analytics capabilities in the context of the supply chain, they analyzed

4
articles from 2008 to 2016 and offered future research and development directions in this area.
The authors looked at 13 maturity models and 82 research papers that were published in the
literature. The findings underlined the necessity of (i) integrating and standardizing diverse data
types in SCM, (ii) understanding big data in supply-chain, and (iii) using a variety of analytics to
incorporate the findings into business operations (Govindan et al., 2018).

Source: Govindan, K., Cheng, T. E., Mishra, N., & Shukla, N. (2018). Big data analytics and application for logistics and
supply chain management. Transportation Research Part E: Logistics and Transportation Review, 114, 343-349.

Logistics management is an efficient system enabling businesses to deliver goods to clients at the
precise location they require at the appropriate time and at the appropriate cost. The planning,
implementation, and control of the flow of raw materials and finished goods and services from
suppliers to customers constitute logistics management. According to Gunasakaran et al (2007),
global competitiveness places more pressure on companies to improve their delivery
performance of products and services to customers. In an effort to improve the quality of delivery
service, companies have outsourced their logistics services, including packing, warehousing
(inventory management), and shipping of goods to customers. Communication plays an
important role in integrating the activities along the logistics value chain. Information
technologies such as electronic data interchange (EDI), the Internet, World Wide Web (WWW,)
and e-commerce have contributed greatly to improving communication with partners in the
logistics chain. E-logistics aids in managing the industry's newest supply chain difficulties. Multi-
channel operations, cross-border tasks, warehouse planning, and inventory, planning,
estimating, and performance management are the core elements of e-logistics (Kanagavalli &
Azeez, 2019).

5
2.1 Impact of Consumer Behaviour and Organisational Structure on Predictive Analytics
Sodero et al. (2019) conducted a study on the exploration of the social processes of Big Data and
Predictive Analytics (BDPA) used for logistics and supply chain management (LSCM), with a focus
on the interactions between technology, human behavior, and organizational context that occur
at the technology’s post-adoption phases in the retail supply chain (RSC) organizations. Results
from their study revealed how user involvement shapes Big Data Predictive Analytics (BDPA) to
fit organizational structures and how changes made to the technology retroactively affect its
design and institutional properties. Findings also reveal previously unreported aspects of BDPA
use for LSCM. These include the presence of temporal and spatial discontinuities in the
technology use across RSC organizations. The study unveils that it is impossible to design a BDPA
technology ready for immediate use. The emergent process framework shows that institutional
and social factors require BDPA use specific to the organization, as the technology comes to reflect
the properties of the organization and the wider social environment for which its designers
originally intended. BDPA is, thus, not easily transferrable among collaborating RSC organizations
and requires managerial attention to the institutional context within which its usage takes place
(Sodero et al., 2019).
Sodero et al. (2019), further posited that the social process through which technology, people
and organizational context interact as their organization uses BDPA. Focusing on the social
process of post-adoption technology use is important for two reasons. First, the technology may
embed different organizational practices and become incompatible with the organizational
context. Thus, organizations need to understand the nature of forces and actors involved in the
social process that shapes BDPA. Second, technologies possess malleable components. Our focus
offers an opportunity for a human-centered investigation of skills required to learn, adjust and
adapt BDPA to achieve a desired level of compatibility with an LSCM context.
According to Awwad et al. (2018), Big Data analytics is the process of using sophisticated analytics
techniques to glean useful information from massive volumes of data, hence enabling data-
driven decision-making. Three different levels of analytics make up big data analytics. Every level
of analytics has a unique job and intended result. Prescriptive Analytics, Predictive Analytics, and
Descriptive Analytics are the three layers of Big Data Analytics that we will be focusing on in this
literature study. Prescriptive analytics currently receives the most attention, followed by
predictive analytics, and descriptive analytics receives the least attention. Diagnostic analytics is
sometimes listed as the fourth form of analytics.
Prescriptive analytics collects application data from existing processes like Cyber-Physical
Systems in the Industry 4.0 trend as well as newly introduced processes like Manufacturing,
Logistics, Transportation, and Warehousing. Procurement, risk analysis, risk management, and
forecasting are just a few areas where predictive analytics is useful. The range of descriptive
analytics is the broadest in terms of how many processes are involved in a system. Descriptive
analytics is used to provide efficient, concise reports on unstructured data that are simple for
people to understand (Awwad et al., 2018).
6
Table 1. categorization of available literature on logistics and Supply Chain Management

Source: Nguyen, T., Li, Z. H. O. U., Spiegler, V., Ieromonachou, P., & Lin, Y. (2018). Big data analytics in supply chain
management: A state-of-the-art literature review. Computers & Operations Research, 98, 254-264.

Figure 2. Classification of sources of data for Logistics and Supply Chain Management

Source: Columbus, L. (2015). Ten Ways Big Data Is Revolutionizing Supply Chain Management.

7
2.2 Social Media Analytics for Improved Logistics and Consumer Satisfaction
With the aid of social media observations, demand forecast updates, and a store with constrained
rationality, Choi (2017) assessed the quick reaction program for fashion. Analytical modeling was
used to discover how important a prompt response is in relation to the likelihood of receiving
favorable product reviews on social media. The impact of such a likelihood is further mitigated
by the fashion retailer's previous perspective on market demand. For the purpose of analyzing
Twitter social media data and finding problems with current food supply chains and logistical
management, Singh et al. (2017) presented a big data analytics technique. They suggested a text
analytics method that uses hierarchical clustering, support vector machines, and multiscale
bootstrap resampling to analyze the content of the Twitter data they had collected. This method
produced a big cluster of phrases that can help decision-makers learn how to improve different
parts of the logistics and food supply chain.
Singh et al. (2016), used a big-data analytics-based approach that considers social media (Twitter)
data for the identification of supply chain management issues in food industries. In particular,
the proposed approach includes text analysis using a support vector machine (SVM) and
hierarchical clustering with multiscale bootstrap resampling. The result of this approach included
a cluster of words that could inform supply-chain (SC) decision-makers about customer feedback
and issues in the flow/quality of food products. A case study in the beef supply chain was analysed
using the proposed approach, where three weeks of data from Twitter were used.
As part of their content analysis, Singh et al. (2016), applied the hierarchical clustering method
individually to the tweets, which are positively and negatively scored. In this research, they
employed hierarchical clustering with p-values via multiscale bootstrap resampling (Suzuki and
Shimodaira, 2006). The clustering method creates hierarchical clusters of words; moreover, it
computes their significance using p-values (obtained after the multiscale bootstrap resampling).
This enables easy identification of significant clusters in the datasets and their hierarchy. The
agglomerative method used was the ward. (Murtagh and Legendre, 2014).

2.3 Big Data Analytics for the Optmisation of E-Procurement Activities

Vaidyanathan, G., & Devaraj, S. (2008) posited that e-Procurement promises to cut operational
costs all across the supply chain, but it also raises the expectations of buyers posing a challenge
for buyer satisfaction and supply chain performance. Using the theoretical lens of Dynamic
Capabilities Theory and Resource-Based View, they postulated that online information and
process act as resources that result in logistics fulfillment capabilities. These capabilities in turn
lead to satisfaction with e-procurement.
According to AlNuaimi et al. (2021), Big data has had a considerable impact on the hotel,
transportation, health, governance, and e-commerce sectors during the past ten years and it is
increasingly thought that it might provide a competitive edge across the supply chain. Big data

8
applications in industries including procurement, transportation, warehouse operations,
marketing, and intelligent logistics can be connected to SCM. Big data can make a noteworthy
contribution in terms of raw material pricing, lead times, environmental concerns, and business
dangers suppliers may encounter in a certain area with the right decision-making and experience,
leading to a noticeable improvement in procurement (Chopra, 2019).
Big Data Analytics (BDA) could aid procurement operations in three crucial decision-making
processes. It starts with risk management. Greater access to historical and current data is made
possible by DBA for procurement functions (Moretto et al., 2017); as a result, they will be better
able to identify issues, monitor supplier behavior, and assess customer satisfaction levels, cash
flow, lead times, and cost (Hickey, 2018). Supplier compliance is the other important area. These
days, suppliers are assessed using a wide range of criteria compatible with their strategic value
(Moretto et al., 2017). BDA enables procurement functions to evaluate each transaction's
performance and compliance by comparing it to the supplier contract it is tied to (Hickey, 2018;
Moretto et al., 2017). Predictive analytics is the third important key area. The use of a significant
amount of structured data for reporting and prediction purposes, as well as a shift toward
predictive rather than reactive procurement processes, would be advantageous (Moretto et al.,
2017). ICT-based procurement systems would therefore be able to anticipate demand for a
particular necessity inside an organization by establishing certain business rules, and then
ensuring that the appropriate resources are sourced based on quality, price, and supply (Hickey,
2018).

2.3.3 Gaps in the Supply Chain E-Procurement Literature

Handfield et al. (2019), predict that the sourcing and supply chain technology environments will
continue to undergo significant changes over the next five years, according to the procurement
analytics landscape outlined in this study. Advanced procurement analytics are currently not
widely used, and problems with data integrity and quality are blocking meaningful analytics
advancements. According to this report, businesses must develop a consistent strategy for
gathering and preserving reliable organizational data that builds on internal sources like spend
analysis and contract databases. In addition, a systematic data governance strategy must take
the place of the current ad hoc methods for acquiring unstructured data. The management of
change and the necessity of fostering an analytical culture are crucial components for companies
in this growth.
While most futuristic studies and papers only predict the broad technological effects of cognitive
analytics and big data, many of them do not offer precise details on functional effects like the
impact on procurement. The analysis of this study gives us a clear understanding of the possible
applications of business analytics and cognitive analytics in procurement processes and helps to
identify related research areas for future investigation. This study also adds to the body of

9
literature and industry studies by offering comprehensive implementation techniques for new
procurement technology.

Theoretical Framework of the Study

The theoretical framework (figure 3) depicts the theoretical approach behind the research
methods that were adopted for the study. E-procurement data were extracted from the
databases of the enterprise resource planner (ERP) for analysis. On the other hand, sensor and
tracking data usually associated with the Internet of Things (IoT) were collected from the IoT
database for analysis. Data (Tweets) from the social media platform Twitter were collected from
the platform using the Application Programming Interface (API). Tweets and Retweets are
typically stored in the cloud or on the Twitter data warehouse. Other sources of data include data
and information elicited from employees and management staff of logistic companies through
the use of structured questionnaires and interviews.
The choice of methods of data analysis adopted for the study is dependent on the theories
espoused for the study. Due to the large quantity of numerical data anticipated from databases
and data warehouses, the use of data mining methods for the scrapping off and extraction of
unstructured data was envisaged, while machine learning algorithms will be used to derive
valuable insights from the said data. Classical statistical methods will be used to analyze the data
derived from structured questionnaires and interviews that were used to elicit information from
employees and management staff of logistics companies.

10
 E-PROCUREMENT
MODULE
INTERNET OF
THINGS [IoT]
ERP IoT
DATABASE DATABASE
ENTERPRISE RESOURCE
PLANNER (ERP)
EMPLOYEES &
MANAGEMEN STAFF

SOCIAL MEDIA

STRUCTURED QUESTIONNAIRE
BIG DATA ANALYTICS
[DATA MINING & MACHINE LEARNING] STRUCTURED INTERVIEW

SOCIAL MEDIA & DESCRIPTIVE STATISTICS

SOCIAL NETWORK ANALYSIS MATHEMATICAL MODELING
TWITTER
DATA WAREHOUSE

RESEARCH METHODS: DATA ANALYSIS

FIGURE 3. THEORETICAL FRAMEWORK

Research Methods
The study will borrow extensively from the epistemological theories of management,
econometrics, mathematical modeling, social network analysis, forecasting, descriptive statistics,
theories of computer-mediated communication, and the applications of digital ethnography to
mention but a relevant few in order to have a deeper and clearer understanding of the subject
matter. Due to the large volume of data envisaged from online sources, databases,
Datawarehouse, social media platforms, and identified blogs interest, data mining techniques
shall be used to extract, collect, wrangle, and cleanse the collected data in preparation for further
analyses. Relevant machine learning algorithms will be applied to the prepared and structured
data for predictive (supervised learning outcomes) and prescriptive (un-supervised learning
outcomes) data analysis and established statistical analysis for inferences.
Data collection: There shall be two primary sources of data for the study. Firstly, relevant data
will be extracted from databases and data warehouses associated with the enterprise resource
planning application associated with the e-procurement module and the transportation module.
Data will also be collected from the Internet of Things (IOT) database for interconnected sensors
and monitoring devices.
Data will also be scrapped off the social media pages and blogs that are of interest to the study,
using hashtags and keywords. Specifically, posts and conversations are scraped off and/or
observed from Twitter and LinkedIn depending on the platform architecture, using Application
Programming Interfaces (APIs) specific to the social media platform. The APIs of Facebook and
Instagram have been closed to the public. On social media platforms where data could not be
scraped and collected due to imposed limitations, digital ethnography shall be applied to study
identified communities of interest following imposed protocols and established ethical guidelines
for internet research.
Secondly, information will be elicited offline from selected individuals such as logistic staff,
workers in the supply chain management, procurements officers, truck drivers and delivery men,
warehouse workers and attendants, etc., through the use of structured interviews and
questionnaires.
Data Analysis: The statistical software application SPSS has highly evolved features for the
analysis of questionnaires and interview transcripts. Information elicited from identified
respondents shall be analysed on the SPSS environment using the relevant modules. The
collected data and methods of analysis shall both be validated to reduce errors to the barest
minimum. The study will apply data mining techniques and machine learning algorithms for the
cleansing, wrangling, and analyses of data scraped from the databases and data warehouses of
the many digital platforms involved in the logistics and supply chain management using modules
from the Python library such as Pandas, SciPy, SciKitlearn, MatPlotLib, TextBlob for doing
sentiment analysis on the collected data, etc.
Predictive Analytics: For supervised learning outcomes, classification algorithms such as the
Naïve Bayes (NB), Support Vector Machines (SVM), Classification & Regression Trees (CART), K
Nearest Neighbour (KNN), and the Neural Networks (NN) have individual characteristics that will
be exploited and applied to the structured data, textual and non-textual datasets for building
dynamic models that can predict possible trends (outcomes. OpenCV for Python shall be used to
analyse images of interest in an unstructured dataset.
Clustering Analysis: The collected data will be subjected to clustering analysis for pattern
recognition and identification in textual and non-textual documents. There exist many machine
learning algorithms for unsupervised learning such as K-Means, K-Nearest Neighbour and
Hierarchical for clustering analysis. Through cluster analysis, we can detect patterns of clusters
in texts, word counts, similarities, and dissimilarities in texts etc.
Evaluation of Research Results
It is important to have a common performance evaluation metric for the supervised and
unsupervised learning outcomes. Metrics such as the classification accuracy (CA), Area Under the
Receiver Operating Characteristic Curve (AUC), Percentage Error, Precision and Recall, will be
accorded cognisance for predictive outcomes for data and text classification. For clustering
analysis, appropriate metrics will be applied for measuring and understanding the clusters found
in research results. Below is a brief description of each performance metric mentioned.
Area Under the ROC Curve (AUC): The Area under the Curve is a metric (usually less than 1.0)
that measures the value of the area under the Receiver Operating Characteristic Curve (ROC).
The AUC is a generalized measure of a classification algorithm’s separation power for two or
more classes. The AUC is considered to be a reliable metric, and hence more acceptable than the
Classification Accuracy.
Classification Accuracy: Classification accuracy (CA) is a metric that measures the performance
of a classification algorithm with respect to its ability to correctly classify a binary or multiclass
response. In other words, classification accuracy is a measure of the proportion of data instances
for which the class prediction was correct.
Precision: Precision can be conceived as a measure of exactness, which is the percentage of
instances labeled as positive that is actually positive. Precision in other words is the fraction or
percentage of detected or retrieved instances that are relevant by the classification algorithm.
Recall: Recall can be conceived as a measure of completeness, which is the percentage of positive
instances that are actually labeled as positive. Recall in other words is the fraction or percentage
of relevant instances that are detected and retrieved by the classification algorithm.
Mean Absolute Error: In regression analysis for numerical output, the MSE is the average of the
difference between the Original Values and the Predicted Values. It gives us the measure of how
far the predictions were from the actual output. However, they don’t give us any idea of the
direction of the error i.e. whether we are under-predicting the data or over-predicting the data
(https://towardsdatascience.com/metrics-to-evaluate-your-machine-learning-algorithm-
f10ba6e38234).
Mean Squared Error (MSE): is quite similar to Mean Absolute Error in regression analysis, the only
difference being that MSE takes the average of the square of the difference between the original
values and the predicted values. The advantage of MSE is that it is easier to compute the gradient,
whereas Mean Absolute Error requires complicated linear programming tools to compute the

13
gradient. As, we take the square of the error, the effect of larger errors become more pronounced
then smaller error, hence the model can now focus more on the larger errors
(https://towardsdatascience.com/metrics-to-evaluate-your-machine-learning-algorithm-
f10ba6e38234).

Table 1.0: Software Requirements

S/N Software Usage Availability
Application
1 Anaconda/Spyder Data mining and machine learning operations Free
2 R-Studio Statistics and Data Analytics Free
3 Gelphi Social Network Analysis Free
4 MongoDB Non-Relational database management Free
5 KNIME Workbench Predictive & Prescriptive data analysis Free
6 MATLAB Simulink Image, audio and Video analysis Subscription
7 Tableau Data Visualisation & Presentation Subscription
8 SPSS Analysis of Questionnaires Subscription
9 RapidMiner Studio Text Mining & Sentiment Analysis Subscription
10 SocNetV Social Network Analysis Subscription

Ethical considerations
Guidelines and ethical considerations put in place by the Association of Internet Researchers
(www.aoir.org) will be followed throughout the study to streamline the scraping of data from
social media. This is important to safeguard the privacy and security of user information.
The social media ethics framework outlined by Tounsend and Wallace
(https://www.gla.ac.uk/media/media_487729_en.pdf) is also instructive for the study.
Ethnographic approaches to digital media would be considered in the study, especially on those
social media where privacy policies make the scraping of data difficult or impossible. In addition,
the ethical guidelines of the post-graduate school in which the research will be conducted shall
be followed to the letter throughout the conduct of the project.

Publication of Results
Results obtained from the study will be published in peer-reviewed journals of international
repute and in scientific conference proceedings in line with the Doctoral Studies rules and
regulations of the University or institution in which the studies are conducted with input from
the thesis/project supervisor and academics involved in the project. Conscious efforts shall be
made to exhibit domain knowledge in the fields of social network analysis, organizational
behaviour, organizational social media use, digital ethnography, and core competence in online
data collection, data mining, machine learning capabilities, and quantitative and qualitative
statistical analysis.

14
Relevance of Proposed Study
The advent of the World Wide Web (WWW), the evolution of digital technology, and
advancements in storage such as the availability of databases, data warehouses and cloud
computing have made it possible to store valuable information in retrieval form.
Online shopping has become an integral part of our lives, especially during the covid-19 lockdown
that cuts across geographical boundaries. In the wake of the said lockdown, consumers had to
rely on the delivery of goods and services to their doorsteps. This involves a lot of logistical
operations from the order (via the World Wide Web), through inventory control, warehousing,
and order picking, to sorting, packaging, shipping, and delivery. While attempts have been made
to automate and monitor every aspect of the listed operations, there are still gaps and
bottlenecks that need to be fixed not-withstanding in the supply chain. Big data analytics has the
potential of providing concrete solutions to these problems through predictive, prescriptive,
descriptive, and diagnostic analysis. With the moderate success achieved through the application
of big data analytics to specific aspects of logistic operations and supply chain management, it
has become pertinent to determine how best e-procurement activities and transportation
scheduling could be best served.

NOTE: Please be informed that this research proposal is by no means definite. It is subject to
changes, improvements, upgrades, and any form of criticism and correction as deemed fit by
the proposed Ph.D. supervisor to suit her/his research interests and research themes of the
faculty/school/institution/research center in which the study is to be conducted. This proposal
amongst others is to show that the author and prospective Ph.D. candidate have the basic
domain knowledge for the proposed topic and research areas.

15
REFERENCES
AlNuaimi, B. K., Khan, M., & Ajmal, M. M. (2021). The role of big data analytics capabilities in
greening e-procurement: A higher order PLS-SEM analysis. Technological Forecasting and Social
Change, 169, 120808.

Arunachalam, D., Kumar, N., Kawalek, J.P., (2017). Understanding big data analytics capabilities
in supply chain management: Unravelling the issues, challenges and
implications for practice. Transp. Res. Part E: Logistics Transp. Rev.
http://dx.doi.org/10.1016/j.tre.2017.04.001.
Awwad, M., Kulkarni, P., Bapna, R., & Marathe, A. (2018, September). Big data analytics in supply
chain: a literature review. In Proceedings of the international conference on industrial
engineering and operations management (Vol. 2018, pp. 418-25)
Basole, R.C., Nowak, M., 2017. Assimilation of tracking technology in the supply chain. Transp.
Res. Part E: Logistics Transp. Rev. http://dx.doi.org/10.1016/j.tre.2016.08.003.

Choi, T.M., (2017). Incorporating social media observations and bounded rationality into fashion
quick response supply chains in the big data era. Transp. Res. Part E:
Logistics Transp. Rev. http://dx.doi.org/10.1016/j.tre.2016.11.006
Columbus, L. (2015). Ten Ways Big Data Is Revolutionizing Supply Chain Management.
Ghalehkhondabi, I., Ahmadi, E., & Maihami, R. (2020). An overview of big data analytics
application in supply chain management published in 2010-2019. Production, 30.
Govindan, K., Cheng, T. E., Mishra, N., & Shukla, N. (2018). Big data analytics and application for
logistics and supply chain management. Transportation Research Part E: Logistics and
Transportation Review, 114, 343-349.
Gunasekaran, A., Ngai, E. W., & Cheng, T. E. (2007). Developing an e-logistics system: a case
study. International Journal of Logistics, 10(4), 333-349.
Graham, D., Manikas, I., & Folinas, D. (Eds.). (2013). E-logistics and e-supply chain management:
Applications for evolving business. IGI Global.
Handfield, R., Jeong, S., & Choi, T. (2019). Emerging procurement technology: data analytics and
cognitive analytics. International journal of physical distribution & logistics management.
Hickey, R., (2018). Big data analytics - The impact on procurement. Retrieved from. https:
//softco.com/blog/big-data-analytics-procurement/.
Kanagavalli, G., & Azeez, R. (2019). Logistics and e-logistics management: benefits and
challenges. International Journal of Recent Technology and Engineering (IJRTE), 8(4), 12804-
12809.

16
 Leveling, J., Edelbrock, M., & Otto, B. (2014, December). Big data analytics for supply chain
management. In 2014 IEEE international conference on industrial engineering and engineering
management (pp. 918-922). IEEE.
Moretto, A., Ronchi, S., Patrucco, A.S., (2017). Increasing the effectiveness of procurement
decisions: The value of big data in the procurement process. International Journal of RF
Technologies: Research & Applications 8 (3), 79–103. https://doi.org/10.3233/ RFT-171670
Murtagh, F., Legendre, P., (2014). Ward’s hierarchical agglomerative clustering method: which
algorithms implement Ward’s criterion? J. Classif. 31 (3), 274–
295
Nguyen, T., Li, Z. H. O. U., Spiegler, V., Ieromonachou, P., & Lin, Y. (2018). Big data analytics in
supply chain management: A state-of-the-art literature review. Computers & Operations
Research, 98, 254-264.
Sarkis, J., Meade, L., & Talluri, S. (2017). E-logistics and the natural environment. In The ecology
of the new economy (pp. 35-51). Routledge.
Singh, A., Shukla, N., Mishra, N., (2017). Social media data analytics to improve supply chain
management in food industries. Transp. Res. Part E: Logistics Transp. Rev.
http://doi.org/10.1016/j.tre.2017.05.008 (in press).
Sodero, A., Jin, Y. H., & Barratt, M. (2019). The social process of Big Data and predictive analytics
use for logistics and supply chain management. International Journal of Physical Distribution &
Logistics Management.
Suzuki, R., Shimodaira, H., (2006). Pvclust: an R package for assessing the uncertainty in
hierarchical clustering. Bioinformatics 22 (12), 1540–1542.
Vaidyanathan, G., & Devaraj, S. (2008). The role of quality in e-procurement performance: An
empirical analysis. Journal of Operations Management, 26(3), 407-425.
Waller, M. A., & Fawcett, S. E. (2013). Data science, predictive analytics, and big data: a revolution
that will transform supply chain design and management. Journal of Business Logistics, 34(2), 77-
84.

17

View publication stats