Publication Search

This research aims to develop a multiclass flower image classification system using the Convolutional Neural Network (CNN) algorithm with the EfficientNet architecture. The main problem addressed is the difficulty of manual identification of flower species that share high visual similarity. The research stages include collecting 17,299 flower images across 19 classes, performing data preprocessing such as image resizing, pixel normalization, and augmentation, followed by model training using the EfficientNet transfer learning approach. The model was trained for 10 epochs with an 80:20 training-validation data split. The evaluation results show that the model achieved a validation accuracy of 98.05% with a loss value of 0.0968, and an average precision, recall, and F1-score of 0.98. The trained model was then implemented into a web-based application built using the Next.js framework, enabling users to upload flower images and obtain real-time classification results via the Hugging Face API. The system successfully identified flower species with a confidence level of 99.87%. These findings demonstrate that combining a modern CNN architecture with transfer learning provides efficient and highly accurate flower classification performance, which can be effectively implemented for educational and digital conservation purposes.

Waste has become a serious environmental problem in Indonesia, which continues to increase along with population growth. The issue of waste management poses serious challenges for the environment, especially in the process of separating organic and inorganic waste. In the field of computer vision, recognising the type and shape of waste through camera images remains a challenge due to variations in shape, colour, and complex lighting conditions. Therefore, this problem utilises Deep Learning technology, which is expected to be widely applied in Indonesia, especially in large cities with high waste volumes. This study aims to distinguish between organic and inorganic waste using the Convolutional Neural Network (CNN) method based on digital images. The developed CNN model was trained to recognise the visual patterns of each type of waste and tested to measure its accuracy. The test results show that the CNN-based classification system is capable of achieving an accuracy rate of 95%, thus proving the effectiveness of this method in supporting artificial intelligence-based automatic waste sorting systems.

The problems in rice fields are complex and varied, depending on geographic location, rice variety, and growing season. Pests often cause serious economic losses. The Solar Sonic Repeller is an innovative portable pest control device designed to address pest problems by utilizing renewable energy, specifically solar energy. This product aims to offer an environmentally friendly and efficient solution. It works by emitting ultrasonic sound waves with a frequency of 30,000–40,000 Hz. The device's advantages lie in its portability and energy independence, thanks to the use of a charging module powered by an integrated photovoltaic (PV) panel with automatic battery charging during the day. The first test measured the output frequency using an oscilloscope to verify that the oscillator circuit produced waves at the specified frequency. The second test measured the device's effectiveness by examining the pest response to the device at various distances. This test was effective within a maximum radius of approximately 14 m from the center point, covering a rice field area of ​​250 m2.

The main problem with the water heating system on offshore platforms is the absence of water level monitoring and automatic overflow detection. This has the potential to cause hot water spills that endanger workplace safety and operational efficiency. This research designs and implements a water level monitoring system based on the Arduino Uno microcontroller with HC-SR04 ultrasonic sensors. The system is equipped with LED indicators, a buzzer alarm, and a 16x2 LCD to display water level status in real-time. Water levels are classified into three zones (low, medium, high), and overflow is detected if the water is within 3 cm of the sensor. Testing was conducted on a 5-liter simulation tank representing actual 500-liter tank conditions. Test results showed a reading accuracy of 96% and a quick system response to overflow conditions (<1 second). This system is economical, easy to develop, and highly applicable for offshore environments. In addition, this system can be integrated with IoT technology for remote monitoring.

Drug abuse (Narcotics, Psychotropics, and other Addictive Substances) is a serious problem that can threaten the younger generation, especially in the Panggung Kidul Village area. The lack of public understanding, especially teenagers, regarding the dangers and negative impacts of drug abuse is one of the factors that influence the high risk of substance abuse. Therefore, innovative and interactive educational media are needed to increase public awareness and understanding regarding the prevention of drug abuse. This study offers a solution by designing and developing educational media based on Augmented Reality (AR) technology as a visual and interactive tool that conveys information in an interesting and easy-to-understand manner. The use of smartphones as the main device in AR applications makes this media more easily accessible to various groups of people. The test results using the System Usability Scale (SUS) method showed a user satisfaction level of 96% which is included in the Acceptable category. Thus, this AR-based educational media is expected to be an effective means of increasing public understanding of the dangers of drug abuse and encouraging early preventive efforts.

Information technology is a technology used to manage data, including processing, acquiring, organizing, storing, and manipulating data in various ways to produce high-quality information—namely, information that is relevant, accurate, and timely. This information is used for personal, business, and governmental purposes, serving as strategic information in decision-making. To anticipate changes in weather conditions, particularly rainfall, a valid and accurate report is needed that can be useful for the public. So far, the correlation or relationship between the factors influencing weather conditions—especially rainfall—has not been precisely determined, making it mathematically difficult to create a model that can describe the correlation among all these factors. This is where Artificial Neural Networks (ANN) come into play: to create such models and map out the existing problems purely based on the input data provided. One of the capabilities of neural networks is to make predictions based on previously learned data using the backpropagation method.

The Message Queue Telementary Transport (MQTT) protocol is able to adjust the sending and receiving of messages to monitor in accordance with the user's preferences because the sending and receiving of messages is topic based on a specified topic, making it necessary to routinely monitor the condition of patients who have been diagnosed with heart problems from a distance. With the aim to perform a Quality of Service (QoS) analysis with throughput, delay, and packet loss parameters using Unshielded Twisted Pair internet transmission media (UTP) and Wireless, the goal of this research is to design and implement (MQTT) a heart rate monitoring device with an EKG module as a sensor and ESP32 as a microcontroller. On the Ubidots website, EKG signals are transmitted over the internet and shown in real time. QoS analysis is performed using the Wireshark application. Data was collected on two scenarios at intervals of 30 minutes, 1 hour, 2 hours, 5 hours, 8 hours, 12 hours, 18 hours, and 23 hours. The throughput, latency, and packet loss metrics used in this study's results cause different value variations; these are influenced by the weather, internet bandwidth, computer, and router specifications. According to testing, the tool is portable and has a 3000mAh battery, but it has the restriction that it can only be used with reliable internet and bandwidth.

Terrestrial digital television transmitter stations are the main facilities in the transmission of digital television broadcasts to the public. The quality of the transmitted signal is expected to reach the Central Java-1 service area well so as to provide optimal and reliable quality of digital television broadcast performance according to the needs of the community, but currently, complaints about signal problems such as service coverage and reception quality still occur a lot, coverage and signal quality received by community-owned television transmitters cannot be separated from the influence of the quality performance of digital television transmission stations. The purpose of this research is to analyse the coverage performance of digital television services of digital television transmitter stations using the K-Means Clustering Method, identify areas with the best signal coverage and group areas based on the level of signal performance. The data used includes field strength parameters collected through field measurements at 25 service area location points, topography factors and transmitter distance were found to be the main causes of signal quality differences. Data analysis shows that the K-Means Clustering method effectively clusters areas with signal reception quality categories of very good cluster 3 areas, good cluster 8 areas, fair cluster 5 areas and poor cluster 9 areas. The results of this study can be recommended in the evaluation and optimisation of tele-transmitting station networks.

The digital revolution has transformed retail industry into a new era full of opportunities through e-commerce but also presenting challenges in online furniture sales. Due to their inability to precisely measure and see furniture in their spaces, many customers are hesitant to make purchases. Augmented Reality (AR) offers solutions through realistic visualizations, but it’s costly and has a lot of accessibility problems. WebAR solves these issues by delivering AR features directly on mobile browsers without requiring additional apps. This study investigates the impact of WebAR innovation on e-commerce furniture by examining the roles of interactivity (IN), vividness (VI), augmentation (AU), customization (CT), and ease of access (EA) on purchase intention (PI), mediated by spatial presence (SP), decision comfort (DC), and satisfaction (SF). Using the Stimulus-Organism-Response (SOR) model, data from 100 respondents were analyzed with the PLS-SEM algorithm using SmartPLS 4.0. The findings tell us that WebAR significantly enhances consumers’s shopping experience and ultimately boosts purchase intention.

Due to the busy activities outside the home, some people do not have time or forget to water their plants, so when they return home they find that the plants have died from drought. Therefore, to overcome this problem by conducting research using Internet of Things technology. The aim of this research is to design an automatic watering tool to overcome manual watering and support learning. This prototype uses a NodeMCU ESP8266 as the main controller, a soil moisture sensor is used to read soil moisture.  Soil moisture sensors are used to detect soil moisture. If the soil humidity is below the minimum limit, the watering process will be active and the watering process will be active if the NodeMCU ESP8266 receives commands from the smartphone.

Penelitian ini bertujuan untuk mengetahui pengaruh kWh meter yang bermasalah terhadap efektivitas kWh Jual. Penggambilan data pemakaian kWh meter bermasalah pelanggan di PT. PLN (Persero) ULP Muara Enim. Metode yang digunakan ACMT (Aplikasi Catat Meter Terpusat) dan anlisa RCPS (Root Cause Problem Solving). Hasil penelitian ini Terdapat peningkatan nilai kWh setelah dilakukan penggantian pada kWh meter bermasalah, yang sebelumnya rata-rata pemakaian hanya 3.282 kWh setelah di lakukan penggantian rata-rata pemakaian meningkat menjadi 4.011 kWh   maka terdapat selisih rata-rata pemakaian sebesar 729 kWh. Penggantian kWh meter macet dan buram meningkatkan efektifitas kWh   jual sebesar 24,97 %. Terdapat selisih biaya yang merupakan kerugian yang selama ini ditanggung oleh PT. PLN sebesar Rp.824.560,-/bulan. Dengan dilakukannya penggantian kWh meter bermasalah ini diharapkan akan meminimalisir kesalahan yang dapat menimbulkan kerugian pada PT. PLN (Persero).

Pendistribusian bantuan yang efisien menjadi tantangan utama dalam proses penyaluran. Penelitian ini bertujuan untuk merancang sistem cerdas yang mampu memetakan dan menyalurkan bantuan secara optimal dengan pendekatan algoritmik. Metodologi yang digunakan meliputi pemodelan sistem dengan menggunakan Unified Modeling Language (UML), perancangan struktur basis data relasional, dan perancangan algoritma penyaluran berdasarkan kriteria prioritas dan efisiensi logistik. UML digunakan untuk menggambarkan arsitektur sistem secara visual, meliputi use case, class, dan diagram aktivitas. Perancangan basis data dilakukan untuk memastikan integritas data dan kemudahan pengelolaan informasi bantuan, lokasi, dan kebutuhan penerima. Algoritma yang dikembangkan memanfaatkan pendekatan heuristik untuk menentukan rute penyaluran dan prioritas penerima berdasarkan parameter lokasi geografis. Hasil penelitian ini berupa prototipe model konseptual yang dapat digunakan sebagai dasar pengembangan sistem cerdas berbasis teknologi untuk mendukung proses penyaluran bantuan yang adaptif dan responsif.

CV. Apris Expressindo is a private company engaged in the construction and distributor of U-PVC frames. In its business process activities, CV. Apris Expressindo still does not utilize technology optimally, so that the transfer of information regarding offers, sales, purchases, product stock, suppliers, customers, and financial reports is still done manually. This causes inefficient business processes including: slow fulfillment of customer orders, product pricing errors between divisions, product marketing that is not optimal, financial reports that cannot be seen and controlled by company leaders. Therefore, it is necessary to implement Enterprise Resource Planning (ERP) which aims to overcome these problems. The methodology used in this study was carried out in several stages, namely, data collection: interviews, observation and literacy studies, business process analysis: activity diagrams and use case diagrams, implementation: installation, configuration and module customization, testing: blackbox and UAT (User Acception Test) . The results of this study indicate that the ERP software that has been selected and implemented, namely the Odoo application, can improve and simplify business process activities and can minimize any problems faced today. It is hoped that with this ERP system business operations can be more effective and efficient so that it is expected to provide more financial benefits for CV Apris Expressindo.

This study discusses the design of a goods storage information system at the Medan Class A Search and Rescue Office to improve the efficiency and accuracy of goods management management. This system is designed to automate the process of storing and tracking goods, which has been done manually and often causes various problems such as tracking difficulties and data inaccuracies. The results of the implementation show that this information system is able to minimize errors, reduce the risk of loss of goods, and ensure the availability of important equipment in real-time. Despite challenges such as user resistance and resource requirements, careful planning and proper training overcame these obstacles. In conclusion, this system contributes significantly in improving the operational performance of the Medan Class A Search and Rescue Office.

PT Astra International BMW Semarang is a company operating in the automotive sector with 3 supporting pillars, namely Sales, Aftersales and Spare Parts for BMW car units. The availability of spare parts is one of the determining factors for consumer satisfaction with the company because if the spare parts stock is empty it will cause consumer disappointment with the company. By using spare parts sales transaction data for the period January 2019 – June 2023, totaling 52,162, it will be utilized using data mining association techniques with the a priori algorithm and the eclat algorithm. The problem in this research is how to find out consumer purchasing patterns so that there is no shortage or empty stock of spare parts in the warehouse. This research aims to determine the association of spare parts purchasing patterns in sales transactions so that partman get recommendations in making decisions about providing priority types of spare parts. This research methodology uses CRISP-DM (Cross-Industry Standard Process for Data Mining) and is implemented with the R programming language with R studio software. In 3 trials using the Apriori algorithm and 3 trials with the Eclat algorithm, The result with the highest confidence appears in a combination of 3 itemsets with minimum support 0.01 and confidence 0.9, namely if a customer buys B11.42.8.593.186 (Set oil-filter Mx) and B83.12.5.A1A.683 (Washer Cleaner) then they will also buy Z99000000333 ( BMW Engine Oil) with confidence 1.00 or 100%. From the results of this association's analysis, it can be used as advice for the management of PT Astra International BMW Semarang in managing spare parts stock.

Most dairy farmers in our country still rely on human labor to monitor the condition of their farm areas. The current problem is the manual security system for barns, meaning that the barns need to be constantly checked around the area at all times. This is especially risky at night, when livestock theft is more likely to occur. This research implements PIR sensor technology to monitor movements around the barn, MC-38 magnetic sensors for gate security, and utilizes Android smartphones as devices to display barn data processing results, allowing real-time monitoring of barn security. The research is conducted using the Research and Development (R&D) method by Borg & Gall, encompassing only six steps. The results of the system evaluation by experts show a score of 3, falling between 2.1 – 3.0, categorized as Valid. User validation results show a score of 27.2, also falling between 2.1 – 3.0, categorized as Effective.

The insurance business within an insurance company offers insurance products owned by the insurance company. In every insurance product there is a premium payment and the premium is the income of an insurance company at the rate of the amount insured. The problem that PT BNI Life Insurance has is that there are many stops in premium payments such as policy redemptions due to errors in the benefits received or incorrect selection of the insurance product, this can reduce the achievement of targets for an insurance company. The aim of this research is to find out the best classification algorithm compared between K-Nearest Neighbor and Naive Bayes to predict the type of insurance product that customers will choose. In this research, data mining methods are applied to compare two different methods, namely the K-Nearest Neighbor method and the Naïve Bayes method. The level of accuracy results for the K-Nearest Neighbor method is 80% and the Naïve Bayes method is 70.53%, which means that the K-Nearest Neighbor method is the best method to apply to an insurance product classification system based on the demographics of prospective customers.

At this moment, IoT technology was implemented in a broader area. The development of IoT technology can help human life become more manageable. IoT systems can help human activity through the natural time control system provided before. But there are some drawbacks when implementing the IoT system. One of them is the transmission process in the networking area. It’s happened because of weather, and other natural problems. Sometimes, IoT transmission can’t reach on time. It can make errors reading data on the user side. We proposed a transmission system, namely a Hybrid MP-QUIC transmission protocol, for helping in sending the data. Using this transmission protocol can reduce the response time between client and server.

All PT Astra International BMW Semarang transactions are recorded in the database but the problem is that the stock management is  efficientless so  the part stock that buyers are interested is not available. This research aims to conduct a comparative mining results using the association rule with apriori algorithm for year 2021, 2022 and 2023 sales transaction dataset with total of 43.694 records using the Rstudio. Data mining process in each year uses the same parameters for each itemset combination. The best association pattern occurs in 2023 with support value 0.05913841 and confidence value 100%. This can be concluded that the rules formed from each year could be different eventhough using same parameters. The item that always appears in the association rule from 2021 – 2023is Z99000000333 (BMW Engine OIL) which is often purchased with items named “Set fil-oil” so it can be a recommendation for  item stocking  in the warehouse.

Based on the observations that have been made at the Arri’ayah Islamic Boarding School in Lembak Village, Lembak District,Muara Enim Regency, the problems found are. Teaching and learning activities that are less interactive, in learning the law of tajwid in hidayatus sibyan’s book, students are still difficlut to understandt the material and there needs to be other alternatives to help students understand the materia being studied, one alternative is the use of technology for education, namely by using learning media. Especially now that may students are not interested in learning to use books, they reason that learning to use book is boring which ultimately impacts on students’ lack of understanding of the material that has been presend, tharefore that researcher designed learning media for the law of tajwid hidayatus sibyan, so that students are more interested when study, in collecting data researchers used descriptive qualitativ methods including conducting interviews, direct observation, and literature study and are built using UML (Unifield Modelling Language).