JOINS (Journal of Information System) https://publikasi.dinus.ac.id/joins <p align="justify"><span style="vertical-align: inherit;"><span style="vertical-align: inherit;"><span style="vertical-align: inherit;"><span style="vertical-align: inherit;"><strong>JOINS (Journal of Information System)</strong> is a national journal, which emphasizes the aspects of theory, research, and intellectual development of information systems in organizations, institutions, economics, and society. </span></span></span></span>This journal is intended to be dedicated to the development of sustainable knowledge related to the application of information technology in organizations and management, and moreover to improve economic and social welfare. <span style="vertical-align: inherit;"><span style="vertical-align: inherit;"><span style="vertical-align: inherit;"><span style="vertical-align: inherit;">All submitted manuscripts must be research-based and use appropriate methodology.</span></span></span></span></p> <p align="justify"><span style="vertical-align: inherit;"><span style="vertical-align: inherit;"><span style="vertical-align: inherit;"><span style="vertical-align: inherit;"><strong>Topics include</strong>: Design and Implementation System, Information Systems and Change Management, Decision Support System, E-Learning, E-Commerce, Methodology and Software Development, Knowledge Management System, Geographic Information System, Healtcare Information System, Enterprise Application Integration, Enterprise Resourse Planning, IT Governance, IT Management, Information Security, Information Science and Technology.</span></span></span></span></p> <p align="justify"><span style="vertical-align: inherit;"><span style="vertical-align: inherit;"><span style="vertical-align: inherit;"><span style="vertical-align: inherit;"><strong>JOINS</strong> is published by Study Program of Information Systems, Dian Nuswantoro University, every six months, namely in <strong>May</strong> and <strong>November</strong>. Starting from <strong>2027 (Volume 12),</strong> JOINS is published four<strong> (4) times</strong> a year, in <strong>January, April, July, and October.</strong> There is no submission deadline, and manuscripts are published in the order in which they complete the review process. All submitted manuscripts undergo a double-blind peer review process. Papers may be submitted in either <strong>INDONESIAN</strong> or <strong>ENGLISH</strong>. JOINS has P-ISSN: <a href="https://portal.issn.org/resource/ISSN/2528-0228" target="_blank" rel="noopener">2528-0228</a> and E-ISSN: <a href="https://portal.issn.org/resource/ISSN/2528-0236" target="_blank" rel="noopener">2528-0236</a>.</span></span></span></span></p> en-US <p dir="ltr">Authors who publish with JOINS (Journal of Information System) agree to the following terms:</p> <ol dir="ltr"> <li>Authors retain copyright and grant JOINS the right of first publication. The work is simultaneously licensed under a <a href="https://creativecommons.org/licenses/by/4.0/">Creative Commons Attribution 4.0 International License (CC BY 4.0)</a>, which allows others to share and adapt the work, provided they give appropriate credit to the authors and the initial publication in this journal, provide a link to the license, and indicate if changes were made.</li> <li>Authors may enter into separate, additional contractual arrangements for the non-exclusive distribution of the published version of the work (e.g., posting it to an institutional repository or publishing it in a book), with an acknowledgement of its initial publication in JOINS.</li> <li>Authors warrant that the manuscript is original, has not been published elsewhere, is not under review by another journal, and does not infringe the rights of any third party. Material from other sources must be properly cited, and permission obtained where required.</li> <li>Authors may post a preprint of the manuscript, but must inform the editor at the time of submission so that the double-blind review process can be handled appropriately.</li> </ol> jurnal@joins.dinus.ac.id (Editorial Joins) jurnal@joins.dinus.ac.id (Ramadhan Rakhmat Sani) Fri, 13 Nov 2026 00:00:00 +0000 OJS 3.2.1.4 http://blogs.law.harvard.edu/tech/rss 60 Transforming Work Processes in the Medium-Scale Garment Industry through an Integrated Task Management System https://publikasi.dinus.ac.id/joins/article/view/15678 <p><em>Work transformation has become a crucial necessity for medium-scale garment industries in addressing operational complexity and increasing demands for production timeliness. The garment industry often faces challenges related to interdepartmental coordination, delays in task completion, and limited monitoring of work progress by management due to the absence of an integrated digital task management system. This study aims to analyze work transformation in a medium-scale garment industry through the implementation of an integrated task management system. The research employs a Research and Development approach using the ADDIE model, which consists of the analysis, design, development, implementation, and evaluation stages, with a case study conducted at Tariz Konveksi in Jombang. The task management system was developed as a web-based application to support task allocation, interdepartmental coordination, and real-time monitoring of work progress. The results indicate that the implementation of an integrated task management system improves workflow organization, enhances transparency in interdepartmental coordination, and increases the effectiveness of work monitoring. This system contributes to minimizing miscommunication and task delays, thereby supporting a more structured and efficient work transformation in medium-scale garment industries.</em></p> Moh. Anshori Aris Widya, A'tini Hidayatika, Nur Khafidhoh Copyright (c) 2026 JOINS (Journal of Information System) https://creativecommons.org/licenses/by-sa/4.0 https://publikasi.dinus.ac.id/joins/article/view/15678 Thu, 15 Oct 2026 00:00:00 +0000 Design of an Internet of Things (IoT)-Based Automatic Irrigation System for Cassava Plants Using Fuzzy Logic and the MQTT Protocol https://publikasi.dinus.ac.id/joins/article/view/16265 <p><em>Cassava (Manihot esculenta Crantz) requires effective irrigation management to maintain soil moisture according to plant growth requirements. Manual irrigation is considered less efficient because it cannot adapt to environmental changes in real time and may lead to excessive water consumption. This study aimed to develop and implement an Internet of Things (IoT)-based automatic irrigation system that integrates the Fuzzy Logic Mamdani method with the Message Queuing Telemetry Transport (MQTT) protocol. The proposed system was designed around an ESP32 microcontroller integrated with Capacitive Soil Moisture Sensor V1.2, DHT22, BH1750, and YF-S401 water flow sensors. The evaluation results indicated that the proposed system successfully performed real-time environmental monitoring and automatic irrigation contro. Soil moisture values ranged from 33.2% to 84.8% with an average of 73.31%, air temperature ranged from 28.6°C to 29.4°C with an average of 28.96°C, air humidity ranged from 84.7% to 88.3% with an average of 86.48%, and light intensity ranged from 225.17 to 362.01 lux with an average of 303.05 lux. By applying the Fuzzy Logic Mamdani method, the system was able to adjust irrigation decisions according to changes in soil moisture and ambient temperature. In addition, the MQTT protocol provided lightweight, stable, and responsive data communication during system operation. Overall, the proposed system shows potential to enhance water-use efficiency and promote the adoption of smart farming practices in cassava cultivation</em></p> Ananda Adhi Surya, Alauddin Maulana Hirzan Copyright (c) 2026 JOINS (Journal of Information System) https://creativecommons.org/licenses/by-sa/4.0 https://publikasi.dinus.ac.id/joins/article/view/16265 Thu, 15 Oct 2026 00:00:00 +0000 Implementing ERP Digitalization in Supply Chain Strategy to Improve the Quality and Competitiveness of Culinary MSMEs https://publikasi.dinus.ac.id/joins/article/view/15505 <p>To improve the product quality and competitiveness of Roodfoodie eateries in Semarang, this study aims to develop a model for the integration of ERP digitalization and supply chain management. A synthesis matrix is used to assess seven relevant scholarly articles selected through the PRISMA diagram as part of the Systematic Literature Review (SLR) procedure. The study concludes that supply volatility, raw material price fluctuations, uneven product quality, and intense competition are the main challenges faced by culinary SMEs. According to the research findings, supply instability, raw material price fluctuations, inconsistent product quality, and stiff competition constitute the primary obstacles. Based on these findings, a three-level framework (upstream, internal, downstream) was developed. Three objectives arise from the quality improvement strategy based on the Analytical Hierarchy Process, contract agreements with suppliers, standard operating procedures, and the enhancement of human resource skills. Management dedication, stakeholder cooperation, quality standardization, and gradual technology implementation are crucial elements for success. By integrating perspectives of digitalization, resilience, and competitive advantage into a unified model that can be adopted by Indonesian culinary SMEs to improve operational effectiveness and sustainable competitiveness, this study makes a contribution.</p> Inas Salwa Nuraini, Zalfa Adinda Cahaya Arisyah, Ikhsani Dewi Rahmasari, Rochimatul Habibah, Asih Rohmani Copyright (c) 2026 JOINS (Journal of Information System) https://creativecommons.org/licenses/by-sa/4.0 https://publikasi.dinus.ac.id/joins/article/view/15505 Thu, 15 Oct 2026 00:00:00 +0000 Implementation of Data Mining for Classifying Indonesian Migrant Worker Complaints Using the Decision Tree Algorithm https://publikasi.dinus.ac.id/joins/article/view/16656 <p>The increasing number of Indonesian Migrant Worker (PMI) complaint records has made the process of data management and classification increasingly complex. Manual data processing is considered less effective because it requires more time for case identification and follow-up. This study aims to implement data mining techniques using the Decision Tree algorithm to classify PMI complaint data. The dataset used consists of 130 complaint records that have undergone the preprocessing stage. The classification process was carried out using RapidMiner, while model performance was evaluated using a confusion matrix based on accuracy, precision, and recall. The results show that the best model was obtained using the information gain criterion, a maximal depth of 3, and a minimal leaf size of 1, achieving an accuracy of 86,92%, precision of 81,94% for the PENDING class and 93,10% for the COMPLETED class, and recall of 93,65% for the PENDING class and 80,60% for the COMPLETED class. These findings indicate that the Decision Tree algorithm is capable of accurately classifying PMI complaint data, thereby supporting a faster, more structured, and more efficient complaint classification process.</p> Muhammad Ali Ridla , Sofiatun Mukarroma, Nargis Yahya Copyright (c) 2026 JOINS (Journal of Information System) https://creativecommons.org/licenses/by-sa/4.0 https://publikasi.dinus.ac.id/joins/article/view/16656 Thu, 15 Oct 2026 00:00:00 +0000 Classification of Environmental Materials: A Comparison of Random Forest, Decision Tree, and Support Vector Machine Algorithms https://publikasi.dinus.ac.id/joins/article/view/16747 <p><em>Soil, air, rocks, minerals, and metals have almost similar material characteristics, so they need to be classified. The classification process is carried out by developing and testing models using Decision Tree, Random Forest, and Support Vector Machine (SVM). The types of materials classified are Air, Metal, Mineral, Organic, Rock, Sediment, Soil, and Water. The dataset is taken from the Earth Intelligence Dataset (EID) with 5,000 samples and 42 features. The dataset contains chemical composition data, physical properties, and environmental data. Data processing uses missing value imputation, Z-score normalization, and coding labels. The data is split 80:20, the evaluation process uses accuracy, recall, and F1-score, while testing uses cross-validation. The results of the study obtained 98% accuracy for Decision Tree, 97% for Random Forest, and 94.60% for SVM. Mineral and rock classes have similar geological characteristics, making the classification process more complicated. There are 5 important features in this classification process, namely density, oxygen content, magnesium content, thermal properties, and calcium content. It is known that the Decision Tree achieved the highest accuracy value, with a standard deviation of 0.0036, indicating good model stability. This research contributes to the literature by providing several combinations of algorithms that have not been widely explored in the process of identifying environmental materials based on chemical composition and physical </em><em>properties.</em></p> Ellen Proborini, Fajar Husain Asy’ari, Ratna Herawati, Bijanto Bijanto Copyright (c) 2026 JOINS (Journal of Information System) https://creativecommons.org/licenses/by-sa/4.0 https://publikasi.dinus.ac.id/joins/article/view/16747 Thu, 12 Nov 2026 00:00:00 +0000 Application of the TOPSIS Method to Determine a Healthy Lifestyle Based on the Sleep Health and Lifestyle Dataset https://publikasi.dinus.ac.id/joins/article/view/15563 <p>Healthy lifestyle is an important factor affecting an individual's quality of life and productivity. Assessing a healthy lifestyle can be subjective because it involves multiple interrelated criteria, including sleep quality, sleep duration, physical activity, and other health indicators. This study aims to apply the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method to assess and rank individuals based on lifestyle and sleep-health indicators using the Sleep Health and Lifestyle dataset available on Kaggle. Three individuals were selected as alternatives, representing the Sleep Disorder categories of None, Insomnia, and Sleep Apnea. The criteria consisted of Quality of Sleep, Sleep Duration, Physical Activity Level, Stress Level, and Age, with respective weights of 0.30, 0.25, 0.20, 0.15, and 0.10. The TOPSIS procedure included decision matrix construction, normalization, weighting, determination of positive and negative ideal solutions, distance calculation, and preference value calculation. The results showed that A1 obtained the highest preference value of 1.0000, followed by A2 with 0.5709 and A3 with 0.1099. A1 ranked first because its distance from the positive ideal solution was zero. These results indicate that TOPSIS can rank alternatives based on multiple lifestyle and sleep-health indicators. However, the study was limited to three individuals, so the results cannot be generalized to the entire dataset.</p> Najma Clara Bella, Edifto Putra Leksmana, Muhammad Ghifar Itaqillah, Amiq Fahmi Copyright (c) 2026 JOINS (Journal of Information System) https://creativecommons.org/licenses/by-sa/4.0 https://publikasi.dinus.ac.id/joins/article/view/15563 Thu, 12 Nov 2026 00:00:00 +0000 A Multi-Criteria ELECTRE Approach for Stock Investment Decision-Making https://publikasi.dinus.ac.id/joins/article/view/15593 <p>The development of Indonesia’s capital market has increased public interest in stock investment, but selecting the right stocks remains challenging due to multiple financial criteria. This study aims to implement the Elimination Et Choix Traduisant La Réalité (ELECTRE) method as a Decision Support System (DSS) for stock selection in the banking industry. A quantitative approach using secondary data was applied to four stocks: Bank Mandiri (BMRI), Bank Rakyat Indonesia (BBRI), Bank Central Asia (BBCA), and Bank Negara Indonesia (BBNI). The criteria include Dividend Yield, Earnings per Share (EPS), and Return on Equity (ROE). The ELECTRE process involves normalization, weighting, concordance and discordance set formation, and alternative ranking. The results show that BBCA achieved the highest net concordance score of 16, followed by BBRI (4), BMRI (−4), and BBNI (−16). Based on average rank, BBCA and BBRI both scored 2, while BBNI scored 2.5 and BMRI scored 3.5. These findings indicate that ELECTRE can systematically compare stock alternatives based on multiple criteria. However, the tied rankings of BBCA and BBRI highlight the need for consistent final ranking rules. The results may serve as a reference for investment decisions but do not guarantee profits.</p> Valentino Arizona Putra, Najwa Jenar Aulia, Putri Rossa Ananta, Fara Anzhura Yudhitasari, Amiq Fahmi Copyright (c) 2026 JOINS (Journal of Information System) https://creativecommons.org/licenses/by-sa/4.0 https://publikasi.dinus.ac.id/joins/article/view/15593 Thu, 12 Nov 2026 00:00:00 +0000 Evaluation of Hyperparameter Optimization for Gaussian Naïve Bayes Using a Genetic Algorithm in Liver Disease Classification https://publikasi.dinus.ac.id/joins/article/view/17705 <p>Liver disease requires early detection to support appropriate treatment. This study evaluates the effect of Genetic Algorithm (GA)-based hyperparameter optimization on Gaussian Naïve Bayes (GNB) performance in liver disease classification. The dataset was the Indian Liver Patient Dataset (ILPD) from the UCI Machine Learning Repository, comprising 570 samples after preprocessing. Optimization was performed on the var_smoothing parameter within the range of to , using accuracy as the fitness function. The GA used a population of 10 individuals and five generations, while model evaluation employed 10-fold cross-validation. The baseline GNB achieved mean accuracy of 56.14%, precision of 96.60%, recall of 39.93%, and F1-score of 56.04%. After GA optimization, mean accuracy increased to 56.32%, precision remained at 96.60%, recall reached 40.18%, and F1-score reached 56.31%. These improvements were small, while the high number of false negatives indicated limitations in identifying liver disease samples. The results show that GA did not improve GNB performance substantially under the evaluated configuration and search range.</p> Nurjanna Nurjanna, Taghfirul Azhima Yoga Siswa, Muhammad Ridwan Copyright (c) 2026 JOINS (Journal of Information System) https://creativecommons.org/licenses/by-sa/4.0 https://publikasi.dinus.ac.id/joins/article/view/17705 Thu, 12 Nov 2026 00:00:00 +0000 The Impact of Implementing the Economic Order Quantity (EOQ) Method on Procurement and Inventory Strategies for Construction Materials: A Literature Review https://publikasi.dinus.ac.id/joins/article/view/15514 <p>Good inventory management is very important for building supply stores to maintain product availability and reduce operating costs. However, many small and medium-sized building supply stores still make purchases based on experience without a structured calculation method, putting them at risk of overstocking or understocking. This study aims to analyze procurement and inventory control strategies for building materials using the Economic Order Quantity (EOQ) method based on a literature review. The research method used is a Systematic Literature Review (SLR) of 5 relevant national journal articles. The results of the study show that the application of EOQ can determine the optimal order quantity, reduce storage and ordering costs, and increase stock availability when combined with reorder points and safety stock. In the context of building supply stores, EOQ is considered appropriate due to the characteristics of the products, which are heavy, relatively high in value, and have a fairly stable demand pattern. This study concludes that the EOQ method can be used as the basis for a more effective and efficient procurement and inventory control strategy in building supply stores.</p> Zhia Eggy Shahzida, Jolang Alif Khan, Nayoan Diaz Ahmad, Asih Rohmani Copyright (c) 2026 JOINS (Journal of Information System) https://creativecommons.org/licenses/by-sa/4.0 https://publikasi.dinus.ac.id/joins/article/view/15514 Thu, 12 Nov 2026 00:00:00 +0000 Hybrid Genetic Algorithm and Support Vector Machine for Hyperparameter Optimization in Liver Disease Classification https://publikasi.dinus.ac.id/joins/article/view/17703 <p>Liver disease remains a significant health concern in Indonesia, making early detection essential for reducing the risk of complications and improving treatment outcomes. Advances in machine learning have enabled the automated classification of diseases based on patients’ clinical data. Support Vector Machine (SVM) is a widely used classification algorithm; however, its performance can be affected by hyperparameter selection and class imbalance. This study employed the Indian Liver Patient Dataset (ILPD), which initially contained 583 records. Following data preprocessing, including data selection, cleaning, transformation, and standardization, 570 records remained after duplicate removal. Model performance was evaluated using stratified 10-fold cross-validation based on accuracy, precision, recall, and F1-score. Two classification approaches were compared: standard SVM and SVM optimized using a Genetic Algorithm (SVM-GA). The results showed that the SVM-GA model achieved an accuracy of 70.88%, precision of 72.14%, recall of 96.31%, and F1-score of 82.49%. In comparison, the standard SVM model achieved an accuracy of 70.18%. These findings indicate that GA-based hyperparameter optimization yielded a modest improvement in classification accuracy on the ILPD dataset, although improvements were not observed across all evaluation metrics.</p> Maliq Yazid Imani, Taghfirul Azhima Yoga Siswa, Muhammad Ridwan Copyright (c) 2026 JOINS (Journal of Information System) https://creativecommons.org/licenses/by-sa/4.0 https://publikasi.dinus.ac.id/joins/article/view/17703 Thu, 12 Nov 2026 00:00:00 +0000