Internet of Things and Artificial Intelligence Journal

Implementation of Real-Time Reservoir Water Turbidity Monitoring System with Automatic Alarm and Control Using Blynk

Sasa Agustina, Muhammad Yahya, Jumadi Mabe Parenreng — Fri, 01 Aug 2025 00:00:00 +0000

The quality of reservoir water can deteriorate due to the accumulation of particles such as mud, sand, and microorganisms, making the water cloudy and unusable. This condition is difficult to monitor directly, especially in reservoirs located at high altitudes. This research developed a real-time IoT-based reservoir water turbidity monitoring system equipped with alarms and automatic drain features through an application. The system is designed using the ESP32 microcontroller, the SEN-0175 turbidity sensor, the HC-SR04 ultrasonic sensor, the buzzer, the DN20 solenoid valve, and the Blynk platform. The turbidity sensor is capable of detecting increased turbidity from 2 NTU to 50 NTU. The ultrasonic sensor shows high accuracy with an average error of 0.41%. The buzzer activates automatically when the turbidity exceeds 25 NTU. The solenoid valve responds to application commands within 2 seconds. The app can display real-time data, notify, and control drainage. This system has been proven to be effective in monitoring water quality, providing early warnings, and facilitating the process of draining reservoirs.

Prototype of a Solar Panel Voltage Monitoring Tool Using IoT

Fri, 01 Aug 2025 00:00:00 +0000

The use of solar energy as a renewable energy source continues to grow, but many solar panel systems are still not equipped with real-time performance monitoring tools. This research aims to design and implement an Internet of Things (IoT)-based solar panel monitoring system using a NodeMCU microcontroller and the ThingSpeak platform. This system is capable of monitoring electrical parameters such as the voltage generated by solar panels, then periodically sending the data to the ThingSpeak server via a WiFi connection to be displayed in the form of graphs that can be accessed online. The INA219 sensor is used to measure voltage and current values directly, while the NodeMCU functions as a data processor and sends information to the cloud. Test results show that the system can work well and stably in displaying monitoring data in real-time, and allows users to monitor the condition of solar panels remotely easily. This system is considered effective, simple, and economical as a solar energy monitoring solution, especially for small to medium-scale needs. The results of the experiments conducted on the first day showed an average voltage of 15.52 V. On the second day, the average voltage was 15.05 V, and on the third day, it was 14.99 V, with productive hours from morning to evening.

Smart Barcode-Based Sorting Using YOLO and Automatic Conveyor Belts

Fawwas Aydin Rafif, Evelina Ginting, Johansyah Al Rasyid — Mon, 28 Jul 2025 00:00:00 +0000

Automatic sorting of goods is an effective solution for improving efficiency and accuracy in distribution and logistics processes. This study developed an automatic barcode-based sorting system that utilizes cameras and image processing technology using Raspberry Pi and actuator control via Arduino. The camera is used to capture images of barcodes on objects moving on a conveyor belt, which are then processed by a Raspberry Pi to detect and read barcode data. The read data is sent to the Arduino to activate actuators such as DC motors and servos, which direct the objects to the appropriate sorting path. Proximity sensors are used to detect the presence of objects before scanning begins. Test results show that the system is capable of performing automatic sorting with a high success rate and quick response. This system has great potential for implementation in industrial production lines to replace manual sorting processes, which still frequently result in errors.

The Comparison of the Performance of Telegram and Blynk as Monitoring Media on the Prototype of Internet of Things-Based Soybean Planting System

Fri, 01 Aug 2025 00:00:00 +0000

Indonesia is the country with the second-largest soybean consumption in the world after China. Meanwhile, to meet the needs, Indonesia still depends on imports since the soybean production has still been under the national demand. Therefore, the Indonesian government has included soybeans in Prioritas Riset Nasional. Agriculture in Indonesia is able to use IoT to increase soybean production; therefore, this research has built a prototype of an IoT-based soybean planting system with a soil humidity sensor and network time protocol (NTP) as the tools for automation. Some elements essential to be observed were informed to the user through Telegram and Blynk applications. This study ran well as the humidity sensor NTP timer could control the watering and fertilizing system, and notifications could be sent to the user. From a QoS standpoint, Blynk exhibits a delay of 62 ms, while that of Telegram was 59 ms. Regarding throughput metrics, Blynk’s performance was nearly equivalent to that of Telegram.

Design and Development of a Wireless Technology-Based Automatic Lighting System and IoT at Madrasah Aliyah Bayyinul Ulum Santong

Hendra Lalu Syamsul Mahendra, Emi Suryadi, Muh. Nasirudin Karim — Mon, 28 Jul 2025 00:00:00 +0000

Madrasah Aliyah or MA Bayyinul Ulum Santong is one of the Madrasahs or High Schools that still uses manual switches for classrooms, where the switches used are in each class of 12. The use of this manual switch can take a long time to turn on and off the lights in each classroom. This research designs and builds a light automation system that can reduce the amount of time needed to turn on and off the lights. This system is made using two switch methods, namely a virtual manual switch and a time switch, where this time switch can turn on and turn off the lights as a whole according to the time specified by the operator. In addition to the time switch, there is also a manual switch that can be used to turn on and off the lights one by one. This system can help reduce the use of electricity, which is usually wasteful because of negligence in forgetting to turn off the lights. With this system, the finished lamp can always be monitored by the operator. This system is designed from the arrival of the main tools, namely NodeMCU as a microcontroller, RTCDS3231 as a timer, and Relay as an ON/OFF switch. With the IoT-based real-time monitoring system using NodeMCU, a flexible monitoring and control system will be created.

Classification of Oil Palm Leaf Diseases Using YOLOv8-Nano Algorithm

Enda Putri Atika, A.Salky Maulana — Fri, 01 Aug 2025 00:00:00 +0000

Early detection of diseases in oil palm leaves is crucial to prevent a decline in productivity and to maintain the quality of crop yields. This study aims to develop an automatic classification model for oil palm leaf images using the YOLOv8-Nano algorithm. The dataset used consists of three classes—Healthy, Fungal, and Brown Spot—which were divided into training, validation, and testing sets with a ratio of 80:10:10. The training process was conducted over 10 epochs using image dimensions of 224×224 pixels, leveraging pretrained weights from YOLOv8n-cls. Evaluation results show that the model was able to classify the images perfectly, achieving 100% in accuracy, precision, recall, and F1-score. These findings indicate that YOLOv8-Nano is a lightweight yet highly effective algorithm for the classification task of oil palm leaf images. However, further testing with field data is necessary to ensure the model’s generalization ability in real-world scenarios.

Segmentation of Circular Economy Adoption in East Java-Indonesia Based on Barriers and Motivations Using K-Means and Multilayer Perceptron

Fri, 01 Aug 2025 00:00:00 +0000

The Circular economy is becoming increasingly relevant in addressing global challenges related to sustainability and natural resource management. While globally recognized, its implementation in East Java faces significant barriers, such as limited understanding, inadequate infrastructure, cultural resistance, and insufficient involvement from both the industrial sector and the public. This study aims to fill this gap by segmenting circular economy adoption in East Java based on motivations and barriers. Segmentation uses the K-Means algorithm combined with the Multilayer Perceptron (MLP) model. The analysis identifies three clusters: (1) highly motivated and proactive individuals, (2) moderately aware but less engaged individuals, and (3) individuals constrained by barriers and passive. The MLP model with 300 iterations delivered the best performance, achieving 92% accuracy, along with high precision and recall across all clusters. Chi-Square testing indicates that access to recycling, government support, and economic incentives significantly influence cluster formation, while product discounts and waste quantity have minimal impact. These findings provide insights for policymakers to design strategies to promote circular economy adoption, confirming that MLP is an effective tool for supporting.

LSTM-Based Forex Trading Bot Using Python and MetaTrader 5: Design, Simulation, and Evaluation

Reymark-John Macapanas, Mary Ann Gliefen Bermudo — Fri, 01 Aug 2025 00:00:00 +0000

This paper presents the development of an AI-driven forex trading bot that utilizes a Long Short-Term Memory (LSTM) neural network to forecast short-term price movements and automate trading decisions. The objective of the study is to create a scalable, data-driven system capable of improving trade accuracy using historical USD-JPY price data in conjunction with the MetaTrader 5 platform. The proposed system integrates a time-series preprocessing pipeline, LSTM-based price prediction, and a logic-driven trade simulation model to assess performance under controlled conditions. The model achieved a directional accuracy of 88.4%, a profit accuracy of 78%, and a cumulative simulated profit of USD 797.50 over 100 trades. Additionally, training and validation losses stabilized after 50 epochs, indicating effective learning without overfitting. Visual comparisons between actual and predicted prices further validated the model’s forecasting ability. The results highlight the potential of LSTM models to support intelligent financial automation and provide a foundation for future enhancements, including real-time deployment and hybrid AI-based trading strategies.

Analysis of the Role of Organizational Culture and Innovation on Adaptability of the West Sumatra Provincial Education Office

Rika Rohim, Tuti Susanti, Rahmat Zainul Rezki — Mon, 28 Jul 2025 00:00:00 +0000

Rapid changes in the environment of public organizations demand a work culture that is able to encourage innovation and increase adaptability. In the context of the education bureaucracy, the ability to innovate and adapt is the key to successful institutional reform. This study aims to analyze the effect of organizational culture on employee innovation and adaptability in the Education Office of West Sumatra Province. This research uses a quantitative approach with Partial Least Squares-Structural Equation Modeling (PLS-SEM) analysis technique processed using SmartPLS software. Data were obtained from 100 respondents through a five-point Likert scale questionnaire, with indicators developed based on relevant theories. The instrument was tested through validity and reliability tests using the loading factor value, AVE, composite reliability, and Cronbach's alpha. The results showed that organizational culture has a significant effect on adaptability, while innovation also has a significant effect but with a smaller coefficient. A strong, participatory, and adaptive organizational culture is proven to encourage organizational readiness in the face of change. This finding confirms the importance of integration between organizational culture and innovation in shaping adaptive public organizations. One form of Digital Adaptation is also shown by the use of Artificial Intelligence which is applied in various forms of tasks and obligations that need to be completed.

Design and Implementation of Augmented Reality-Based Housing Promotion Media for Cipta Bangun Khatulistiwa Ltd.

Fri, 01 Aug 2025 00:00:00 +0000

Information media continues to evolve with the advancement of time. It is needed at all times because through information media, people can access various types of current and emerging information, as well as exchange ideas and interact with one another. Numerous technologies have been developed to support the dissemination of information, one of which is Augmented Reality (AR). Augmented Reality (AR) can be defined as an environment that merges two realms: the virtual world and the real world. AR allows information to be embedded into the virtual world and displayed in the real-world using markers or barcodes via smartphones or other devices. Augmented Reality (AR) has been widely implemented across various sectors, one of which is promotional media. In the business world, AR can serve as a powerful promotional tool. Therefore, with the advancement of Augmented Reality (AR) technology, a promotional media application was developed using AR for the housing brochures of Cipta Bangun Khatulistiwa Ltd. This AR-based promotional media is expected to assist current and prospective consumers in gaining a more realistic view of Cipta Bangun Khatulistiwa Ltd. housing details through specially designed brochures, without the need to travel to the development site, which is relatively far from the city center. From the results of the system test, 80% mastery and very good assessment were obtained.

Smart Home Monitoring House Fence Using Face Recognition Based On The IoT

Ega Nurwani Faisal, Jumadi Mabe Parenreng , Muliadi Muliadi — Fri, 01 Aug 2025 00:00:00 +0000

Technological developments have become very important in modern life, including the security sector. Currently, there is more and more sophisticated equipment and security systems based on the latest technology; the increasing crime rate, especially theft and robbery, encourages the need for a more effective and efficient security system. This research aims to build a prototype for smart home monitoring of home devices, namely house fences, using facial recognition based on the Internet of Things. The dataset required in this research is 11,500 facial images in 5 categories. Training of the machining learning model using a convolutional neural network was carried out several times to produce a model with the best accuracy. The test was carried out on 122 samples and produced an accuracy value of 86% and an average telegram response of 7 seconds, so that it could monitor house fences in real time.

Personalized Facial Wrinkle Distribution Analysis Using Backpropagation Neural Network (BPNN)

Magfiratul Jannah, Desi Anggreani, Muhyiddin A. M Hayat — Fri, 01 Aug 2025 00:00:00 +0000

Facial wrinkle distribution is an important indicator of aging and lifestyle. This study proposes a personalized wrinkle classification system using a Backpropagation Neural Network (BPNN) based on segmented facial areas such as the forehead, eyes, cheeks, and mouth. After preprocessing and feature extraction, the BPNN model is trained to classify wrinkle severity into two categories: high and medium. Evaluation results show that the model performs well, particularly in detecting the Medium class, achieving a precision of 0.8438 and a recall of 0.9310, while for the High class, the precision is 0.8333 and the recall is 0.6667. These findings indicate that the BPNN architecture is effective and reliable in facial wrinkle classification, with potential applications in dermatology, cosmetic analysis, and digital forensics.

Enhancing Network Security and Scalability through RADIUS and IP Segmentation

Fri, 01 Aug 2025 00:00:00 +0000

The rapid expansion of information technology necessitates effective methods for managing network security and scalability. This paper explores the integration of Remote Authentication Dial-In User Service (RADIUS) for Authentication, Authorization, and Accounting (AAA), and IP segmentation to optimize network performance and security. By analyzing the case study at the company, we investigate the practical application of these technologies to enhance user access management and address the growing need for IP address resources. The results show that RADIUS effectively manages network access, while IP segmentation ensures the network can handle increasing traffic without compromising security.

Early Detection of Disease Outbreaks: A Monitoring System for Sleman Regency

M. Tsana'uddin Farid, Rifqi Anugrah, Rickhy Artha Octaviyana — Fri, 01 Aug 2025 00:00:00 +0000

In early 2025, Indonesia faced a surge in infectious diseases, with 6,000 Dengue Fever and 28 deaths by January, and 889,000 tuberculosis cases by March. An outbreak or “Kejadian Luar Biasa” (KLB) is marked by a significant rise in illness or death within a period. However, according to the Sleman Regency Health Office, the existing Early Warning and Response System remains suboptimal in detecting such events. To improve early detection, a Disease Outbreak Monitoring System was developed using the Waterfall Method. This system features an interactive dashboard, data storage for patients and health centers, automatic KLB detection, and a feedback mechanism. Testing has demonstrated that the system improves accuracy and responsiveness, providing a promising solution for early outbreak detection and prevention.

Simple Math Learning Application for Children Aged 7–11 Using Scratch

Indah Tri Handayani — Mon, 28 Jul 2025 00:00:00 +0000

Mathematics cannot be separated from numbers, counting, reasoning, and so on. This application is made using Scratch, which is open source. The learning method begins to recognize abstract numbers. The author also makes learning media using computers, which are expected to be more fun. The research method used for writing this scientific research is the SDLC (Software Development Life Cycle) method. Learning Application to Solve Simple Math Problems Using Scratch has been completed, and children aged 7 to 11 years can increase their interest in learning about simple math. This application contains 10 questions about simple math problems. Based on the results of testing using a black box, it can be concluded that this application functions properly following the planning and design that have been made. Development suggestions that can be done in this scientific writing application are adding levels or levels from the easiest to the most difficult, animated characters that can move, and adding other interesting features so that users are more entertained when using this application.