Journal Sensi: Strategic of Education in Information System

Identify Traffic Flow Volume using Image Processing for Intelligent Transport System

Fitria Claudya Lahinta — 2025-08-31

Detection of traffic flow volume is needed to get information about the number of vehicles on a street section. This information is required for traffic control. This study uses video data taken from four adjution street section in Manado City. Video data are processed using the background subtraction method in the segmentation process. Then morphological operation techniques are applied to improve segmentation results. The binarization method with a threshold value of 220 to eliminate shadows on vehicle objects. Vehicle shadows must be removed because it can reduce system accuracy especially if vehicle shadows are too large and connected to other vehicles. Accuracy results from vehicle detection are 95.04% for the Toar street, 94.62% for the Diponegoro street, 91.47% for the Lumimuut street and 95.18% for the 14 Februari street. Vehicle detection results will be calculated to get the number of vehicles then divided by the duration of observation time to get the traffic flow volume. The results of the traffic flow volume information are expected to be implemented on Smart Traffic Light.

Logistic Regression and TF-IDF Models for Early Detection of Indonesian-Language Fake News

Abdul Hamid Arribathi — 2025-08-31

The spread of false information, particularly on social media platforms, has become a significant challenge due to the rapid flow of digital content. Most existing fake news detection systems are primarily designed for the English language, making them less effective when applied to Indonesian contexts. This study proposes a web-based hoax detection system that combines the Term Frequency-Inverse Document Frequency (TF-IDF) method with the Logistic Regression classification algorithm. The dataset, consisting of both real and fake news articles, was obtained from Kaggle and processed through several stages including normalization, stopword removal, and stemming. The TF-IDF vectorization results were then used to train a binary classification model. The system allows for user input either in the form of raw text or a URL, and delivers real-time classification results. Evaluation of the system indicates a high level of accuracy and strong potential in improving public digital literacy. These findings demonstrate a lightweight yet effective approach to mitigating the spread of hoaxes in the Indonesian language.

Performance Evaluation of ARIMA and LSTM Models with Product Inventory Demand in Production Companies

Aris Martono — 2025-08-31

This study aims to evaluate and compare the performance of two time series forecasting approaches: the classical statistical ARIMA model and the deep learning-based LSTM model, in the context of forecasting product inventory demand in a production company. The data used consists of historical daily demand records, totaling 100 and 200 records, which were analyzed to identify linear and non linear patterns. The ARIMA model was selected for its reliability in modeling stationary and seasonal data, while the LSTM model was utilized to capture complex temporal patterns through its layered neural network architecture. The test results using the MSE and RMSE metrics show that in both datasets, the ARIMA model has better prediction performance (100 records, RMSE=45.61% and 200 records, RMSE=44.72%) compared to LSTM, namely 100 records, RMSE=45.93% and 200 records, RMSE=49.54%. Although LSTM excels in handling non-linear dynamics, ARIMA outperformed it on data with linear. This study highlights the importance of selecting forecasting models based on data characteristics and suggests opportunities for future exploration of hybrid models. The theoretical and empirical foundations of this research are supported by the works of Hyndman & Athanasopoulos (2018), Hochreiter & Schmidhuber (1997), and Makridakis et al. (2018), which provide critical insight into predictive modeling for time series analysis.

Decision Support System for Employee Bonus Recommendation Using Fuzzy Logic

Alfian Ardhiansyah — 2025-08-31

A decision support system is indeed something that should be used to make it easier for organizations to determine a policy. With the existence of information technology, all data analysis and calculation is carried out automatically through computers. Similarly, in making recommendations to give bonuses to an employee in a company or institution. To speed up the decision-making process, a system is needed that can provide recommendations like calculations made by human intelligence. The system was developed using the fuzzy logic method that expresses classical logic into linguistic forms. The advantage offered by this logic is that it produces a more just and humane decision such as a decision that results from human feelings and thoughts. This system uses four variables used to determine the receipt of bonus wages, namely the age of the employee, the length of service, the amount of salary and productivity in one month. Each of these variables has a linguistic variable that is used to represent a certain state or condition that utilizes natural language. This research produces a system that can provide recommendations for organizations or companies to use in determining the receipt of bonus wages in accordance with the rules applied.

Application of Material Judgment Monitoring The TBR Plant Division Uses RFID and MES (Manufacturing Execution System) Methods

Fitri Andriyani — 2025-08-31

With great advances in the world of information technology in the industrial era 4.0, efficiency and accuracy in the production process are crucial factors to increase the company's competitiveness. One of the important aspects is material monitoring, especially in the material judgement process in the TBR Plant (Truck and Bus Radial) Division, which requires an information system that supports the material assessment, starting from the hold process, judgement, making reports on the results of the TBR judging plant monitoring to superiors. This study aims to design and develop a material judgement monitoring application based on RFID (Radio Frequency Identification) technology integrated with the MES (Manufacturing Execution System) system to unravel errors in data control, inaccurate reports, and delays in finding the required data are all the results of manual systems. With the application of RFID, every material can be tracked in real time from the inspection process to the production line. Integration with MES enables data synchronization between production systems and enterprise information systems, supporting fast and accurate decision-making. The results of the implementation show an increase in the efficiency of material recording time, the accuracy of data judgement, and the reduction of human error in the tracking process. This application is expected to be an effective digital solution in supporting a more transparent, efficient, and digitized production system in the manufacturing industry environment.

Smart Tour Travel Purwokerto's Service Quality Improvement Strategy for Customer Satisfaction

Ali Nurdin — 2025-08-31

This research discusses strategies for improving the quality of Smart Tour Travel Purwokerto services in an effort to increase customer satisfaction. The research method used is a qualitative approach with a case study design. Data was collected through interviews, direct observation, and analysis of internal company documents. The research results show that the main factors in increasing customer satisfaction include service speed, staff privacy, travel comfort, and transportation safety. The strategies implemented include employee training to increase professionalism, development of an application-based ordering system, and rejuvenation of the vehicle fleet. Survey results showed a significant increase in customer satisfaction after this strategy was implemented. Improving service quality not only has an impact on customer loyalty, but also strengthens the company's competitiveness in the travel industry. Therefore, innovation and continuous improvement are key in maintaining customer satisfaction and business desires.

An ESP32-Based IoT System for Blind Spot Detection in Logistics Vehicles

Ilamsyah Ilamsyah — 2025-08-31

The increasing number of vehicles in the logistics industry has led to a higher risk of accidents due to vehicle blind spots. PT. Aerofood ACS requires a real-time, cost-effective, and easily deployable monitoring system to enhance the safety and operational efficiency of their EBRO trucks. To address this need, this research develops an Internet of Things (IoT)-based blind spot monitoring system using ESP32. The device design methodology includes a prototyping approach and data collection through observation, interviews with stakeholders, and literature studies in system development. This system utilizes HC-SR04 ultrasonic sensors to detect objects in blind spot areas, where the obtained data is processed by ESP32 and displayed on an LCD. Additionally, information is transmitted to the cloud for real-time monitoring. The primary objective of this study is to design and implement a monitoring system capable of providing early warnings to drivers to reduce accident risks. Alerts are provided through LED indicators, buzzers, and IoT notifications, ensuring that drivers can promptly respond to potential hazards. The research findings indicate that the developed system can accurately detect blind spots and effectively issue warnings. Thus, this system serves as an innovative solution that can be implemented for various logistics needs in compliance with PT. Aerofood ACS's safety standards.

Implementation of RevoU Tech Academy's Independent Campus Program to Increase Digital Talent Capacity

Erna Astriyani — 2025-08-31

The Merdeka Belajar Kampus Merdeka (MBKM) program aims to enrich students' hard and soft skills, preparing them better for the ever-changing challenges of the times. One of the independent study programs is RevoU Tech Academy "Learn Data Analytics & Software Development With AI" which focuses on developing skills in the fields of Data Analytics and Software Engineering. This journal describes the implementation of the RevoU Tech Academy program and its effects on participants using a qualitative descriptive approach. The program is conducted for 5 months, with a 17-week curriculum that covers Data Processing in Data Analytics, Data Visualization & Communication in Data Analytics with AI Tools, Introduction to Web Development, Front-End Development with AI Tools, and Career Development program. Participants are not only given lectures but also practice their knowledge through various assignments and a capstone project. The results of the study showed that the program was successful in developing participants' skills in Data Analytics and Software Engineering. They were able to clean data, analyze data, build visualizations, develop web applications.

Web-Based Slaughterhouse Monitoring Information System Using ERD Method

Eka Purnama Harahap — 2025-08-31

The advancement of information technology drives the livestock sector, especially slaughterhouses (RPH), to adopt digital solutions. This study develops a web-based RPH monitoring information system to support real-time recording, animal health monitoring, scheduling, and reporting. Data collection was conducted through interviews and literature studies, while the database design applied the Entity Relationship Diagram (ERD) method to ensure organized and scalable data structures. The system was developed using PHP and MySQL and is accessible to both operational staff and relevant institutions. The results show that the system improves efficiency, recording accuracy, and operational transparency of slaughterhouses. Additionally, it enhances public trust and facilitates data-driven decision-making.

Design and Construction of an IoT-Based Automatic Fish Feeder Using Telegram Bot

Ahmad Roihan — 2025-08-31

This research presents the design and development of an automatic fish feeder based on the Internet of Things (IoT), integrated with the Telegram application. The device is intended to assist fish farmers, particularly those farming catfish, in feeding their fish automatically or remotely while away. The research adopts the Research and Development (R&D) method, utilizing the Software Development Life Cycle (SDLC) Waterfall model in the system development process. The stages include requirement analysis, system design, implementation, testing, and maintenance. The system is capable of dispensing feed automatically based on a predefined schedule using an RTC module, and it enables control and monitoring through commands sent via a Telegram bot. Testing results indicate that the system operates effectively, enhancing the efficiency and accuracy of fish feeding. The implementation of this device is expected to serve as an innovative solution to support productivity and sustainability in modern catfish farming practices.