Volume 1 Issue 1 2 MISDS

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/379893157
THE IMPACT OF MACHINE LEARNING ON PRESCRIPTIVE ANALYTICS FOR

OPTIMIZED BUSINESS DECISION-MAKING
Article · April 2024

DOI: 10.62304/ijmisds.v1i1.112
CITATIONS READS
0 59
1 author:
Ashiqur Rahman
Lamar University
6 PUBLICATIONS 34 CITATIONS
SEE PROFILE
All content following this page was uploaded by Ashiqur Rahman on 17 April 2024.
The user has requested enhancement of the downloaded file.

International Journal of Management Information Systems and Data Science, 2024; 1(1): 7-18
THE IMPACT OF MACHINE LEARNING ON

PRESCRIPTIVE ANALYTICS FOR OPTIMIZED
BUSINESS DECISION-MAKING
Anjuman Ara1; Md Abdul Ahad Maraj2; Md Ashiqur Rahman3; Md Hasanujamman Bari4
1
Management Information Systems, College of Business, Beaumont, Texas, USA.
2
Graduate Researcher, Management Information Systems, Lamar University, Texas, USA.
3
Management Information System, College of Business, Lamar University, Beaumont, Texas, US
4
Graduate Researcher, Management Information Systems, Lamar University, Texas, USA
Keywords ABSTRACT
This study investigates into the integration of Machine Learning

Machine Learning (ML) (ML) with Prescriptive Analytics, showcasing the enhancement
Prescriptive Analytics of decision-making processes in business through this
combination. By analyzing contemporary methodologies and
Decision-Making practical applications, it delves into how ML algorithms
Business Operations significantly improve the precision, efficiency, and forecasting
capabilities of prescriptive analytics. Highlighting case studies
Competitive Advantage across a variety of sectors, the research underscores the
competitive edge businesses can gain by adopting these
sophisticated analytical tools. Moreover, it addresses the array of
technical and organizational hurdles that arise with the
implementation of ML-enhanced prescriptive analytics, such as
challenges in data handling, system integration, and the demand
for specialized skills. Leveraging the latest advancements and
insights from experts, the paper offers a compilation of best
practices and strategic methodologies to effectively overcome
these obstacles. Conclusively, it emphasizes the critical role of
continuous innovation in ML and prescriptive analytics,
encouraging firms to adopt these cutting-edge technologies to
maintain a competitive stance in the fast-evolving, data-centric
business landscape.
1 Introduction facilitating the conversion of unprocessed data into a

potent tool for decision-making. This process is stratified
The ability to distil actionable insights from this data is into a hierarchical structure comprising descriptive,
paramount for organisational success in the contemporary predictive, and prescriptive analytics (Kotsiantis, 2007).
business environment, characterised by abundant data Descriptive analytics summarises past data to understand
(Chawla, 2009). Analytics serves as a pivotal framework, what happened, serving as the foundation for more
International Journal of Management Information Systems and Data 7

Science, Vol 1, Issue 1, April, 2024
advanced analytical techniques (Menon & Rahulnath, data drawn from various organisational sources, allowing
2016). Building upon the insights provided by descriptive businesses to integrate and analyse data across different
analytics, predictive analytics employs statistical models departments (Lepenioti et al., 2020). This shift from
and forecasts to anticipate future events. Prescriptive siloed views to more comprehensive analysis fostered
analytics, the most advanced tier within this hierarchy, better-informed reporting and decision-making. Data
leverages the insights derived from descriptive and warehousing laid the groundwork for the subsequent
predictive analytics. It uses sophisticated algorithms and evolution of analytical practices.
machine learning techniques to suggest actionable
2.3 Advancements in Statistical Modeling (2000s)
strategies and optimise decision-making processes
(Punnoose & Ajit, 2016). This progression from The 2000s witnessed marked progress in statistical
descriptive to prescriptive analytics exemplifies the modelling, paving the path for predictive analytics.
evolving sophistication of data analytics, underscoring its Increased accessibility to affordable computing power
critical role in guiding strategic business decisions in an and the emergence of sophisticated software solutions
increasingly complex and data-driven marketplace enabled businesses to employ various statistical
(Bappy, 2024). Machine learning, a pivotal subset of techniques, including regression analysis, time-series
artificial intelligence, allows systems to learn and evolve forecasting, and classification models (McIver et al.,
from data autonomously without explicit programming 2018). The ability to gain probabilistic insights into
(Bari et al., 2024; Susto et al., 2015). This attribute of ML, potential future outcomes allowed organisations to ask,
to discern intricate patterns within extensive datasets, “What might happen?”. Predictive analytics found
render predictions, and incessantly refine its models, diverse applications ranging from demand forecasting and
renders it an invaluable asset in predictive analytics and inventory management to customer segmentation and risk
bolstering prescriptive decision support. The symbiosis of assessment.
prescriptive analytics and machine learning heralds a new 2.4 Big Data and Early Machine Learning (2010s)
era of unparalleled optimisation and data-driven decision-
making in the corporate sphere (Chawla, 2009). The 2010s were characterised by the explosion of big data
and the increased adoption of nascent machine learning
2 The Evolution of Business Analytics (ML) techniques in analytics. Businesses can now collect
and process massive, diverse datasets generated from
2.1 Early Foundations (pre-1990s) social media, IoT sensors, and online transactions (Mehta
The nascent stages of business analytics emerged in the et al., 2013). ML algorithms started to gain prominence
pre-1990s, where the focus lay squarely on descriptive with their ability to handle complex and unstructured
analytics. Organisations relied on spreadsheets and data. This empowered organisations to uncover hidden
rudimentary reporting software to summarise and patterns, refine predictions, and make better data-driven
visualise historical data, offering essential insights into decisions in fraud detection, personalised marketing, and
past performance (Davenport et al., 2010). Key questions algorithmic trading.
centred around understanding, "What happened?" Data 2.5 Prescriptive Analytics and Mature Machine
storage constraints and limited computational power Learning (2020s - Present
significantly restricted the depth and breadth of analysis
The current decade has witnessed the rise of prescriptive
possible during this period.
analytics and the increasing maturity of machine learning
2.2 Rise of Data Warehousing (1990s) (Ahmed et al., 2024). Optimisation models, simulations,
The 1990s ushered in the era of data warehousing, a and advanced ML algorithms, including deep learning
pivotal development for business analytics. Data and reinforcement learning, now drive actionable
warehouses were centralised hubs for storing structured recommendations, empowering organisations to answer
the central question, "What should we do?". Prescriptive
analytics is pivotal in complex decision-making processes fueled by ongoing research and innovation, promises even
across domains such as resource allocation, supply chain greater sophistication and accuracy in predictions, leading
optimisation, and dynamic pricing (Samuel & Chipunza, to more comprehensive prescriptive insights that will
2009). The continuous evolution of machine learning, reshape business decision-making in the years to come.
Era Focus Key Developments Technologies and Tools Representative
Applications
Pre-1990s Descriptive Analytics ▪ Data storage limitations ▪ Spreadsheets ▪ Basic reporting
▪ Limited computing power ▪ Rudimentary ▪ Summarizing past
reporting software performance
1990s Rise of Data Warehousing ▪ Centralized data storage ▪ Data warehouses ▪ Improved reporting
▪ Integration of data across ▪ Data integration ▪ Cross-departmental
departments tools analysis
2000s Predictive Analytics ▪ Growth of statistical ▪ Statistical software ▪ Demand
modelling ▪ Forecasting and forecasting
▪ Increased accessibility to modeling tools ▪ Risk assessment
computing power ▪ Customer
segmentation
2010s Big Data and Early ▪ Explosion of data ▪ Big data platforms ▪ Fraud detection
Machine Learning ▪ Unstructured data from (Hadoop, etc.) ▪ Personalized
diverse sources ▪ Early ML libraries marketing
▪ Early adoption of ML and tools* ▪ Algorithmic trading
algorithms
2020s - Prescriptive Analytics & ▪ Action-oriented ▪ Prescriptive ▪ Resource allocation
Present Mature Machine Learning recommendations analytics platforms ▪ Supply chain
▪ Advanced ML (deep ▪ Cloud-based ML optimisation
learning, etc.) services ▪ Dynamic pricing
▪ Optimization and ▪ Optimization
simulation software
Table 1: The evolution of business analytics
into meaningful categories, and detect anomalies or
outliers that could signify potential issues. In today's
3 The Rise of Machine Learning
rapidly evolving and hypercompetitive business
Machine learning (ML) has become a transformative environments, organisations that rely solely on intuition
force within the modern business landscape. This branch or outdated decision-making strategies risk falling behind
of artificial intelligence enables computer systems to data-driven competitors. A data-centric approach has
learn from data, identify patterns, and improve their become essential for sustained success (Strohmeier &
performance over time without explicit programming Piazza, 2013). By harnessing the insights from rigorous
(Skowronski, 2019). The remarkable growth of ML can data analysis, businesses can move away from guesswork,
be attributed to several factors, including the increasing making informed decisions aligned with market realities
availability of computational power needed to process and trends.
complex models, vast troves of data to train algorithms, Optimisation sits at the heart of effective business
and ongoing advancements in algorithm design. At the operations. Optimisation is identifying the best possible
core of machine learning's effectiveness lies its superior solution or course of action to achieve a desired outcome
ability to process and extract insights from vast amounts while respecting constraints or limitations. Within
of complex data. ML algorithms recognise patterns that business analytics, optimisation techniques are central to
might be too subtle or intricate for traditional analytical prescriptive analytics models. By mathematically
methods to detect (Wang & Shun, 2016). This capability formulating business problems, optimisation tools are
is fundamental to predictive analytics. ML models learn used to find solutions that maximise desirable outcomes,
from historical data to forecast future trends, classify data like revenue or customer satisfaction, while minimising
negative factors such as costs or delays (Susto et al., businesses to weigh different strategies and comprehend
2015). Machine learning and optimisation work the implications of their choices prior to executing real-
synergistically to drive significant competitive world decisions (Lin & Tseng, 2004; McIver et al., 2018;
advantages. For example, consider a manufacturing Rocchetta et al., 2019).
operation relying on just-in-time inventory management.
The synergy between descriptive, predictive, and
ML-powered demand forecasting can help optimise
prescriptive analytics forms a cohesive analytical
inventory levels by accurately predicting future demand,
continuum, each stage contributing uniquely to the
ensuring sufficient materials are on hand to meet
understanding and shaping business trajectories (Menezes
production requirements without excessive storage costs
et al., 2019). Descriptive analytics lays the groundwork
from overstocking. Similarly, within distribution
by providing a retrospective glance at historical data,
networks, optimisation models can leverage ML-
setting the stage for predictive analytics to employ
generated insights to identify the most efficient delivery
statistical modelling and machine learning in forecasting
routes, reducing transportation expenses and ensuring the
future trends. The insights garnered from these stages are
timely fulfilment of customer orders (Tursunbayeva et al.,
pivotal, feeding into the prescriptive analytics phase and
2018).
synthesising this accumulated knowledge to strategise
4 Prescriptive Analytics actionable paths forward. Such a hierarchical structure
ensures a comprehensive data analysis, from
Prescriptive analytics represents the zenith of data-driven understanding past occurrences to devising plans that
decision-making processes within the spectrum of influence future outcomes (Lepenioti et al., 2020). By
business analytics, transcending the foundational layers leveraging the insights from its analytical predecessors,
of descriptive and predictive analytics. Unlike its prescriptive analytics establishes a strategic framework
precursors, which focus respectively on articulating past that anticipates what might happen and advises on the
events and forecasting future possibilities, prescriptive steps an organisation should take to steer towards its
analytics employs advanced methodologies to furnish objectives (Harding et al., 2005; Kotsiantis, 2007). As the
actionable guidance to achieve optimal business field of analytics progresses, the integral role of
outcomes (Menezes et al., 2019). This realm of analytics prescriptive analytics in facilitating informed decision-
tackles complex questions, ranging from determining the making continues to gain prominence. The integration of
most effective pricing strategies for sales maximisation to descriptive and predictive analytics provides a robust
designing resilient supply chains for timely deliveries, foundation for prescriptive analytics to thrive, enhancing
facilitating a proactive approach to sculpting an the accuracy and relevance of its recommendations. With
organisation's future. The application of prescriptive the advancements in machine learning algorithms and
analytics is grounded in using sophisticated mathematical optimisation techniques, the capabilities of prescriptive
models and optimisation algorithms (Rocchetta et al., analytics are expanding, making it an increasingly vital
2019). These models meticulously encapsulate the component in strategic decision-making arsenals across
intricacies of business scenarios by accounting for various industries. This evolution underscores the
objectives, constraints, and the multitude of variables at potential of prescriptive analytics in mapping out
play, providing a structured framework for decision- immediate steps and charting a course for long-term
making. Optimisation algorithms further enrich this strategic direction (Janitza et al., 2016). As such,
analytical domain by methodically sifting through prescriptive analytics emerges as a critical tool in the data-
potential solutions to pinpoint the best action that adheres driven business environment, capable of harnessing past
to the predefined models. In tandem, simulation insights and future projections to craft actionable
techniques enable examining diverse decision outcomes recommendations that drive success.
under varied conditions, offering a sandbox for "what-if"
analyses. This predictive exploration empowers
5 Prescriptive Analytics and Decision-Making prescriptive analytics significantly. Particularly,

reinforcement learning algorithms, which refine their
The future of decision-making in business analytics is decision-making capabilities through trial and error, are
increasingly intertwined with the evolution of machine emerging as a cornerstone for optimising decisions within
learning (ML) technologies, where advances in deep dynamic and complex environments characterised by
learning, reinforcement learning, and natural language constantly changing variables. This progression in ML
processing are poised to bolster the capabilities of technologies is anticipated to result
Figure 1: Comparative aspects of prescriptive analytics, machine learning advancements
in enhanced accuracy of predictions and more nuanced industries; for instance, manufacturing could streamline
recommendations within the sphere of prescriptive production schedules, minimise inventory waste, and
analytics, thereby enabling more sophisticated and bolster supply chain resilience, while in healthcare, it
dynamic decision-making frameworks (Li et al., 2019). could enhance patient care through optimised treatment
Concurrently, the trend towards democratising plans and efficient scheduling. Retail sectors could also
prescriptive analytics tools is set to lower the barriers to see the adoption of dynamic pricing strategies tailored to
entry for businesses seeking to leverage this advanced real-time market dynamics and consumer behaviours,
analytical capability. The ongoing development of more thereby optimising revenue streams and enhancing
accessible platforms, including cloud-based services and customer satisfaction. Beyond operational efficiencies,
user-friendly analytical tools, is expected to expand the the strategic application of
reach of prescriptive analytics, making it feasible for
prescriptive models in risk management and strategic
businesses of varying sizes to adopt these sophisticated
planning allows businesses to simulate various decision
models without necessitating substantial in-house
scenarios, enabling a more informed and proactive
technical expertise or significant investments in
approach to navigating the complexities of the modern
computational infrastructure (Liu et al., 2015)
business landscape (Minbaeva, 2017). This paradigm
As businesses across many sectors witness the tangible shift underscores a broader movement towards a future
benefits of data-driven decision-making facilitated by where organisations are not merely reactive but are
prescriptive analytics, a cultural shift is anticipated empowered to sculpt their trajectories through informed
towards embracing data-informed strategies. This shift data-driven strategies proactively.
will likely embed data literacy and analytical
competencies as pivotal attributes across all
6 Finance and Risk Management
organisational functions and leadership roles, fostering an
environment where data-driven insights are integral to Machine learning is reshaping financial operations and
strategic decision-making processes (Maynard et al., risk management, bringing greater efficiency, accuracy,
2014). The application of prescriptive analytics is and proactive insight to these functions. In fraud
expected to yield transformative impacts across diverse detection, ML analyses vast datasets of financial
transactions, identifying subtle patterns, anomalies, and et al., 2015). Within risk assessment, ML is applied to
deviations from usual behaviour that might indicate loan underwriting, analysing borrowers' financial history,
fraudulent activity. These sophisticated models surpass credit scores, and numerous other data points to assess
the capabilities of traditional rule-based systems, their creditworthiness and the likelihood of default. This
reducing financial losses and protecting consumers (Susto
Figure 2: Comparative roles of ML in different sectors
advanced analytical approach informs better lending ads. This precision-driven approach leads to more
decisions, aiding risk management (Tozer et al., 2017). effective ad placements, increased conversion rates, and a
Furthermore, ML-powered portfolio optimisation offers better return on advertising investment (Kotsiantis, 2007;
dynamic asset allocation and investment strategy Lee et al., 2013).
solutions. ML algorithms can recommend portfolio
adjustments by continually analysing market trends and
identifying patterns in large datasets, aiming to maximise 7 Machine Learning's Role in Enhancing
returns while managing risk in a constantly changing Prescriptive Analytics
financial landscape (Valle et al., 2012).
Machine learning has become a cornerstone of Machine learning (ML) has emerged as a pivotal
personalised marketing and creating superior customer enhancement to prescriptive analytics, offering a broad
experiences. Personalised recommendation engines, spectrum of techniques that significantly augment its
ubiquitous on e-commerce platforms, leverage ML efficacy. Within this symbiotic relationship, supervised
models that learn from past purchases, browsing history, learning techniques such as regression models, decision
and user preferences to suggest products or content trees, and neural networks play a crucial role by
aligned with individual interests. This leads to a highly leveraging labelled data to predict future trends, thus
tailored customer experience, increasing engagement and providing a solid foundation for prescriptive models to
the likelihood of purchase (Vater et al., 2019). Dynamic build upon (Kirimi & Moturi, 2016). Similarly,
pricing is another domain where ML flourishes. ML- unsupervised learning algorithms excel in detecting
powered optimisation models analyse real-time data on patterns and anomalies in unlabeled data, which is
market demand, competitor actions, inventory essential for segmentation and anomaly detection tasks
availability, and customer behaviour to suggest optimal that contribute to a more nuanced prescriptive analysis.
pricing strategies. The ability to adjust prices based on Reinforcement learning, characterised by its trial-and-
these factors dynamically empowers businesses to error learning process, shows promise in optimising
maximise revenue and profitability (Wang & Shun, decisions for complex problems within prescriptive
2016). ML also enhances the effectiveness of targeted models, further broadening the scope of prescriptive
advertising campaigns. By analysing user demographics, analytics applications (Lepenioti et al., 2020). This
online behaviour, and search trends, ML models predict integration of ML into prescriptive analytics is not merely
which individuals are most likely to respond to specific theoretical but is evidenced through real-world
applications across various industries. For example, models that recommend personalised care interventions to
manufacturers utilise ML-based demand forecasting to prevent readmissions. Furthermore, the financial industry
inform prescriptive models for optimal production benefits from ML-powered fraud detection systems that,
scheduling, thereby minimising inventory waste and coupled with optimisation models, help make informed
ensuring timely product delivery. In the healthcare sector, decisions regarding suspicious transactions and
ML algorithms that analyse patient records to identify implement verification measures to protect consumers
high-risk patients enable the development of prescriptive (Levenson & Fink, 2017).
Figure 3: The architecture of the proposed approach
They are integrating ML with prescriptive analytics, planning, and risk management. For instance, ML-
which yields numerous benefits, notably enhancing powered supply chain models can assess the impact of
predictive accuracy and enabling more reliable and unforeseen events like supplier delays or port congestion
actionable recommendations. ML algorithms are adept at under various scenarios, providing businesses with the
processing complex datasets, uncovering non-linear insights needed to simulate and implement optimal
relationships, and adapting through continuous learning. responses (McIver et al., 2018). Additionally, the
This capability leads to more accurate forecasts automation and scalability afforded by ML algorithms
concerning market dynamics, customer behaviours, and streamline the decision-making process, enabling the
potential risks, thereby enhancing the reliability of analysis of vast datasets and the generation of complex
prescriptive recommendations (Li et al., 2019). Moreover, recommendations with unprecedented speed and
ML facilitates the exploration of various potential efficiency, especially in data-intensive environments.
outcomes by simulating "what-if" scenarios, thereby This automation not only enhances operational efficiency
allowing businesses to evaluate the implications of but also scales decision-making capabilities across
different strategies before their implementation. This multiple business functions, underscoring the
data-driven experimentation and trade-off evaluation transformative potential of ML in bolstering prescriptive
empower prescriptive models to optimise decisions in analytics (Menon & Rahulnath, 2016).
critical areas such as resource allocation, supply chain

8 Method significant strategic benefits, encapsulated in five primary

themes: Data Integration and Quality, Technological
The study employed a mixed-methods approach, Infrastructure, Organizational Culture, Talent
beginning with a comprehensive literature review to map Acquisition, and Strategic Outcomes. The critical need for
the evolution of machine learning (ML) and its integration high-quality, integrated data emerges as a fundamental
with prescriptive analytics in the context of business challenge, with inaccuracies and inconsistencies directly
decision-making. This phase identified key trends, undermining ML's potential to deliver reliable insights.
technological advancements, and pivotal moments in the This challenge is compounded by the difficulties of
history of analytics through an exhaustive examination of melding ML technologies with existing, often outdated,
academic journals, industry reports, and case studies. technological infrastructures, necessitating substantial
Following the literature review, a series of semi- investments in IT modernization. Moreover, the transition
structured interviews were conducted with industry towards ML-driven analytics is frequently hindered by
experts and academicians specializing in ML, prescriptive organizational resistance to change and a pervasive lack
analytics, and business strategy. These interviews of data literacy, highlighting the imperative for a cultural
provided invaluable insights into the practical shift towards embracing data-driven innovation. The
applications, challenges, and future directions of ML- competitive market for skilled ML professionals further
driven prescriptive analytics across various sectors. exacerbates the challenges, underscoring the urgency for
Additionally, quantitative data from several businesses initiatives aimed at cultivating a skilled workforce.
that have implemented ML in their prescriptive analytics Despite these obstacles, the strategic benefits of
processes were gathered and analyzed. This data included integrating ML into prescriptive analytics shine through,
performance metrics before and after the implementation offering enhanced decision-making capabilities,
of ML algorithms, enabling the quantitative assessment of operational efficiencies, and the agility to adapt to rapidly
the impact of ML on business decision-making efficacy changing market conditions. These benefits collectively
and efficiency. underscore the transformative potential of ML-enhanced
prescriptive analytics in securing competitive advantages
9 Findings
and driving business success, provided organizations can
navigate the intricate web of challenges inherent in this
The integration of Machine Learning (ML) into
technological integration.
prescriptive analytics, as gleaned from interviews with
industry experts and academicians, unfolds a nuanced .
narrative characterized by both formidable challenges and
Theme Challenges Opportunities
Data Integration and ▪ Inaccuracies and inconsistencies in data ▪ Implement robust data governance
Quality ▪ Lack of real-time data integration frameworks
▪ Utilize advanced data processing
technologies
Technological ▪ Legacy systems incompatible with ML ▪ Modernize IT infrastructure to support
Infrastructure complexities ML
▪ Difficulties in integrating advanced ▪ Invest in scalable and flexible
analytics technology solutions
Organizational Culture ▪ Resistance to change and innovation ▪ Foster a data-driven culture
▪ Low data literacy among staff ▪ Emphasize education and training in
data literacy
Talent Acquisition ▪ Shortage of skilled ML and analytics ▪ Develop strategic education and
professionals training programs
▪ Competitive market for talent ▪ Focus on workforce development and
retention strategies
Strategic Outcomes ▪ Initial high investment costs and ▪ Enhanced decision-making capabilities
implementation challenges ▪ Operational efficiencies
▪ Personalized customer experiences
▪ Market agility
Table 2: Summary of the findings

enhanced prescriptive analytics in the modern business
environment (Fan et al., 2012).
10 Discussion
In the context of deploying machine learning (ML)-
The integration of Machine Learning (ML) into enhanced prescriptive analytics, this study has identified
prescriptive analytics represents a pivotal evolution in a constellation of technical and organizational challenges
business analytics, marked by a dual narrative of that significantly impact the success of such initiatives.
significant challenges and substantial strategic benefits Central to these challenges is the establishment of a robust
(Chawla, 2009). This duality, characterized by the themes data infrastructure, which encompasses the processes of
of Data Integration and Quality, Technological data collection, storage, and cleaning. This foundational
Infrastructure, Organizational Culture, Talent requirement not only demands substantial investment but
Acquisition, and Strategic Outcomes, encapsulates the also specialized expertise, underscoring the pivotal role of
complex journey organizations undertake in harnessing data integrity in the effectiveness of prescriptive analytics
the power of ML-driven analytics. The fundamental (Jantan, 2009). The study further illuminates the
challenge lies in the critical need for high-quality, complexities involved in integrating prescriptive
integrated data, where inaccuracies and inconsistencies analytics into existing organizational systems and
can severely undermine ML's capability to deliver reliable workflows, where the fidelity of the underlying data is
insights, necessitating a paradigm shift towards robust critical.
data governance and advanced processing technologies
(Choudhary et al., 2008). Concurrently, the integration of Moreover, the ethical dimensions of deploying advanced
ML technologies with existing infrastructures poses analytics systems, particularly in terms of data privacy
significant hurdles, often requiring extensive IT and the security of sensitive information, are highlighted
modernization efforts to accommodate the complex as paramount concerns (Tursunbayeva et al., 2018). The
demands of advanced analytics (Gröger, 2018). discussion also delves into the challenge of the opacity of
Additionally, the cultural and workforce-related barriers, ML algorithms, which poses barriers to their broader
including resistance to change and a pervasive lack of data acceptance due to the difficulty in understanding their
literacy, highlight the urgent need for organizations to decision-making processes. The emergence of the field of
cultivate a data-driven culture and invest in education and explainable AI (XAI) is discussed as a significant effort
workforce development to bridge the talent gap in ML and to address this challenge, aiming to demystify the
analytics (Henke et al., 2016). Despite these challenges, workings of complex ML models and thereby foster trust
the strategic benefits of ML integration into prescriptive and accountability in automated decision-making
analytics—ranging from enhanced decision-making and processes (Li et al., 2019; Menezes et al., 2019). The
operational efficiencies to market agility—underscore the narrative extends to the crucial aspect of skilled
transformative potential of this technological synergy in personnel, emphasizing the increasing demand for
securing competitive advantages and driving business professionals with a hybrid of technical skills and
success (Jantan, 2009). As organizations navigate this business acumen, such as data scientists, machine
intricate landscape, the balance between overcoming the learning engineers, and business analysts (Lee et al.,
inherent challenges and leveraging the opportunities 2013). This demand highlights a pressing skill gap that
becomes critical in realizing the full potential of ML- organizations must bridge to harness the full potential of
ML-enhanced prescriptive analytics effectively. The navigate complex decision-making landscapes with
study concludes that addressing these multifaceted greater precision and foresight. This integration promises
challenges is essential for organizations seeking to a competitive edge and heralds a new era of strategic
leverage prescriptive analytics for competitive advantage planning and operational efficiency. As the field
in the increasingly data-driven business environment continues to evolve, embracing these technologies
(Kirimi & Moturi, 2016; Lepenioti et al., 2020). becomes essential for businesses aiming to remain at the
forefront of innovation. The challenges associated with
11 Recommendations
implementing such sophisticated analytics systems—
The landscape of prescriptive analytics is poised for from data infrastructure to talent acquisition—underscore
significant transformation, driven by integrating the importance of a strategic approach to technology
emerging technologies such as artificial intelligence (AI), adoption. However, the opportunities to harness the
the Internet of Things (IoT), and big data, which power of ML-enhanced prescriptive analytics far
collectively promise to elevate the scope and precision of outweigh these hurdles. Thus, businesses are called upon
data-driven decision-making. Future trends suggest that to invest in developing their ML capabilities, ensuring
machine learning will play an increasingly central role in they are well-positioned to make informed, data-driven
business analytics, offering unprecedented opportunities decisions that can propel their operations to new heights
for innovation and securing competitive advantage by of success in the dynamic landscape of intelligent
deploying advanced analytics strategies for businesses analytics.
aiming to harness the full potential of machine learning .
within their analytics frameworks; adopting best practices
is crucial. This includes establishing robust data References:
infrastructure, proactively engaging in talent acquisition Ahmed, H., Al Bashar, M., Taher, M. A., & Rahman, M.
to bridge the skills gap, and forging strategic technology A. (2024). Innovative Approaches To Sustainable
Supply Chain Management In The
partnerships that can enhance analytical capabilities.
Manufacturing Industry: A Systematic Literature
Furthermore, successfully implementing machine Review. Global Mainstream Journal of
learning-driven prescriptive analytics necessitates a Innovation, Engineering & Emerging
holistic approach, focusing on technological and Technology, 3(02), 01-13.
organisational dimensions to ensure seamless integration Bappy, M. (2024). Exploring the Integration of Informed
into existing systems and workflows. A critical factor in Machine Learning in Engineering Applications:
A Comprehensive Review. American Journal of
realising the value of these analytics investments lies in
Science and Learning for Development, 3(2), 11-
the ability to measure their impact on business outcomes 21.
effectively, affirming the strategic importance of Bari, M. H., Arif, N. U. M., Hasan, M. M., & Maraj, M.
advanced analytics in driving organisational success and A. A. (2024). Comparative Analysis Of Digital
innovation in the digital era. Payment Platforms And E-Commerce Giants: A
Five-Year Performance And Strategic
12 Conclusion Development Study Of Visa, Mastercard,
Amazon, And Ebay. Global Mainstream Journal
of Innovation, Engineering & Emerging
The symbiotic relationship between machine learning Technology, 3(01), 01-10.
(ML) and prescriptive analytics represents a pivotal shift Chawla, N. V. (2009). The Data Mining and Knowledge
in business analytics, offering a transformative potential Discovery Handbook - Data Mining for
beyond traditional decision-making processes. By Imbalanced Datasets: An Overview (Vol. NA).
leveraging ML's advanced capabilities, prescriptive https://doi.org/10.1007/978-0-387-09823-4_45
Choudhary, A. K., Harding, J. A., & Tiwari, M. K. (2008).
analytics can provide businesses with unprecedented
Data mining in manufacturing: a review based on
insights and recommendations, thus enabling them to the kind of knowledge. Journal of Intelligent
Manufacturing, 20(5), 501-521. Levenson, A., & Fink, A. A. (2017). Human capital
https://doi.org/10.1007/s10845-008-0145-x analytics: too much data and analysis, not enough
Davenport, T. H., Harris, J. G., & Shapiro, J. (2010). models and business insights. Journal of
Competing on talent analytics. Harvard business Organizational Effectiveness: People and
review, 88(10), 52-58, 150. https://doi.org/NA Performance, 4(2), 145-156.
Fan, C.-Y., Fan, P.-S., Chan, T.-Y., & Chang, S.-H. (2012). https://doi.org/10.1108/joepp-03-2017-0029
Using hybrid data mining and machine learning Li, G., Gomez, R., Nakamura, K., & He, B. (2019).
clustering analysis to predict the turnover rate for Human-Centered Reinforcement Learning: A
technology professionals. Expert Systems with Survey. IEEE Transactions on Human-Machine
Applications, 39(10), 8844-8851. Systems, 49(4), 337-349.
https://doi.org/10.1016/j.eswa.2012.02.005 https://doi.org/10.1109/thms.2019.2912447
Gröger, C. (2018). Building an Industry 4.0 Analytics Lin, C.-C., & Tseng, H.-Y. (2004). A neural network
Platform. Datenbank-Spektrum, 18(1), 5-14. application for reliability modelling and
https://doi.org/10.1007/s13222-018-0273-1 condition-based predictive maintenance. The
Harding, J. A., Shahbaz, M., Srinivas, N. A., & Kusiak, A. International Journal of Advanced
(2005). Data Mining in Manufacturing: A Manufacturing Technology, 25(1), 174-179.
Review. Journal of Manufacturing Science and https://doi.org/10.1007/s00170-003-1835-3
Engineering, 128(4), 969-976. Liu, C., Xu, X., & Hu, D. (2015). Multiobjective
https://doi.org/10.1115/1.2194554 Reinforcement Learning: A Comprehensive
Henke, N., Bughin, J., Chui, M., Manyika, J., Saleh, T., & Overview. IEEE Transactions on Systems, Man,
Wiseman, B. (2016). The age of analytics: and Cybernetics: Systems, 45(3), 385-398.
competing in a data-driven world. NA, NA(NA), https://doi.org/10.1109/tsmc.2014.2358639
NA-NA. https://doi.org/NA Maynard, D. C., Brondolo, E. M., Connelly, C. E., &
Janitza, S., Celik, E., & Boulesteix, A.-L. (2016). A Sauer, C. E. (2014). I'm Too Good for This Job:
computationally fast variable importance test for Narcissism's Role in the Experience of
random forests for high-dimensional data. Overqualification. Applied Psychology, 64(1),
Advances in Data Analysis and Classification, 208-232. https://doi.org/10.1111/apps.12031
12(4), 885-915. https://doi.org/10.1007/s11634- McIver, D., Lengnick-Hall, M. L., & Lengnick-Hall, C.
016-0276-4 A. (2018). A strategic approach to workforce
Jantan, H. (2009). Towards Applying Data Mining analytics: Integrating science and agility.
Techniques for Talent Management. NA, Business Horizons, 61(3), 397-407.
NA(NA), NA-NA. https://doi.org/NA https://doi.org/10.1016/j.bushor.2018.01.005
Kirimi, J. M., & Moturi, C. A. (2016). Application of Data Mehta, S., Pimplikar, R., Singh, A., Varshney, L. R., &
Mining Classification in Employee Performance Visweswariah, K. (2013). EDBT - Efficient
Prediction. International Journal of Computer multifaceted screening of job applicants.
Applications, 146(7), 28-35. Proceedings of the 16th International Conference
https://doi.org/10.5120/ijca2016910883 on Extending Database Technology, NA(NA),
Kotsiantis, S. (2007). Supervised Machine Learning: A 661-671.
Review of Classification Techniques. Informatica https://doi.org/10.1145/2452376.2452453
(lithuanian Academy of Sciences), 31(3), 249- Menezes, B. C., Kelly, J. D., Leal, A. G., & Le Roux, G.
268. https://doi.org/NA A. C. (2019). Predictive, Prescriptive and
Lee, J., Lapira, E., Bagheri, B., & Kao, H.-A. (2013). Detective Analytics for Smart Manufacturing in
Recent advances and trends in predictive the Information Age. IFAC-PapersOnLine, 52(1),
manufacturing systems in big data environment. 568-573.
Manufacturing Letters, 1(1), 38-41. https://doi.org/10.1016/j.ifacol.2019.06.123
https://doi.org/10.1016/j.mfglet.2013.09.005 Menon, V. M., & Rahulnath, H. A. (2016). A novel
Lepenioti, K., Bousdekis, A., Apostolou, D., & Mentzas, approach to evaluate and rank candidates in a
G. (2020). Prescriptive analytics: Literature recruitment process by estimating emotional
review and research challenges. International intelligence through social media data. 2016
Journal of Information Management, 50(NA), International Conference on Next Generation
57-70. Intelligent Systems (ICNGIS), NA(NA), 1-6.
https://doi.org/10.1016/j.ijinfomgt.2019.04.003 https://doi.org/10.1109/icngis.2016.7854061
Minbaeva, D. (2017). Human capital analytics: why aren't Applications, 39(11), 9939-9945.
we there? Introduction to the special issue. https://doi.org/10.1016/j.eswa.2011.11.126
Journal of Organizational Effectiveness: People Vater, J., Harscheidt, L., & Knoll, A. (2019). Smart
and Performance, 4(2), 110-118. Manufacturing with Prescriptive Analytics. 2019
https://doi.org/10.1108/joepp-04-2017-0035 8th International Conference on Industrial
Punnoose, R., & Ajit, P. (2016). Prediction of Employee Technology and Management (ICITM), NA(NA),
Turnover in Organizations using Machine 224-228.
Learning Algorithms. International Journal of https://doi.org/10.1109/icitm.2019.8710673
Advanced Research in Artificial Intelligence, Wang, K.-Y., & Shun, H.-Y. (2016). Applying Back
5(9), NA-NA. Propagation Neural Networks in the Prediction of
https://doi.org/10.14569/ijarai.2016.050904 Management Associate Work Retention for Small
Rocchetta, R., Bellani, L., Compare, M., Zio, E., & and Medium Enterprises. Universal Journal of
Patelli, E. (2019). A reinforcement learning Management, 4(5), 223-227.
framework for optimal operation and https://doi.org/10.13189/ujm.2016.040501
maintenance of power grids. Applied Energy,
241(NA), 291-301.
https://doi.org/10.1016/j.apenergy.2019.03.027
Samuel, M. O., & Chipunza, C. (2009). Employee
retention and turnover: Using motivational
variables as a panacea. African Journal of
Business Management, 3(9), 410-415.
https://doi.org/NA
Skowronski, M. (2019). Overqualified Employees: A
Review, A Research Agenda, and
Recommendations for Practice. Journal of
Applied Business and Economics, 21(1), NA-NA.
https://doi.org/10.33423/jabe.v21i1.661
Strohmeier, S., & Piazza, F. (2013). Domain driven data
mining in human resource management: A review
of current research. Expert Systems with
Applications, 40(7), 2410-2420.
https://doi.org/10.1016/j.eswa.2012.10.059
Susto, G. A., Schirru, A., Pampuri, S., McLoone, S., &
Beghi, A. (2015). Machine Learning for
Predictive Maintenance: A Multiple Classifier
Approach. IEEE Transactions on Industrial
Informatics, 11(3), 812-820.
https://doi.org/10.1109/tii.2014.2349359
Tozer, B., Mazzuchi, T. A., & Sarkani, S. (2017). Many-
objective stochastic path finding using
reinforcement learning. Expert Systems with
Applications, 72(72), 371-382.
https://doi.org/10.1016/j.eswa.2016.10.045
Tursunbayeva, A., Di Lauro, S., & Pagliari, C. (2018).
People analytics—A scoping review of
conceptual boundaries and value propositions.
International Journal of Information
Management, 43(NA), 224-247.
https://doi.org/10.1016/j.ijinfomgt.2018.08.002
Valle, M. A., Varas, S., & Ruz, G. A. (2012). Job
performance prediction in a call center using a
naive Bayes classifier. Expert Systems with
View publication stats

Volume 1 Issue 1 2 MISDS

Uploaded by

Copyright:

Available Formats

Volume 1 Issue 1 2 MISDS

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Volume 1 Issue 1 2 MISDS

Uploaded by

Copyright:

Available Formats

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

THE IMPACT OF MACHINE LEARNING ON PRESCRIPTIVE ANALYTICS FOR

Article · April 2024

The user has requested enhancement of the downloaded file.

THE IMPACT OF MACHINE LEARNING ON

This study investigates into the integration of Machine Learning

1 Introduction facilitating the conversion of unprocessed data into a

International Journal of Management Information Systems and Data 7

5 Prescriptive Analytics and Decision-Making prescriptive analytics significantly. Particularly,

Figure 1: Comparative aspects of prescriptive analytics, machine learning advancements

Figure 2: Comparative roles of ML in different sectors

Figure 3: The architecture of the proposed approach

International Journal of Management Information Systems and Data 13

8 Method significant strategic benefits, encapsulated in five primary

Table 2: Summary of the findings

View publication stats

You might also like