Optimizing inventory management and demand forecasting system using time series

algorithm
Alexandra An Pavon Arnaiz 1, Lovely Samuele Cristal 1, Antonette Obligado Fernandez 1, Mark Rowie Flores
Gubaton 1, *, Domingo Valenzuela Tanael 2 and Criselle Jose Centeno 2
1 College of Computer Studies, Global Reciprocal Colleges, Philippines.
2 Information Technology Department, Faculty of Assistant Professor, Pamantasan ng Lungsod ng Maynila, Philippines.

World Journal of Advanced Research and Reviews, 2023, 20(03), 021–027

Publication history: Received on 19 October 2023; revised on 28 November 2023; accepted on 30 November 2023

Article DOI: https://doi.org/10.30574/wjarr.2023.20.3.2456

Abstract
In the rapidly evolving business landscape, effective inventory management and meeting customer demands rely
heavily on accurate forecasting. While technology automates parts of inventory control, human expertise remains vital
in decision-making for forecasting. Building supplier relationships, monitoring market trends, and adaptable supply
chains are crucial too. Accurate demand forecasting reduces costs, streamlines operations, and boosts customer
satisfaction. Therefore, companies must carefully review their forecasting methods to stay competitive. Researchers are
addressing the lack of data on inventory and forecasting by focusing on implementing time series algorithms,
recognizing their crucial role in optimizing these processes. This academic pursuit has led researchers to develop a user-
friendly system tailored for improved inventory management, integrating a feature set dedicated to demand forecasting.
The project aims to streamline user operations by offering an intuitive platform that facilitates seamless navigation. By
encompassing forecasting capabilities, the system empowers businesses to accurately predict their future product
requirements. The primary objective of this initiative is to simplify inventory procedures while enabling users to
proactively meet upcoming demands effectively. While conducting the study, the researcher considered the first
problem in how the user will use the inventory system in a more user-friendly manner. The second problem that the
researchers conducted was manual input, and it will cost more when the documents are not organized. Lastly, the
highest problem that the inventory management conducted was the overseers of the products by excessive inventory,
low stocks, and expired products. The researchers use some of the sub-characteristics of ISO 25010 that are appropriate
for evaluating inventory management. After evaluation, the sub-characteristics of functional stability garnered an
overall weighted mean of 3.90. The compatibility and usability garnered an overall weight of 3.89. Reliability garnered
an overall weight of 3.66. Lastly, maintainability was overall weighted at 3.63. The confusion matrix was used with the
help of the tool of Weka Software using the scheme of function. Simple Logistics. The evaluation on the training set has
a summary of correctly classified instances of 89.4737% and incorrectly classified instances of 10.5263%, which
indicates that the application has an accurate algorithm.

Keywords: Inventory; Forecasting; Time-series; Supplies; Reorder Point

1. Introduction
Nowadays, business is hard and fast-paced, making it increasingly difficult to keep up with changing market conditions
and customer demands. Advances in technology, such as machine learning and predictive analytics, have made it
possible to automate some aspects of forecasting and streamline inventory management processes. However,
technology alone cannot replace the human expertise and judgment needed to develop effective forecasts and make
informed decisions about inventory. In short, while forecasting is a key component of inventory management,


Corresponding author: ARNAIZ ALEXANDRA AN.
Copyright © 2023 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0.
 World Journal of Advanced Research and Reviews, 2023, 20(03), 021–027

businesses must also invest in developing strong relationships with suppliers, monitoring market trends, and
maintaining a nimble, responsive supply chain to succeed in today's fast-paced business environment.

Inventory management is the process of overseeing the flow of goods in and out of a company's warehouses and retail
locations. Effective inventory management is essential for companies to maintain optimal levels of inventory to meet
customer demand while minimizing costs associated with holding and storing inventory. Forecasting plays a critical
role in inventory management as it helps companies anticipate future demand for their products and services.
Forecasting involves analyzing historical sales data, market trends, and other relevant factors to predict future demand
for a product.

Effective inventory management with accurate demand forecasting can help companies reduce costs, optimize
operations, and improve customer satisfaction. The selection of the appropriate forecasting technique depends on a
variety of factors, and companies must carefully evaluate their options to ensure that they are using the most effective
method for their specific needs.

1.1. Statement of the problem

 Due to manual labor, the store needs to enhance its inventory management system to improve inventory
visibility and optimize inventory turnover.
 The store is struggling due to inadequate demand prediction, leading to stockouts, excess inventory, and
financial loss.
 The store is facing challenges in meeting customer preferences due to inefficient forecasting of seasonal
demand resulting in slow inventory sales.
 Existing systems evaluate historical sales data and inventory levels to assess product demand and optimize
stock replenishment based on past trends and business rules.

1.2. Objectives of the study

1.2.1. General Objectives

The general objective of this study is to address issues with inaccurate demand prediction and inadequate inventory
management and to improve forecasting accuracy, reduce costs, enhance customer satisfaction, and optimize inventory
levels for efficiency and profitability.

1.2.2. Specific Objectives

1. To establish an inventory monitoring system with a control dashboard to help improve inventory visibility and
track inventory levels, turnover.

 Login authorization
 Suppliers, employees, and products information
 User-friendly interface

2. To set up a system for automatic inventory forecasting models that uses real-time inventory monitoring to
predict future demand for products.

 Automatically highlight low stock levels and prevent excessive inventory

 Stock aging tracking to determine how long the inventory has been in stock
 Automatic disposal of expired products

3. Implement an inventory feature that uses forecasting to improve the accuracy of demand forecasting by using
time series and algorithm techniques.

 Generate predictions of future demand for products every monthly and yearly
 Generate and export purchase order
 Options to generate and export inventory reports in different file formats

4. To test and evaluate the developed system using ISO 25010 standards to enhance demand forecasting accuracy,
minimize stockouts and overstock, reduce costs, and ensure scalability and adaptability to changing market
conditions.

22
 World Journal of Advanced Research and Reviews, 2023, 20(03), 021–027

1.3. Scope
The system will be designed to support forecasting techniques for inventory management. It will enable businesses to
predict demand for their products, forecast future inventory levels, and generate purchase orders. The system will
support various forecasting methods such as time series. The system will enable businesses to track their inventory
levels efficiently. It will allow users to create and manage inventory items, monitor product movement, and set reorder
points. It will allow users to generate reports on inventory levels, product movement, and stockouts.

1.4. Delimitation
The system will not be responsible for physical movements of inventory, such as shipping and receiving, or the
management of warehouse operations. It will not include features for managing other aspects of business operations,
such as accounting, customer relationship management, or human resources management. Additionally, the system will
not offer any physical inventory management capabilities, such as barcode scanning or RFID tracking. The system will
not make decisions on behalf of the business regarding inventory management or purchasing. It will provide data and
recommendations, but ultimately, the business owner or manager will be responsible for making decisions. The system
will also not include features for managing supply chain relationships or supplier information.

2. Methodology

2.1. Conceptual Framework

Figure 1 Conceptual Model of the System

Figure 2 shows the input, process, and output. The input must contain such things as knowledge, software, and hardware
requirements. Knowledge Requirements is the key to the development of the system Software Requirements are the
applications that should be used during the development of the system, Hardware Requirements are the physical
components of the device to run the system smoothly. During the process the system will undergo agile development,
such as meet and plan, design, code and test, release, and feedback. Lastly is the output, the system will be published or
released to use by the user.

2.2. System Development

Figure 3 illustrates the entire process, incorporating an Agile development process involving analyzing user
requirements, designing a program with flexibility for future changes, coding incrementally, documenting and testing
the system, and maintaining it. This approach emphasizes collaboration with clients and stakeholders, allowing for

23
 World Journal of Advanced Research and Reviews, 2023, 20(03), 021–027

continuous integration and testing. The system is then operational, ensuring its functionality and continuous
improvement. By adopting Agile principles, the team aims to deliver a high-quality system that meets evolving needs.

Figure 2 System Development Life Cycle

2.3. Instrument Used

The researchers utilized a survey interview as the primary instrument to gather data and insights from participants
regarding inventory management and forecasting. The survey interview was thoughtfully designed to include a
structured set of questions specifically aimed at exploring various aspects of inventory management and forecasting
practices. Through this instrument, the researchers were able to obtain valuable data from a diverse pool of
respondents, shedding light on their experiences, perspectives, and strategies related to inventory management and
forecasting. The survey interview facilitated the gathering of detailed information, opinions, and perspectives on the
challenges, strategies, and best practices associated with inventory management and forecasting. Overall, the survey
interview proved to be an effective tool for gathering crucial insights into inventory management and forecasting,
enhancing the researchers' contribution to the existing body of knowledge in this area.

3. Results
The researchers used ISO 25010 as a complete framework for quantifying and evaluating the inventory management
system. To determine the system's capabilities, comprehend the user response, and collect useful information from the
respondents for the system's improvement, the researchers conducted a project review. The researchers deploy a paper
survey as well as an online survey using Google Forms. Fifty (50) people have responded. Twenty-nine (29) people
responded in a paper survey and Sixteen (16) people responded to an online survey via Google Form. To test the system,
five (5) indicators were available.

3.1. Functional Suitability Overall Weighted Mean: 3.91

 Functional Completeness: 3.80
 Functional Correctness: 3.90
 Functional Appropriateness: 4.02

3.2. Compatibility Overall Weighted Mean: 3.89

 Co – Existence: 3.92
 Interoperability: 3.86

3.3. Usability Overall Weighted Mean: 3.89

 Appropriateness Recognizability: 4.06
 Learnability: 3.82
 Operability: 3.96
 User Error Protection: 3.62
 User Interface Aesthetics: 4.10
 Accessibility: 3.78

24
 World Journal of Advanced Research and Reviews, 2023, 20(03), 021–027

3.4. Reliability Overall Weighted Mean: 3.66

 Maturity: 3.72
 Availability: 4.00
 Fault Tolerance: 3.26

3.5. Maintainability Overall Weighted Mean: 3.63

 Modularity: 3.80
 Reusability: 3.76
 Analyzability: 3.48
 Modifiability: 3.58
 Testability: 3.54

3.6. Evaluation
Table 1 Evaluation ISO 25010

EVALUATION (ISO 25010)

Functional Overall
P F G VG E Sub - Characteristics
Stability Weighted Mean
Q1 1 2 11 28 8 Functional Completeness
Q2 2 0 10 27 11 Functional Correctness 3.91
Q3 2 1 7 24 16 Functional Appropriateness
Overall
Compatibility P F G VG E Sub - Characteristics
Weighted Mean
Q4 1 0 13 24 12 Co – Existence
3.89
Q5 1 2 11 25 11 Interoperability
Overall
Usability P F G VG E Sub - Characteristics
Weighted Mean
Q6 0 1 12 20 17 Appropriateness Recognizability
Q7 0 1 17 22 10 Learnability
Q8 1 0 15 18 16 Operability
3.89
Q9 0 4 17 23 6 User Error Protection
Q10 0 3 8 20 19 User Interface Aesthetics
Q11 0 2 15 25 8 Accessibility
Overall
Reliability P F G VG E Sub - Characteristics
Weighted Mean
Q12 1 2 17 20 10 Maturity
Q13 1 2 8 24 15 Availability 3.66
Q14 5 5 18 16 6 Fault Tolerance
Overall
Maintainability P F G VG E Sub - Characteristics
Weighted Mean
Q15 1 5 9 23 12 Modularity
Q16 2 5 9 21 13 Reusability
Q17 4 5 13 19 9 Analyzability 3.63
Q18 2 7 12 18 11 Modifiability
Q19 4 4 15 15 12 Testability

25
 World Journal of Advanced Research and Reviews, 2023, 20(03), 021–027

It shows how effective and efficient when the respondent’s uses the system.

3.7. Confusion Matrix

Figure 3 Confusion Matrix

4. Discussion
Providing an easy-to-use inventory management system with the ability to apply time-series technique forecasting is
the main goal of this project. The researchers discovered the benefits and drawbacks of creating an inventory
management procedure through this research. The researchers set out to give users a safe and efficient method using
demand forecasting and time series algorithms that can be useful for inventory management, as it can help to forecast
the future demand for products or services and adjust the inventory levels accordingly. By using time series analysis,
inventory managers can avoid overstocking or understocking, and improve customer satisfaction and profitability.
Despite its apparent simplicity, it is the method by which items are tracked along the whole supply chain, from purchase
through manufacturing to final sales. It is important to forecast because it can help to reduce inventory costs, meet
customer demand, and improve workflow efficiency using an online website. It is a system that can deliver the required
outcomes in the future and provide users with streamlined, and well-organized data that can help them save time and
effort. The researchers decided to take advantage of the opportunity to fully utilize the potential of the inventory
management system.

For future researchers, after the development of the system has been effective, there is always room for improvement.
The researchers would like to suggest that in order to make the system even better, it should be improved further by
incorporating auto-generated reports or product sorting’s that are more convenient and adaptable, in line with current
trends. This technology has the power to significantly increase productivity and reduce user burdens. These
recommendations are meant to help future researchers make sure their own research is accurate and valuable by using
this study as a reference.

26
 World Journal of Advanced Research and Reviews, 2023, 20(03), 021–027

5. Conclusion
In conclusion, even though the evaluation's results indicate that the system's failure tolerance is not particularly great,
the researchers can still conclude that the system's user interface aesthetics were accurate and safe. Moreover, among
all the results, the system's recognizability comes in second. Furthermore, the system is adaptable and will continue to
function even in the event of changes. It is known to the researchers that users can still utilize the system despite its
flaws. The researchers set up an inventory monitoring system with a controllable dashboard and improved interface to
help track inventory levels, log-in authorization, supplier, employee, and product information and improve inventory
visibility. In some ways, the system was in line with the first intended objective. In addition, with regard to objective 2,
the researchers put in place a system for automatic inventory forecasting models that make use of real-time inventory
monitoring to forecast future demand for products. These models include the following features: automatic expiration
of products; stock aging tracking to ascertain how long a product has been in stock; automatic highlighting of low stock
levels; and prevention of excessive inventory. Additionally, the researchers implemented an inventory feature that
makes use of time series and algorithmic techniques to forecast demand more accurately. Users will also be able to
generate predictions of future product demand on a monthly and annual basis, generate and export purchase orders,
and generate inventory reports in a variety of file formats, including CSV, JSON, and XLS. Lastly, in order to accurately
assess user satisfaction with time-series algorithm-based demand forecasting and inventory management systems, the
researchers employed a survey as part of Objective 4. The researchers employed ISO 25010 as their evaluation
methodology.

Compliance with ethical standards

Acknowledgements
First and foremost, the researchers would like to express a sincere gratitude and appreciation to all those who have
provided support and assistance throughout the completion of this thesis. The researchers extend heartfelt thanks to
the supervisor, Ms. Centeno, Criselle, for their invaluable guidance, unwavering support, and continuous
encouragement. The supervisor's expertise and insights have significantly contributed to the quality and refinement of
this research. Lastly to thank God Almighty been a part of this process, making it a fulfilling and rewarding experience.

Disclosure of conflict of interest

No conflict of interest to be disclosed.

References
[1] Blunsdon, B. “Inventory management in retail: A case study on Walmart.” Researchgate, 2016
https://www.researchgate.net/publication
[2] Arora, R., & Kumar, A. "Importance of demand forecasting in inventory management” Researchgate, 2020
https://www.researchgate.net/publication/327793184_A_Study_of_Inventory_Management_System_Case_Stud
y
[3] Jonathan Hans Soeseno, Sergio González, Trista Chen “Streamlined Framework for Agile Forecasting Model
Development towards Efficient Inventory Management.” Researchgate 2023
https://www.researchgate.net/publication/370001220_Streamlined_Framework_for_Agile_Forecasting_Model
_Development_towards_Efficient_Inventory_Management
[4] Satur Dorado Apolonio, Marvin I. Norona (2021) Automating Inventory Management in Distribution Centers of a
Leading Fast Moving Consumer Goods (FMCG) Industry Player in the Beverage Market: A Supply Chain 4.0
Journey https://ieomsociety.org/proceedings/2021indonesia/277.pdf
[5] Larry M. Ichon Jr., Jerd M. Dela Gente “Time Series Analysis of Mean Temperature using SARIMA: An Example
from Davao Oriental, Philippines” 2023 https://doi.org/10.59120/drj.v14i1.91

27