Inform : Jurnal Ilmiah Bidang Teknologi Informasi dan Komunikasi Vol.
11 No.
January 2026.
P-ISSN : 2502-3470.
E-ISSN : 2581-0367
Integrating RAD and Design Thinking for Developing a Web-Based POS and Inventory Management System for MSMEs: A Case Study Imelda Dimentieva1.
Erba Lutfina2.
Galuh Wilujeng Saraswati3.
Resha Meiranadi Caturkusuma4
1,2,3,4
Information System Department.
Universitas Dian Nuswantoro PSDKU Kediri.
Indonesia 612202200069@mhs.
id (*) .
lutfina, galuhwilujen.
@dsn.
reshameiranadi@gmail.
Received: 2025-11-12.
Accepted: 2026-01-20.
Published: 2026-01-29 AbstractAiManual transaction and inventory recording in micro, small, and medium enterprises (MSME.
often leads to recording errors an d delayed decisions.
This study developed a web-based point-of-sale (POS) and inventory management system for Fresh Market Klatak to streamline transactions and stock control.
Development followed Rapid Application Development (RAD) and was integrated with Design Thinking to elicit user needs and iterate on prototypes rapidly.
The system was evaluated using scenario-based functional testing and user acceptance testing (UAT).
All functional test scenarios passed .
%).
UAT with 16 users produced an overall acceptance score of 93% (Very Goo.
, indicating the system is usable and meets operational requirements.
Future work will develop a mobile application and integrate paymentgateway services to improve accessibility and transaction efficiency.
KeywordsAiPoint of Sale.
SMEs.
Rapid Application Development.
Design Thinking.
Inventory Management System.
User Acceptance Testing.
decisions are based on invalid or outdated data, thereby impacting the store's operational performance .
One solution to overcome operational problems and improve business efficiency is implementing a POS system.
The POS system not only serves as a transaction recording tool but also plays an important role in supporting operational efficiency by enabling real-time management of stock, customer data, and suppliers .
Several traditional merchants have implemented integrated POS systems as part of their digital transformation.
These systems help speed up transactions, reduce input errors, and improve the accuracy of operational data .
Therefore, it is proposed to develop a website-based POS system that integrates transaction and inventory management on a single Several studies have demonstrated the benefits of web-based POS and inventory systems for MSMEs.
Research by .
found that implementing a website-based POS system improves operational efficiency and speeds up transactions through automated sales recording.
At the same time, .
reported similar results for a website-based inventory information Research .
also highlights that web-based, configurable information systems are crucial for supporting operational activities and improving data accuracy in small- to medium-scale businesses.
In previous studies, the Rapid Application Development (RAD) method has been proven to accelerate the system development cycle.
This research .
demonstrated that RAD is more flexible and time-saving than the linear, rigid waterfall Another study by .
showed that combining the RAD method with projects enables manual sales systems to be converted into faster, more responsive website-based systems.
However, the main weakness of RAD lies in the user design
stage, where low user involvement can result in a system
INTRODUCTION
Technological and communication developments have had a major impact on various aspects of life, including business and trade .
Digitalization has driven business model innovation in the retail sector, with the potential to improve operational efficiency, strengthen competitiveness, and enhance adaptability to market changes .
Digital transformation in Indonesia's Micro.
Small, and Medium Enterprises (MSME.
sectors is also being driven by the implementation of website-based information systems that support efficient management and empowerment strategies .
The implementation of cloud-based Point of Sale (POS) systems has been proven to improve the operational efficiency and non-financial performance of MSMEs in Indonesia, as these systems accelerate transaction processes, improve recording accuracy, and facilitate business control .
Although the application of digital technology has been proven to improve business efficiency and productivity, its adoption rate in Indonesia's MSME sector remains relatively This condition aligns with .
, which confirms that limited human resources, organizational readiness, and digital infrastructure are the main factors affecting the low adoption of technology in small and medium-sized enterprises.
At Fresh Market Klatak, for example, stock and transaction management are still conducted manually using physical books.
This manual process often leads to delayed updates, errors in stock records, and difficulties tracking sales, hindering efficient inventory management and timely procurement decisions.
These operational challenges highlight the need for a digital system that can automate both transaction recording and inventory Inaccurate inventory data often hinders stock management and procurement processes, as purchasing DOI : https://doi.
org/10.
25139/inform.
Inform : Jurnal Ilmiah Bidang Teknologi Informasi dan Komunikasi Vol.
11 No.
January 2026.
P-ISSN : 2502-3470.
E-ISSN : 2581-0367 design that fails to meetser needs .
In an effort to address this shortcoming, the Design Thinking approach offers a solution by emphasizing deep user understanding, problem exploration, and solutions that focus on the user experience .
Previous studies have explored the RAD method and the Design Thinking approach in developing information systems.
However, most of these studies tend to emphasize only one aspect, either accelerating development through RAD or improving user experience through Design Thinking, without clearly explaining how both approaches are systematically integrated, particularly in the context of web-based POS and inventory systems for MSMEs.
This indicates a research gap:
a lack of structured methodological integration that simultaneously supports rapid development and alignment with user needs.
To address this gap, the present research aims to design and evaluate a web-based POS and inventory system that integrates RAD and Design Thinking in a complementary manner.
The system is designed to automate transaction recording and stock updates, minimize errors, provide real-time sales and inventory reports to support timely procurement decisions, and improve overall operational efficiency.
Through this approach.
MSME operators are expected to manage transactions and inventory more effectively, reduce recording errors, and make faster, data-driven procurement decisions, while benefiting from a development process that is both efficient and practical for By clarifying the integration of these two approaches, this research contributes to existing studies by demonstrating a practical, user-focused system development approach in which RAD and Design Thinking are commonly applied separately.
The RAD method is applied in POS and inventory systems to accelerate the process of needs analysis, interface design, and system evaluation, which are carried out in stages.
strengthen the design and user experience.
Design Thinking is applied early to understand user needs, identify problems, and create prototypes that reflect user expectations.
Once the prototype is validated, the project transitions to RAD, enabling rapid, iterative system development, testing, and integration.
Although the RAD method emphasizes speed and iterative development, it still has limitations in deeply understanding user needs, particularly during the user design stage .
shown in Fig.
1, the RAD method is integrated with the Design Thinking method during the User Design stage to strengthen empathetic design and validation.
This integration ensures that the system is developed more efficiently while remaining aligned with user requirements .
The sequence of stages of the proposed method is as follows:
II.
RESEARCH METHODOLOGY
The RAD method is a system development approach that emphasizes a relatively short system development cycle to make the system creation process more efficient .
The main advantage of the RAD method lies in its ability to accelerate development time through rapid iteration and to obtain direct user feedback .
Ideate: Acarried out by developing various creative solution ideas that can answer user needs .
Requirements Planning: A stage of system development, problem identification, by analyzing user needs through interviews and observation .
This technique has been proven effective in previous studies, in which interviews with store owners and observations of sales processes and product management provided a more accurate understanding of the issueshe system needed to resolve .
Empathize: A process carried out to understand user needs and problems, motivations, and activities .
Define: A stage of formulating the main problems based on the results of empathy so that the design direction is more focused .
Prototype: Create an initial design of the interface and system flow for quick testing .
The prototype was created in Figma.
Testing: A prototype is evaluated with users to obtain relevant feedback .
The prototype was evaluated by 4 participants on Figma, all of whom regularly handle sales and inventory tasks at the fresh market, to obtain relevant feedback on the interface and system workflow.
They performed tasks such as navigating menus, adding products, completing transactions, and updating data.
The small sample size is justified because this testing focuses on early-stage prototype evaluation, aiming to identify usability issues quickly before full-scale deployment.
Each task completed was treated as the unit of observation, and each session lasted about 60 minutes per user.
The success rate was calculated using Equation .
At the same time, overall performance was summarized using the mean and standard deviation (SD) to represent both the average level of task completion and the consistency of the The SD was calculated using Equation .
Where, the ycuycn denotes the success rate for each task, the ycuO variable represents the mean success rate, and ycu refers to the total number of observed tasks.
Several validity threats could affect Fig.
Proposed Model DOI : https://doi.
org/10.
25139/inform.
Inform : Jurnal Ilmiah Bidang Teknologi Informasi dan Komunikasi Vol.
11 No.
January 2026.
P-ISSN : 2502-3470.
E-ISSN : 2581-0367 the results of prototype testing.
Users may perform better on tasks over time as they become more familiar with the system, potentially artificially increasing success rates.
Additionally, the small sample size may introduce sample bias, limiting the generalizability of the results.
To mitigate these threats, users were given a brief training session before testing, and each task was clearly defined and observed individually to ensure consistent measurement.
ycIycycaycayceycyc ycIycaycyce (%) = ycIya = Oo ycNycuycycayco ycycycaycayceycyc ycAycycoycayceyc ycuyce ycycyceycyc y 100 Oc.
cuycn Oe ycuI )2 ycu Statement Information on the page is easy to read and not confusing The system helps speed up the transaction process.
The system simplifies product and inventory data management.
Overall, the system meets user needs.
The UAT instrument consisted of 14 Likert-scale statements, rated on a five-point scale from 1 (Strongly Disagre.
to 5 (Strongly Agre.
The maximum possible score for each respondent is 70.
To obtain objective evaluation results, the scores are normalized into percentages using Equation .
Where mean represents the average score across all Likert-scale responses, and maximum weight refers to the highest possible total score a respondent can achieve.
The resulting percentage values indicate the level of user acceptance of the system.
After that, the average percentage will be assessed based on the following score interpretation Table i.
Construction: The process of developing the system based on the design results that have been validated s in the previous stage.
Testing is carried out in stages to ensure functionality and user comfort .
Work in the Construction phase includes writing program code, integrating modules, and conducting initial testing of system functionality.
Each module is tested in stages to minimize errors before the system is fully The development process uses the Laravel 10 framework for the backend and the Blade Template Engine for the frontend.
The database used is MySQL, and the interface design is based on the prototype results from the previous stage.
ycOyaycN ycEyceycycayceycuycycayciyce = ycAyceycaycu ycoycaycuycnycoycyco ycyceycnyciEayc y 100% .
TABLE i SCORE INTERPRETATION Percentage Description .
Cutover: The final stage that helps ensure the system runs as required and in line with real-world usage scenarios.
Testing uses two approaches: Scenario-Based Testing (SBT) and User Acceptance Testing (UAT) .
The UAT was conducted with 16 respondents who regularly handle store operations to evaluate system performance in real-world Participants were asked to complete a questionnaire consisting of 14 test items to assess directly, using a 1-5 Likert scale, as listed in Table I.
Participation was voluntary, and all responses were kept confidential.
The Likert scale was used to assess the system using 14 statement items covering aspects of usability, functional suitability, and user satisfaction, as shown in Table II.
0% - 20%
Very poor
21% - 40
Poor
41% - 60%
Fair
61% - 80%
Good
81% - 100%
Very good Several factors could affect the validity of SBT and UAT As participants become more familiar with the system, task performance may improve beyond first-time usage levels.
Additionally, the limited number of respondents may reduce the generalizability of the findings.
To reduce these risks, testing tasks and scenarios were randomized, participants were selected to cover a range of experience levels, and clear, stepby-step instructions were provided before testing to ensure consistent and reliable evaluation.
TABLE I
LIKERT SCALE DESCRIPTION
Weight
Description Strongly Disagree (STS) Disagree (D) Somewhat Agree (SA) Agree (A) Strongly Agree (SA) TABLE II
TEST STATEMENT
Statement The system can display product data correctly.
The cashier transaction feature functions as needed.
The system automatically updates stock after transactions.
The login and logout processes work correctly.
The system interface is easy to understand and use.
The menu and navigation buttons are easy to find.
Users can complete transactions without difficulty.
The system responds quickly to every action.
No errors or crashes occur during transactions.
Attractive and user-friendly interface design
P11
P12
P13
P14
RESULTS AND DISCUSSION
Requirements Planning The outcomes of this stage are presented in Table IV, which summarizes the system requirements identified based on user
TABLE IV
FUNCTIONAL AND NON-FUNCTIONAL REQUIREMENTS
Functional Requirements Non-Functional Requirements Users can manage product data The system is equipped with login in the system.
authentication and access restrictions based on user roles .
dmin & cashie.
P10
Users can create and update product categories.
The system automatically reduces product stock according to the number of products sold.
The system interface is designed to be simple and easy to understand, requiring no additional training.
The system can be run on modern browsers (Chrome.
Firefox.
Edg.
and desktop devices.
DOI : https://doi.
org/10.
25139/inform.
Inform : Jurnal Ilmiah Bidang Teknologi Informasi dan Komunikasi Vol.
11 No.
January 2026.
P-ISSN : 2502-3470.
E-ISSN : 2581-0367
TABLE VI
RESULT OF USER ISSUES
Transactions are processed manually using a calculator.
Inventory recording is still done manually using notebooks.
Difficulty monitoring fast-moving inventory.
Difficulty viewing transaction data in real time.
Transaction data is not stored properly.
Functional Requirements The system can display reminder notifications when product stock falls below the minimum threshold.
Non-Functional Requirements The system's code structure is modular, using the Laravel framework for easy maintenance and development.
Key Issues Users can select and add products to the cart for The system can be accessed at any time as long as there is an internet Reasons Users can access the cashier and admin pages according to their assigned roles.
The system can store data consistently without losing data.
Cashiers take longer to calculate the total price and often make calculation errors.
The recording process is inefficient and prone to delays when the store is busy.
There are no notifications or indicators for low stock levels.
Transaction and stock data updates are not automatic.
There is no centralized storage.
Impact Potential errors in transaction calculations.
Discrepancies between physical stock and records Delays in product procurement Difficulty in determining current stock and transaction Risk of data loss or incompleteness.
User Users need a system that automatically calculates the total price and saves transaction data directly.
Users need a feature that automatically updates stock after a transaction is completed.
Users need an automatic reminder feature when inventory reaches the minimum threshold.
Users require a feature for real-time recap and updates of stock and transaction data.
Requires a system integrated with a database to store data User Design .
Empathise: Based on observations, the transaction and product management processes are still carried out manually, as shown in Table V.
This manual process often leads to delays in stock recording, transaction calculation errors, and slow responses to product restocking requests.
The observation results form the basis for designing the POS system to address issues of efficiency and data accuracy.
Says TABLE V
USER EMPATHY MAP
We still calculate the total price manually.
There are often discrepancies between stock records and physical stock.
Forgot to update stock data after a transaction.
Must open today's record recap to see transaction history Inventory data is not stored properly.
Thinks How can transactions be made faster and more accurate? Is it possible for transaction and inventory data to be immediately visible without having to re-enter the data? How can all inventory history be automatically saved and not easily lost? Calculate the total price using a calculator.
Comparing the physical quantity of goods with manual records Does Rewriting stock records after sales are complete Check sales records at the end of each day.
Write down transactions in a logbook.
Tired Confused when there is a stock discrepancy Feels Overwhelmed Troublesome and worried Anxious because the data is not secure Is it possible for transaction and inventory data to be immediately visible without having to re-enter the data? How can all inventory history be automatically saved and not easily lost? TABLE VII
RESULT OF USER REQUIREMENTS DETAILS
System Page Requirements Dashboard Page Displays general information such as transaction history, sales charts, total number of employees, total revenue for the day, and total products sold.
Define: Based on the results of the previous stage, several major issues were identified that hindered operational efficiency at Fresh Market, as summarised in Table VI.
Based on these findings.
Table VII was compiled to map out the details of user requirements.
This was compiled as a basis for designing a website-based POS and inventory management Product and Inventory Management Page Displays and facilitates product management, including adding new products, editing product data, changing product sales status, managing product inventory, and viewing product inventory Transactions and Payments Page Displays products currently being sold and used to serve customers and process User Management Page Supplier Management Page Allows admins to view and manage users when access changes.
Displays and stores information about product suppliers.
Ideate: The user flow diagram in Fig.
2 shows the user's interaction with the Fresh Market Klatak POS system.
The process starts from the login page, then the system validates the user account.
The user flow provides an overview of the system's functions.
It serves as the basis for developing use cases, which describe each user action in detail and its interactions with the system.
The designed Use Case Diagram in Fig.
3 serves as a reference for the initial testing of functionality and user requirements.
DOI : https://doi.
org/10.
25139/inform.
Inform : Jurnal Ilmiah Bidang Teknologi Informasi dan Komunikasi Vol.
11 No.
January 2026.
P-ISSN : 2502-3470.
E-ISSN : 2581-0367 displays a transaction summary that clearly shows the total price, number of products, and change calculation.
In the management section, the dashboard displays sales summaries, the number of products, total revenue, and daily sales performance graphs.
In addition, the stock management page design shows the flow of stock change records, which are projected to be displayed in Tables for easy monitoring.
Overall, the prototype shows the direction of the system design, which focuses on clarity of workflow, reduction of manual errors, and improved ease of monitoring operational data.
The resulting design provides an initial overview of the expected user experience and serves as a basis for evaluation before entering the actual development stage.
Thus, the prototype significantly helps ensure that user needs are accurately mapped before implementation.
Fig.
User Flow Fig.
Prototype System .
Testing Prototype: The results of the prototyping test calculations are presented in Table Vi.
Based on the prototype testing results shown in Table Vi, the five tested features achieved an average success rate of 90% with a standard deviation of 13.
69, indicating moderate variation in task success across features.
Three features (F1.
F4, and F.
achieved a perfect 100% success rate, while F2 and F3 recorded a 75% success rate due to minor execution errors during This variation indicates that although the system's core functions operate reliably, some interaction flows still need improvement, especially for first-time users.
The decrease in success rates observed in F2 and F3 indicates that errors occur more often in features involving sequential steps, suggesting that further improvements in usability design are needed.
Overall, the results indicate that the developed prototype functions properly, supports smoother operational processes, and reduces the potential for errors during system use.
Fig.
Use Case Diagram .
Prototype: The display in Fig.
4 depicts a prototype of a web-based Point of Sale system interface designed to support transaction processing, product management, and sales The prototype shows how the main functions are mapped into a structured interface design.
The login page serves as a simple and secure authentication stage.
After the user logs in, the cashier page provides product selection, search, category filter, and shopping cart management features to support a more efficient transaction flow.
The payment page
Features (F) TABLE Vi
PROTOTYPE TESTING RESULTS
Users Successful Failed Success Rate (%) Mean Standard Deviation DOI : https://doi.
org/10.
25139/inform.
Inform : Jurnal Ilmiah Bidang Teknologi Informasi dan Komunikasi Vol.
11 No.
January 2026.
P-ISSN : 2502-3470.
E-ISSN : 2581-0367 Construction The results of the Construction phase are demonstrated through several main system interface views.
Fig.
illustrates the main gateway of the system, which differentiates access rights for Admin and Cashier users.
Fig.
shows the Dashboard page displayed to users with admin roles after successful login, which shows summary information such as the number of products, products with low stock, and total daily This display can help users quickly monitor the store's condition in real time.
The product management page shown in Fig.
manages all product information, including adding, updating, and deleting product data.
The product management page also displays notifications when stock reaches the minimum stock Fig.
shows the Stock Management page, which records all inventory changes in a structured stock change log.
Automatic and manual stock adjustments are consistently stored, enabling transparent tracking of inventory movements.
This functionality supports more accurate stock monitoring and reduces discrepancies between recorded and physical inventory.
After logging in, in Fig.
, users with the cashier role will be directed to the transaction page.
The system displays a product catalogue equipped with an "Add" button to add products to the cart.
The system can automatically calculate the total price, tax, and total.
After all products have been entered and the total purchase has been calculated, the cashier will press the "Pay" button to proceed to the payment page.
Fig.
shows the payment interface used to finalize transactions.
The page provides a clear transaction summary and payment entry, enabling faster transaction completion and reducing calculation errors during the payment process.
Product Management Page .
Stock Management Page .
Transaction Page .
Login Page .
Payment Page .
Dashboard Page Fig.
System Interface Views DOI : https://doi.
org/10.
25139/inform.
Inform : Jurnal Ilmiah Bidang Teknologi Informasi dan Komunikasi Vol.
11 No.
January 2026.
P-ISSN : 2502-3470.
E-ISSN : 2581-0367 Cutover Scenario-based testing was conducted on the POS and webbased systems at Fresh Market Klatak, using 3 test samples representing the cashier and admin roles.
Each scenario is based on business processes that occur in the field, such as login, product management, sales transactions, transaction cancellations, and system logout, as listed in Table IX.
If an activity can be performed by both roles, namely admin and cashier, then the actor is written as user.
activities that can only be performed by a specific role are written according to that role's access rights.
Test results show that 100% of scenarios run as expected.
The system performs its main functions well, including displaying product data, automatically reducing stock after transactions, and printing payment receipts.
This result indicates that the implemented business processes are well aligned with operational workflows, enabling the system to function reliably in real-world use cases.
Based on these results, it can be concluded that the POS and inventory systems have met user needs functionally and can be used effectively in store operations.
Testing Scenario The admin updates the Expected Results The admin clicks the existing stock button, then updates the product's stock.
The system displays the stock management page and saves the stock changes in the The admin The admin clicks the " Add Supplier " menu, then fills in the desired supplier data.
The system displays the supplier page and saves the supplier data to the database.
Correct The cashier The cashier selects products, adds them to the cart, determines the purchase quantity, and completes the The system displays the cashier page, adds the selected products to the cart, automatically reduces the stock, and displays the total payment.
Correct The cashier cancels the The cashier presses the "Cancel" button before payment The system will return to the transaction page, not save the transaction, and the stock remains Correct The cashier prints the The cashier enters the amount of money and the customer's payment method, then presses the submit button.
The system will save
the amount of money
and payment method
in the database and print the purchase Correct
The user
logs out of the system
The user presses the "Logout" button
The system ends the session and displays the login page
Correct
TABLE IX
SYSTEM TESTING SCENARIOS RESULT
Testing Scenario User logs in to the Results Test Correct Testing Steps Expected Results The user enters their username and password based on their role (Admin/Cashie.
, then presses the "Login" button.
The system displays the dashboard or transaction page according to the User fails to log in due to The user enters the wrong username and The system displays an error message and denies access.
Correct The admin adds new The admin opens the product menu, fills out the add product form, and clicks the save button.
The new product is saved in the database and appears in the product list Correct The admin updates the The admin selects the product to be edited, clicks the edit button, makes changes, and then clicks the save The system displays the product edit page and saves the updated Correct The admin deletes the The admin clicks the delete button on the product list The system displays a confirmation message and deletes the data from the database Correct The admin The admin clicks the add category button available on the product management page, then adds the desired product The system displays the product category page and saves the category in the Correct Results Test Correct Testing Steps User Acceptance Testing (UAT): The results of the User Acceptance Testing (UAT) are presented to describe user acceptance of the developed web-based POS and inventory The assessment involved 16 respondents who interacted directly with the system and subsequently completed a questionnaire comprising 14 evaluation items, rated on a fivepoint Likert scale.
To provide a more representative description of user responses, the analysis employed median values and interquartile ranges (IQR) for each item, as these measures are suiTable for ordinal data and are less sensitive to extreme As shown in Table X, all evaluation items achieved median scores of 4.
5 or higher, indicating strong user agreement across all assessed aspects.
The relatively narrow IQR values, ranging from 0.
00 to 1.
00, indicate low variability in user responses, reflecting consistent perceptions of system performance and usability.
In addition to the item-level analysis, the overall user acceptance level was calculated by aggregating respondents' scores and converting them to a The system achieved an overall acceptance score of 93%, as illustrated in Fig.
TABLE X
UAT ITEM STATISTICS RESULT
Item Median IQR DOI : https://doi.
org/10.
25139/inform.
Inform : Jurnal Ilmiah Bidang Teknologi Informasi dan Komunikasi Vol.
11 No.
January 2026.
P-ISSN : 2502-3470.
E-ISSN : 2581-0367
Item
P10
P11
P12
P13
P14
Median functionally reliable and well accepted by users at Fresh Market Klatak.
Empirical evidence from the evaluation phase shows that all defined business scenarios were successfully executed, with Scenario-Based Testing achieving a 100% pass rate across core workflows, including login, transaction processing, inventory management, and reporting functions (Table IX).
Furthermore.
User Acceptance Testing (UAT) involving 16 respondents yielded an overall acceptance score of 93%, classified as Very Good, indicating a high level of usability, functional suitability, and user satisfaction (Fig.
These results demonstrate that integrating Rapid Application Development (RAD) and Design Thinking effectively supports rapid system development while maintaining alignment with operational needs.
The RAD approach facilitated iterative implementation and continuous refinement based on user feedback.
In contrast.
Design Thinking ensured that system workflows and interface designs addressed real user problems identified during the empathy and prototyping stages.
This methodological integration improved transaction efficiency, enabled automatic stock updates, and provided more accurate operational data to support informed decision-making.
Nevertheless, several limitations should be acknowledged.
The system was evaluated in a single-store context and employed a non-experimental, descriptive evaluation approach, which limits the generalizability of the findings to other MSME settings.
In addition, the evaluation focused on short-term system usage and did not assess long-term performance, scalability, or system resilience under higher transaction volumes.
Future work should address these limitations by integrating a payment gateway to enable secure cashless transactions and by developing a mobile phone application to enhance accessibility, mobility, and real-time operational control for store owners and cashiers.
Further enhancements may include support for multi-branch management, system failure logging, and data quality auditing mechanisms to improve system reliability and enable more accurate, data-driven decisionmaking in MSME operations.
IQR
UAT Percentage Result (%) 96% 95% P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 Fig.
UAT Presentage Result According to the acceptance criteria defined in Table i, this score falls into the "Very Good" category, indicating that the developed POS and inventory system is highly accepTable and well aligned with the users' operational needs at Fresh Market Klatak.
Test results indicate that all defined scenarios were successfully executed, demonstrating that the system consistently supports essential operational workflows, including product management, transaction processing, stock updates, and receipt generation.
Beyond confirming functional correctness, this result reflects the system's robustness in handling real-world operational sequences without interruption.
Compared to previous studies on web-based POS systems for MSMEs .
, which reported occasional inconsistencies in stock synchronization and transaction logging, the proposed system demonstrates improved reliability in maintaining data consistency between transactions and inventory records.
From a practical perspective, this reliability reduces operational risk, minimizes human error, and increases cashier efficiency during peak transaction periods.
The findings also suggest that closely aligning system workflows with actual business processes is a critical factor for achieving sTable system performance in small retail environments.
However, this study has several limitations.
The evaluation was conducted in a single retail environment with a limited number of respondents, and the system has not yet been integrated with advanced features such as a payment gateway, mobile application access, or sales forecasting.
Future work may explore these additions to enhance system functionality and REFERENCES