Journal of Intelligent Decision Support System (IDSS)

NutriSee: A Rule-Based Mobile Application for Personalized Dietary Planning Using Indonesian Food Composition Data

2026-05-30T09:03:00+00:00

Nutritional problems in Indonesia show a concerning trend, with adult overweight and obesity prevalences reaching 14.4% and 23.4% (2023 Indonesian Health Survey). However, existing mobile nutrition applications commonly rely on static macronutrient ratios, lack transparent decision-making mechanisms, and do not incorporate localized Indonesian food composition data, limiting their practical relevance for Indonesian users. This study aims to develop NutriSee, a mobile application for determining dietary patterns based on Body Mass Index (BMI) classification. The application was built using the Rapid Application Development (RAD) method, Flutter framework, and Firebase Firestore. The system implements BMI calculation (WHO Asia-Pacific standards) alongside Basal Metabolic Rate and Total Daily Energy Expenditure calculations utilizing the Mifflin-St Jeor equation. A rule-based system generates five daily meal recommendations based on users' caloric and macronutrient targets, using the Indonesian Food Composition Table (TKPI 2020) database. Black Box Testing with Boundary Value Analysis and Equivalence Partitioning passed all 33 scenarios. User Acceptance Testing involving 30 respondents yielded "Very Good" acceptance rates of 89.33% for functionality and 84.40% for usability. These findings indicate that NutriSee functions effectively and remains accessible for practical use, enabling Indonesian adults to independently plan and manage their daily dietary intake based on localized food composition data

Multivariate Long Short-Term Memory (LSTM) Algorithm for Spatial-Temporal Agricultural Productivity Time Series Forecasting

2026-06-06T01:03:32+00:00

Accurate forecasting of crop productivity is fundamental to contemporary food security planning, yet conventional predictive models frequently underperform when confronted with the multivariate, spatial, and temporal intricacies inherent in agronomic datasets. This study presents a robust deep learning framework leveraging a multivariate Long Short-Term Memory (LSTM) network to forecast yields of principal food crops. The model was developed using a panel dataset from 12 districts in Chhattisgarh and Madhya Pradesh, India (2010–2017), comprising area, production, and yield observations for multiple competing crops. Rigorous preprocessing protocols included the application of separate StandardScalers to mitigate matrix inversion issues, and the derivation of land-allocation features to capture spatial interactions among crops. A lightweight LSTM architecture stabilized by gradient clipping was employed to enhance convergence and prevent exploding gradients. Empirical results demonstrate that the multivariate LSTM notably outperforms simple baseline estimators by effectively modeling non-linear relationships and district-level yield heterogeneity, attaining an RMSE of 494.70 Kg/ha and an R² of 0.8031. These findings suggest that spatial anthropogenic indicators—particularly the allocation of land across commodities—serve as informative proxies for reliable yield prediction in contexts lacking comprehensive weather-sensor data

Flood early warning monitoring system and water depth limit in the Curug Dhuwur river tourist area, Bumisari Village

2026-06-13T07:07:44+00:00

Waterfall tourist attractions are generally located in mountainous areas or at the foothills of mountains, far from residential settlements. Due to their geographical location, tourism managers face difficulties in monitoring environmental conditions and tourist activities. At Curug Dhuwur Waterfall, unpredictable weather changes can cause sudden fluctuations in water quality and water levels, posing safety risks to tourists due to unexpected changes in water flow. Previous studies have mainly focused on monitoring water levels and water quality without integrating tourist activity monitoring into a single system. Therefore, this study aims to design and implement an Internet of Things (IoT)-based early warning and tourist swimming activity monitoring system capable of monitoring water levels and tourist activities in real time to improve visitor safety and support disaster mitigation efforts. The system was developed using the prototyping method and consists of a NodeMCU ESP8266, an ultrasonic sensor, a turbidity sensor, and a laser receiver integrated with a website and Telegram platform. System functionality was evaluated using black-box testing, while user satisfaction was assessed using the System Usability Scale (SUS). The results indicate that the proposed IoT-based system can assist tourism managers in monitoring water conditions and visitor activities more effectively, with an ultrasonic sensor measurement accuracy of 98.33%. The SUS evaluation produced a score within Grade Scale D and the High Acceptability Range, indicating that the system is acceptable for use by tourism managers. In addition, the system successfully delivered monitoring information and warning notifications through both the website and Telegram platform according to the detected conditions. The implementation of the IoT-based monitoring system enables tourism managers to respond more quickly to potential flood hazards and unsafe swimming activities, thereby improving visitor safety and supporting disaster mitigation efforts.

Implementation of the Weighted Product Method in Decision Support Systems Selection of Outstanding Students

2026-06-19T09:15:53+00:00

The selection of outstanding students is an important process in education as it serves as the basis for awarding achievements and motivating students to improve both their academic and non-academic performance. However, the selection process in many schools is still carried out manually, which may lead to subjectivity and a lack of transparency in decision-making. This issue also occurs at SMA Plus Nurul Mubin, creating the need for a Decision Support System (DSS) that can assist the evaluation process in a more objective and systematic manner. This study aims to implement the Weighted Product (WP) method to determine outstanding students based on four criteria: Subject Grades, discipline, attendance, and extracurricular participation. The weight assigned to each criterion was determined based on the school's policies as the basis for the decision-making process. The research employed data collection methods including interviews, observations, and documentation. The collected data were then processed using the Weighted Product (WP) method through weight normalization, vector S calculation, and vector V calculation to obtain the final ranking results. The findings indicate that the Weighted Product (WP) method is capable of providing objective recommendations for outstanding students based on the preference value of each alternative. Based on the calculation results, alternative S3 obtained the highest vector V value of 0.176193538, ranking first as the outstanding student. It was followed by S6 with a value of 0.170283200, S5 with 0.168336098, S2 with 0.166745623, S1 with 0.160578981, and S4 with 0.157862560. These results demonstrate that the Weighted Product (WP) method is capable of producing an objective, effective, and transparent ranking of students based on the predetermined criteria weights. Therefore, the developed system can serve as a recommendation tool for schools to support the outstanding student selection process in a more accurate, structured, and accountable manner

Improving K-Means clustering performance on non-linear data using variance-weighted distance metrics

2026-06-15T14:12:32+00:00

K-Means is one of the most widely used clustering algorithms because of its simplicity and computational efficiency. However, its performance often decreases when handling non-linear data due to the assumption that all attributes contribute equally to the distance calculation process. This study proposes a Variance-Weighted Distance Metrics K-Means (VWDM-KMeans) method that assigns attribute weights based on variance values to improve clustering quality. The proposed approach consists of Min-Max Normalization, variance calculation, weight generation, and integration of variance-based weights into the distance metric used by K-Means. Experiments were conducted on a non-linear dataset containing 103 records and 3 attributes (x, y, and z) with K = 3 clusters. The generated attribute weights were 0.3207, 0.3342, and 0.3451 for attributes x, y, and z, respectively. The performance of VWDM-KMeans was compared with conventional K-Means and K-Medoids using the number of iterations, Sum of Squared Errors (SSE), and Silhouette Score (SS). The results showed that VWDM-KMeans converged in 5 iterations, compared to 6 iterations for K-Means and 3 iterations for K-Medoids. In terms of cluster compactness, VWDM-KMeans achieved the lowest SSE value of 2.7932, outperforming K-Means (8.2429) and K-Medoids (8.9602). Furthermore, VWDM-KMeans obtained a Silhouette Score of 0.4854, equal to K-Means and higher than K-Medoids (0.4696). These findings demonstrate that incorporating variance-based attribute weighting into the distance calculation process improves cluster compactness while maintaining cluster separation quality and stability. Therefore, VWDM-KMeans can serve as an effective and computationally efficient alternative for clustering non-linear data.

Implementation of augmented reality in an application to recognize traditional Papuan musical instruments as a medium for cultural preservation

2026-06-09T07:24:18+00:00

Traditional musical instruments of Papua represent a rich cultural heritage that faces a decline in public awareness due to limited access and physical preservation challenges. Most instruments are stored exclusively in cultural houses (Sanggar) or museums, isolating them from the younger generation who are deeply engaged with modern mobile technologies. This study aimed to develop and evaluate an Augmented Reality (AR)-based cultural preservation medium that improves public access to information on Papuan traditional musical instruments, specifically the Tifa, Tifa Titir, and Gong from the Fakfak region. The novelty of this study lies in its localized integration of three-dimensional visualization, authentic instrument sounds, and contextual cultural information into a mobile AR application designed for indigenous musical heritage preservation. Using a structured Waterfall development approach, cultural data were collected from local cultural custodians and transformed into interactive digital content. The application projects realistic 3D assets, plays authentic instrument sounds, and displays textual cultural information upon scanning unique physical markers. Black-box testing validated that all system features, including marker tracking, audio triggers, and page navigation, functioned successfully across various Android hardware environments. User satisfaction evaluations yielded an overall cumulative usability index of 90.35%, classifying the application as highly effective and user-friendly. This digital intervention bridges the gap between historical preservation and modern education, providing an accessible, interactive, and community-oriented model for cultural digitization and learning

Maturity And Governance Gap Analysis Of Simgos At Royal Maternity General Hospital Using COBIT 5

2026-05-29T07:18:54+00:00

RSU Royal Maternity is a type C healthcare institution in Medan with full accreditation and is committed to high-quality service standards. To support the achievement of organizational objectives, Information Technology (IT) has become a strategic asset, enhancing efficiency, transparency, and service quality. However, implementing IT without proper management processes risks such as business needs disparities, reduced service quality, and system failures. This study aims to evaluate the information system governance at RSU Royal Maternity to ensure that IT investments deliver maximum added value. The research methodology uses the COBIT 5 framework, a comprehensive approach to IT governance. The analysis focuses on measuring maturity, with an emphasis on the Deliver, Service, and Support (DSS) and Monitor, Evaluate, and Assess (MEA) domains. This focus is intended to ensure that the operations of the healthcare support system align with Standard Operating Procedures (SOPs) and professional codes of ethics. The results of this study are expected to provide recommendations for service quality improvements that will proportionally enhance patient satisfaction at Royal Maternity General Hospital.

Escalation-Aware Governance Framework for Autonomous Warfare Decision Support Systems

2026-07-01T13:48:23+00:00

Artificial intelligence increasingly transforms military decision-making through autonomous targeting, predictive battlefield analytics, cyber operations, and algorithm-assisted command systems. Although these technologies improve operational speed, they create governance risks when machine-speed decisions exceed human oversight, legal accountability, and escalation control. This study analyzes governance asymmetry in AI-driven autonomous warfare decision support systems and develops an escalation-aware governance framework. A structured qualitative review using transparent search procedures and thematic synthesis was conducted across 33 peer-reviewed and institutional sources on military AI, autonomous weapons, cybersecurity, and AI governance. The synthesis reveals three interconnected governance failures governance asymmetry, symbolic human control, and escalation compression that reinforce one another by weakening accountability, reducing substantive human judgment, and increasing strategic instability. The study proposes a human-centered escalation-aware governance framework integrating eight operational control layers: human authorization checkpoints, explainability, auditability, cybersecurity resilience, cognitive security, escalation-control mechanisms, emergency override, and post-action review. The framework contributes to intelligent decision support literature by translating autonomous warfare risks into embedded governance layers for preserving accountability and strategic stability

Decision Support System for Selecting Outstanding Students Using a Combination of AHP and SAW Methods

2026-06-26T06:37:48+00:00

The process of selecting outstanding students at MA Badrul Islam Gulung is still done conventionally, mainly by considering report card grades as the key factor, which could lead to subjectivity and be less effective in decision making. This study aims to develop a decision support system for selecting outstanding students using a combination of the Analytical Hierarchy Process (AHP) and Simple Additive Weighting (SAW) methods. AHP is used to determine the priority weight of each criterion, while SAW is used to rank students based on preference scores. Data was collected through observation, interviews, questionnaires, and literature studies, involving 10 respondents and 5 student alternatives as research subjects. This study uses six assessment criteria: academic grades, non-academic achievements, behavior, discipline, activeness, and memorization of the Qur’an. The research results show that the combination of the AHP and SAW methods can produce a more objective, transparent, and systematic evaluation process. Based on the calculations, alternative A3, named Oktaviana, scored the highest at 0.991 and was chosen as the outstanding student because she had excellent and consistent scores across almost all assessment criteria, especially in academic performance, behavior, and Qur'an memorization. Thus, the decision support system that was developed can help the school determine outstanding students more accurately and efficiently

Keyboard circuit trace restoration using conductive silver paste for computer input devices

2026-06-10T12:38:47+00:00

Computer keyboards serve as essential input devices that translate user actions into digital signals. One common failure mechanism in keyboard membranes involves broken or degraded circuit traces, which leads to non-functional keys. This study investigates the restoration of such traces using conductive silver paste. An experimental approach was adopted, involving the repair of damaged traces on flexible membrane keyboard PCBs, followed by electrical testing and durability observation. The measurement outcomes indicate that silver paste offers approximately 7% higher electrical conductivity relative to copper, with measured resistivity at 15.87 nΩ·m and thermal conductivity reaching 430 W/m·K. Open circuit conditions (infinite resistance) were successfully restored to functional conductive paths showing 3.5 ± 0.8 Ω resistance after paste application and controlled heat curing. The post-repair resistance represents only a 25% increase compared to the original intact trace value of 2.8 ± 0.5 Ω, which remains well within the acceptable range for digital keyboard applications. This research contributes practical guidance for computer hardware maintenance personnel and researchers focused on input device reliability, while also supporting sustainable practices through repair-based lifecycle extension

Computational intelligence for solar photovoltaic power plant monitoring and fault diagnosis: a machine learning approach

2026-06-20T12:45:59+00:00

Solar photovoltaic (PV) power plants are increasingly deployed in tropical regions such as Indonesia, yet their performance is often degraded by undetected faults including partial shading, dust accumulation, and module mismatch. This study presents a computational intelligence framework for real-time monitoring and fault diagnosis of grid-connected PV systems from a computer science perspective. The framework consists of three main components: (1) a data acquisition module that simulates 12 months of PV system operation (25 kWp capacity) using meteorological data from Medan, Indonesia, generating 8,760 hourly samples of voltage, current, power, irradiance, and temperature; (2) a machine learning-based fault classifier using Random Forest (RF) and Support Vector Machine (SVM) algorithms to distinguish between four fault types (normal operation, partial shading, dust accumulation, and module mismatch) and one healthy state; and (3) a web-based dashboard built with PHP and MySQL for real-time visualization and alerting. Experimental results show that the Random Forest classifier achieves 97.3% accuracy, 95.8% precision, and 96.2% recall, outperforming SVM (91.6% accuracy). The algorithm detects faults within 1.8 seconds of occurrence, enabling rapid operator response. The proposed system is implemented as an open-source prototype and can be deployed on low-cost hardware (Raspberry Pi 4) with an average response time of 1.8 seconds. The framework is validated using tropical climate data from Medan, Indonesia, addressing a gap in existing PV fault diagnosis research. This research contributes a practical, software-based fault diagnosis tool for PV system operators in tropical environments