Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24FEB1099

Implementation of ABC - XYZ Analysis on Business

Inventory Management using Machine Learning
Model Technology
Hakizimana Alain1; Dr. Wilson Musoni2 (PhD)
Master of Science with honors in Information Technology (MSCIT),
University of Kigali, Rwanda

Abstract:- This paper presents a novel approach to In this paper, we propose a novel approach that
optimize business inventory management through the combines ABC-XYZ analysis with machine learning
integration of ABC-XYZ analysis with advanced algorithms to optimize business inventory management. We
machine learning models. Inventory management plays a first provide an overview of ABC-XYZ analysis and its
critical role in the operational efficiency and profitability limitations in addressing the complexities of modern supply
of businesses across various industries. Traditional chains. Next, we introduce the concept of machine learning-
methods such as ABC analysis and XYZ analysis have based inventory management and discuss its potential
been widely used to classify inventory items based on benefits. We then present our integrated framework and
their importance and demand variability. However, the demonstrate its effectiveness through a case study conducted
effectiveness of these methods can be further enhanced on a real-world business dataset. Finally, we discuss the
by leveraging machine learning techniques to analyze implications of our findings and provide recommendations
historical data and make accurate predictions. In this for future research in this area.
study, we propose a framework that combines ABC-XYZ
analysis with machine learning algorithms to classify II. METODOLOGY
inventory items and optimize inventory control policies.
We demonstrate the effectiveness of our approach  Data Collection Methods and Instruments/Tools
through a case study conducted on a real-world business A research methodology describes the techniques and
dataset, highlighting significant improvements in procedures used to identify and analyze information
inventory turnover, cost reduction, and customer regarding a specific research topic. It is a process by which
satisfaction. researchers design their study so that they can achieve their
objectives using the selected research instruments. It
Keywords:- ABC-XYZ Analysis, Inventory Management, includes all the important aspects of research, including
Machine Learning, Optimization, Business Efficiency. research design, data collection methods, data analysis
methods, and the overall framework within which the
I. INTRODUCTION research is conducted. (Sreekumar, 2023)

Efficient inventory management is crucial for  Data Analysis

businesses to meet customer demands, minimize costs, and Data analysis is a process of inspecting, cleansing,
maximize profitability. Traditionally, businesses have relied transforming, and modeling data with the goal of
on manual methods or basic inventory classification discovering useful information, informing conclusions, and
techniques such as ABC analysis and XYZ analysis to supporting decision-making. Data analysis has multiple
prioritize inventory items based on their importance and facets and approaches, encompassing diverse techniques
demand variability. While these methods provide valuable under a variety of names, and is used in different business,
insights, they often fail to adapt to dynamic market science, and social science domains. In today's business
conditions and may lead to suboptimal inventory control world, data analysis plays a role in making decisions more
decisions. scientific and helping businesses operate more effectively.

Recent advancements in machine learning offer Python programming language was used to analyze
promising opportunities to enhance inventory management secondary data such as sales reports.
by leveraging historical data to predict future demand
patterns more accurately. By integrating machine learning  Research Design
models with traditional inventory classification techniques, This study adopted a research design to investigate the
businesses can gain deeper insights into their inventory implementation of ABC-XYZ analysis on business
profiles and develop more effective inventory control inventory management utilizing machine learning models.
strategies. The research employed a single-method approach,
quantitative analysis of inventory data. The quantitative
aspect involved collecting historical inventory sales data

IJISRT24FEB1099 www.ijisrt.com 1963

Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24FEB1099

from the target business and applying ABC-XYZ analysis to automate and optimize the ABC-XYZ classification process.
classify and forecast inventory items based on their value The research aimed to evaluate the effectiveness of the
and demand variability. Machine learning algorithms such proposed approach in optimizing inventory management
as clustering and classification models were employed to practices and improving overall business performance.

Fig 1 Analysis of Our Framework Models

IJISRT24FEB1099 www.ijisrt.com 1964

Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24FEB1099

III. CONCEPTUAL FRAMEWORK  ABC Classes Analysis

This image shows codes used to classify inventory
products in XYZ classes.

Fig 4 ABC Classes Analysis

ABC-XYZ Classification analysis is a dual-tier

inventory management technique used to categorize items in
a business's inventory based on their value (ABC) and their
demand variability (XYZ). The ABC classification divides
items into three categories: A items, which are the most
Fig 2 Conceptual Framework
valuable and typically account for a significant portion of
revenue; B items, which have moderate value and
IV. DATA PRESENTATION
importance; and C items, which are less valuable and often
make up the majority of items but contribute less to revenue.
 XYZ Classes Classification
The XYZ classification, on the other hand, categorizes items
This image shows codes used to classify inventory
based on their demand patterns: X items have stable and
products in XYZ classes.
predictable demand, Y items have moderate demand
variability, and Z items have highly variable and
unpredictable demand. This combined classification helps
companies apply tailored inventory management strategies
to different item categories, optimizing stock levels and
reducing costs while ensuring adequate supply for critical
items. Businesses can get a far more thorough understanding
of their products and consumers' needs by combining ABC
and XYZ evaluations into an automated procedure.
ABC/XYZ analysis provides a total of nine classifications
for classifying materials, products, and consumption as
opposed to just three. This study is consequently one of the
most crucial tools for a long-term inventory optimization in
material management. A link between the value component
of individual products and the temporal requirement and/or
consumption process of the relevant piece is created by
combining the ABC and XYZ analyses. Important
implications for future assortment planning can be made
from this.

Fig 3 XYZ Classes Classification

IJISRT24FEB1099 www.ijisrt.com 1965

Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24FEB1099

V. IMPREMENTATION AND RESULTS

Fig 5 XYZ Products Demand per Month

In the above graph, XYZ classified products are all arranged in a row. When they are all displayed on the same scale, it
becomes clear that the difficult to forecast Class Y and Class Z SKUs are not just seasonal during the summer but also cause a
significant increase in demand.

Fig 6 Revenue Generated by ABC XYZ Classes

As you can see from the above graph, Class C contains a ton of SKUs, but Class C only accounts for 10% of the company's
overall income.

IJISRT24FEB1099 www.ijisrt.com 1966

Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24FEB1099

Fig 7 Demand by ABC XYZ Classes

The above graph represented the demand for each satisfaction by ensuring availability of key products when
category from our nine combined ABC XYZ classes is the needed but also aids in cost reduction by avoiding
total products demand received by a company from overstocking of low-value and low-demand items, thus
customers. As per the graph, products classified in a contributing to overall operational efficiency and
category that has class AY, AX and AZ are the ones provided profitability.
highest demand to the company compared with others.
 To Demonstrate the Effectiveness of ABC - XYZ Analysis
VI. INTERPRETATION OF FINDINGS in Improving Forecasting Accuracy and Reducing the
Risk of Overstocking or Stockouts During Periods of
 To Explore the Role of ABC - XYZ Analysis in Inventory High Demand at the Case Study
Management and its Impact on Demand Fluctuation at Our exploration into the effectiveness of ABC-XYZ
the Case Study analysis in improving forecasting accuracy and mitigating
The impact of ABC-XYZ analysis on demand the risks of overstocking or stockouts during periods of high
fluctuation objective is profound. It allows the case study to demand has yielded compelling findings. By classifying
prioritize resources, ensuring that high-value, high-demand products based on both their demand patterns (ABC
items are well-stocked while optimizing inventory levels for analysis) and their supply variability (XYZ analysis), we
low-value, low-demand items. By focusing on the demand observed a significant enhancement in our forecasting
characteristics of each item, this approach minimizes the precision. High-demand periods became more predictable,
bullwhip effect, reduces excess inventory costs, and allowing us to allocate resources more efficiently.
enhances overall supply chain efficiency, ultimately Furthermore, the combination of ABC-XYZ analysis
mitigating demand fluctuations and improving the enabled us to tailor inventory management strategies,
organization's responsiveness to changing market demands. resulting in a substantial reduction in both overstocking and
stockouts, thus optimizing inventory levels and ensuring
 To Analyze the use of ABC - XYZ Analysis in Identifying smoother operations even during volatile demand
High-Demand Products and Balancing Inventory Levels fluctuations. These findings underscore the invaluable role
to Meet Customer Demand During Peak Periods at the of ABC-XYZ analysis as a strategic tool for businesses
Case Study seeking to achieve greater supply chain resilience and cost-
The analysis of ABC - XYZ inventory management effectiveness.
methodology reveals its effectiveness in identifying high-
demand products and optimizing inventory levels to cater to VII. CONCLUSION
customer demand, particularly during peak periods. By
categorizing products based on both their demand value In conclusion, the implementation of ABC-XYZ
(ABC) and demand predictability (XYZ), this approach analysis in business inventory management at the case study,
helps the case study to focus their attention on items with powered by machine learning models, has yielded
the greatest impact on sales and enables the allocation of significant improvements in operational efficiency and cost-
resources for adequate stocking and timely replenishment. effectiveness. By classifying inventory items into categories
This targeted approach not only enhances customer based on their importance and demand patterns, this

IJISRT24FEB1099 www.ijisrt.com 1967

Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24FEB1099

technology-driven approach has enabled the case study to [12]. Pathak, A. (2023, July 14). Data Cleaning in Data
optimize its inventory control strategies. Machine learning Science: Process, Benefits and Tools. Retrieved from
models have played a pivotal role in forecasting demand, KnowledgeHut: https://www.knowledgehut.com/
optimizing reorder points, and reducing excess stock levels, blog/data-science/ data-cleaning
leading to reduced carrying costs and improved customer [13]. Potters, C. (2022, July 21). Demand: How It Works
satisfaction. This innovative approach has not only enhanced Plus Economic Determinants and the Demand Curve.
inventory management but has also positioned the case Retrieved from Investopedia:
study for sustainable growth in a competitive market by https://www.investopedia.com/terms/d/demand.asp
leveraging the power of data-driven decision-making. [14]. Pradiptamukherjee. (2023, June 15). What is a
Flowchart and its Types? Retrieved from
REFERENCES GeeksForGeeks: https://www.geeksforgeeks.org/
what-is-a-flowchart-and-its-types/
[1]. Aktunc, E. A., Basaran, M., Ari, G., & Gungor, M. I. [15]. Shahroudi, M. A. (2020). Statistical Forecasting for
(2019). Inventory Control Through ABC/XYZ Inventory Management. Milan: Researchgate.
Analysis. Industrial Engineering in the Big Data Era [16]. Shukla, S. (2020). CONCEPT OF POPULATION
(p. 13). Instanbul: Springer Nature. AND SAMPLE. India: Researchgate.
[2]. Atnafu, D., & Balda, A. (2018). The impact of [17]. SIKMUS, J. (2023, February 13). Pilot Study In
inventory management practice on firms’ Research: Definition & Examples. Retrieved from
competitiveness and organizational performance: Simply Sychology: https://simplypsychology.org/
Empirical evidence from micro and. Cogent Business pilot-studies.html#:~:text=Pilot%20studies%20are
& Management, 16. %20a%20fundamental%20stage%20of%20the,peopl
[3]. Auhl, M. (2021, August 6). What is an ARIMA e%20and%20trying%20out%20the%20study%20on
Mode? Retrieved from TowardsDataScience: %20them.
https://towardsdatascience.com/what-is-an-arima- [18]. Syntetos, A. A., Boylan, J., & Disney, S. M. (2009).
model-9e200f06f9eb Forecasting for inventory planning: A 50-year review.
[4]. Bennett. (2023, February 04). Business. Retrieved Journal of the Operational Research Society, 37.
from The Economic Times: [19]. Thirathon, U., Wieder, B., Matolcsy, Z., & Ossimitz,
https://economictimes.indiatimes.com/definition/busi M.-L. (2017). Impact of Big Data Analytics on
ness Decision Making and Performance. Thessaloniki,
[5]. Chetty, R. J. (2019, July 5). The manual methods of Greece: Open Publication Of UTS Scholars.
calculating the sample size of quantitative research. [20]. TUOVILA, A. (2022, September 21). Forecasting:
Retrieved from Project Guru: What It Is, How It’s Used in Business and Investing.
https://www.projectguru.in/the-manual-methods-of- Retrieved from Investopedia:
calculating-the-sample-size-of-quantitative-research/ https://www.investopedia.com/terms/f/forecasting.as
[6]. Ellen, S. (2020, December 14). Slovin's Formula p
Sampling Techniques. Retrieved from Sciencing: [21]. Wilson, M., Zenkova, Z., & Tarima, S. (2020).
https://sciencing.com/slovins-formula-sampling- Accounting for deficit in ABC-XYZ analysis. Rabat,
techniques-5475547.html Moscow: IEEE.
[7]. Evdokimova, S. A. (2021). Segmentation of store [22]. Zenkova, Z., & abanova, T. (2018). The ABC-XYZ
customers to increase sales using ABC-XYZ-analysis analysis modified for data with outliers. 2018 4th
and clustering methods. J. Phys.: Conf. Ser. 2032 International Conference on Logistics Operations
012117, 7. Management (GOL) (p. 32). Le Havre, France: IEEE.
[8]. HAYES, A. (2022, December 18). Autoregressive
Integrated Moving Average (ARIMA) Prediction
Model. Retrieved from Investopedia:
https://www.investopedia.com/terms/a/autoregressive
-integrated-moving-average-arima.asp
[9]. HAYES, A. (2022, July 30). The Supply Chain: From
Raw Materials to Order Fulfillment. Retrieved from
Investopedia:
https://www.investopedia.com/terms/s/supplychain.as
p
[10]. KENTON, W. (2022, May 27). What is Inventory?
Definition, Types, and Examples. Retrieved from
Investopedia:
https://www.investopedia.com/terms/i/inventory.asp
[11]. Min, H., & Zhou, G. (2002). Supply chain modeling:
past, present and future. In H. Min, & G. Zhou,
Supply chain modeling: past, present and future (p.
43). Louisville: ScienceDirect.

IJISRT24FEB1099 www.ijisrt.com 1968