Publication Search

Manual post-harvest paddy stirring requires significant time and labor and often results in uneven mixing, which can affect grain quality. To address this issue, this study designed and implemented a prototype of an Internet of Things (IoT)-based paddy stirring robot to simplify the process and improve efficiency. The system utilizes an ESP32 microcontroller as the main controller, DC motors as the stirring mechanism, and an IoT module for wireless connectivity to a mobile application. The research stages included hardware design, control system programming, IoT platform integration, and performance testing. Testing was conducted to evaluate response time, mixing uniformity, and power consumption. The results showed that the system could be operated remotely via a local Wi-Fi network with an average delay of less than 1 second, enabling real-time control. The prototype successfully stirred 0.3 kg of paddy with a mixing uniformity rate of 92% and an average power consumption of 12 watts. The application of IoT in the paddy stirring mechanism significantly improved time efficiency, reduced manual labor requirements, and maintained grain quality compared to traditional methods. These findings indicate the potential for further development into a large-scale automated paddy processing system with integrated humidity and temperature sensors for real-time quality monitoring, supporting the modernization of post-harvest processing through digital technology.

Coffee is a leading commodity in Indonesia's agricultural sector, possessing high economic value and providing a livelihood for many farmers. However, coffee plant productivity often declines significantly due to various diseases affecting the leaves, stems, and berries. This situation is exacerbated by the lack of knowledge among most farmers in recognizing early disease symptoms, resulting in delayed treatment. Consequently, crop losses are unavoidable. Based on these challenges, this study aims to design and build an expert system capable of diagnosing coffee plant diseases quickly, precisely, and accurately using the Bayesian Theorem method. This method was chosen because it can calculate the probability of a disease occurring based on observed symptoms in plants. The Bayesian approach allows the system to provide more reliable diagnostic results by updating the probability values ​​as new evidence is introduced. The developed expert system is web-based, making it easily accessible to users, both farmers and other interested parties. Users simply select the symptoms observed in coffee plants, and the system will then provide a diagnostic result in the form of possible diseases and their probability levels. Test results indicate that the system is capable of providing fairly accurate diagnostic results and can be used as a basis for farmers in making initial decisions regarding coffee plant disease management. With this expert system, farmers are expected to improve their ability to detect coffee plant diseases early, thereby maintaining crop productivity. This expert system is expected to be an effective decision support tool for farmers to reduce crop losses and improve agricultural sustainability.

Pulmonary tuberculosis remains a pressing public health problem, particularly in the work area of the Duduk Health Center (UPT Puskesmas). Effective management of this disease requires a thorough understanding of the characteristics of the causes of pulmonary TB in patients. This study aims to classify pulmonary TB cases based on the main causes such as diabetes mellitus, irritant factors, pleural effusion, and family environmental conditions. The research method used is a clustering technique with the K-Means algorithm. The data used are data on pulmonary TB patients in 2020–2025 with variables of age, gender, and causative factors collected from medical records. The analysis process was carried out using MATLAB R2014b software. The clustering model was carried out in 3, 4, and 5 clusters to compare the level of segmentation efficiency. Based on the calculation results, the model with 5 clusters showed the lowest cluster variance value of 0.4889 compared to the 3-cluster model (0.7333) and 4-cluster models (0.6151), which indicates that the division into 5 clusters produces the most compact and representative data group. Each cluster shows a different combination of characteristics of pulmonary TB patients, for example: (1) elderly male patients with comorbid diabetes; (2) adolescent females with the negative influence of environmental factors; (3) adult males exposed to irritants; (4) patients with pleural effusion; and (5) groups with multiple factors. The results of this study can provide strategic input for the Finished Community Health Center UPT in formulating more targeted and targeted intervention policies in order to prevent, control, and handle pulmonary tuberculosis cases in a sustainable and effective manner.

The development of Internet of Things (IoT) technology provides opportunities to automate various devices, including height measuring devices. This research aims to design and build an automatic height measuring device that is integrated with IoT technology. This tool is designed to measure body height automatically and send the measurement data to a cloud-based platform, making it easier for users to monitor data in real-time via smart devices. This system uses an ultrasonic sensor to detect body height, a microcontroller as a data processor, and a Wi-Fi module to send data to the server. Test results show that this tool is capable of measuring body height with a good level of accuracy and provides convenience in storing and monitoring measurement data remotely. Thus, this IoT-based height measuring device can be an innovative solution for the need for height measurement that is more efficient and integrated with digital systems.

Personal Protective Equipment (PPE) is essential for worker safety, especially in oil palm harvesting activities. PT Langkat Nusantara Kepong faces major challenges related to work safety, with analysis showing that work accidents still occur frequently in the Padang Brahrang Plantation. This indicates the need for an in-depth evaluation of the use of Personal Protective Equipment (PPE) to reduce the risk of work accidents. By using the clustering method to group data based on the type of Personal Protective Equipment (PPE) used and aims to provide recommendations for optimizing the use of personal protective equipment based on risk management and reducing the incidence of work accidents. From testing the results of cluster 3, cluster 4 and cluster 5 it can be concluded that clustering with 5 clusters provides the most efficient and precise results, followed by 4 clusters, while 3 clusters provide greater variation within clusters, indicating that clustering with fewer clusters is less able to capture subtle differences in the data.

The development of information technology that is currently developing serves to facilitate, accelerate, benefit and provide other alternatives for people who have businesses and have a big influence in the future. One of the things that is very influential is the sale of MSMEs. MSMEs are productive businesses owned by individuals or business entities that have met the criteria as micro businesses that have an important role in the economy because they provide employment, encourage local economic growth, and create innovation. MSMEs still face challenges such as limited access to financing, digital readiness, and marketing access that hinder the development of MSMEs. Therefore, it is necessary to take action to predict MSME sales in Binjai City using the backpropagation method so that later it can create new innovations and encourage community economic growth. Based on the process carried out using the backpropagation method, it can be seen that the value obtained has reached more than the predetermined target with a target value (t) of 0.26, learningrate 0.2, maximum epoch 10000 target error 0.01.

Dyspepsia is a gastroduodenal disorder that is often characterized by symptoms such as epigastric pain, burning, bloating, and a feeling of fullness after eating. Treatment of dyspepsia often requires examination by a specialist doctor, which may not always be easily accessible due to distance, cost, or time constraints. Therefore, this study aims to diagnose dyspepsia using the Dempster-Shafer method to identify possible dyspeptic diseases such as GERD, gastritis, dyspepsia, and gastric ulcers based on 16 detected symptoms and 4 different treatments. to make it easier for patients to consult and get an initial diagnosis without having to see a specialist doctor directly. From this study, it is expected to help patients get information on the initial diagnosis of the patient.

Cases of biological food poisoning can be caused by several causative factors, one of which is a food processing site that does not meet health requirements. According to the BPOM report (2016) cases of food poisoning in Indonesia in 2016 reached 1,068 cases. In 2016, 60 extraordinary events (KLB) of food poisoning were reported by 31 BB/BPOM throughout Indonesia. From the many cases of food poisoning that occur, it is necessary to take action in prevention by processing data on existing cases of poisoning to follow up on existing problems to reduce the number of cases of food poisoning by using a system on a computer so that the managed data can be processed quickly to obtain further information. Therefore the author wants to use a system with the clustering method to assist in processing data on biological poisoning cases grouping objects based on the characteristics of each object. Based on the research conducted, it can be seen that in cluster 2 in the dasta group of biological poisoning cases there are 11 data with centroid point age (x) 2, namely 12-16 years, centroid point on the type of poisoning (y) 6.36, namely sandwiches, and centroid point on the causative factor (z) 2.9, namely Gram-negative rod-shaped bacteria which are usually found in the intestines of humans and warm-blooded animals.

Stunting is a growth and development disorder in children caused by chronic malnutrition over a long period of time, especially in the first 1,000 days of life, namely from pregnancy to the first 2 years of life. There are more than 149 million (22%) toddlers worldwide who are stunted, of which 6.3 million are Indonesian toddlers. Based on data from the Ministry of Health, the stunting rate in Indonesia in 2023 was recorded at 21.5 percent, only down 0.1 percent from the previous year which amounted to 21.6 percent. Predicting the number of stunted toddlers is very important and necessary to know the stunting rate in Langkat Regency in 2024, and the prediction results can help health workers in handling and preventing the spread of stunting. The method applied to this prediction system is Multiple Linear Regression where this analysis determines whether each independent variable is positively or negatively related, the direction of the relationship between variables, and estimates the value of the dependent variable will increase or decrease.  The prediction system is carried out using the RapidMiner application because this application is very appropriate to produce information output in the form of prediction results for the coming year. The prediction results obtained are an increase and decrease in 2024 in each sub-district and there are sub-districts that do not experience an increase and decrease. The sub-district with the highest number was Secanggang with approximately 177 people, and the sub-district with the lowest number of stunted children was West Berandan with approximately 55 people. Then Stabat sub-district became the sub-district that experienced the most increase in the number of stunting, which was around 15 people, and the sub-district that experienced the most decrease was Kuala sub-district with a total of approximately 23 people. From the overall results it can be calculated that the number of stunting in all districts in Langkat Regency amounted to approximately 2453 people in 2024. And testing the error rate of prediction results using RMSE in the RapidMiner application of 7.63%, where the level of accuracy in the prediction of child stunting in Langkat Regency is 92.46%.

According to academic data, student math ability tests at MTSS PAB 5 Klambir Lima yield mixed results. There are students who understand math well, but there are also those who have difficulty understanding the mathematical concepts themselves. Math teachers at this school have difficulty designing lessons that can meet the needs of students with different levels of understanding. So, it is necessary to group student data to produce educational decision-making and improve learning effectiveness, such as through data mining. Data mining is a semi-automated process that uses machine learning techniques, mathematics, statistics, and artificial intelligence to identify and organize information contained in large databases. The process of finding information can be done by determining the decision rule based based on the level of student understanding in mathematics lessons using the Decision Tree Algorithm C4.5 method. The use of the Decision Tree algorithm C4.5 aims to make it easier to determine decision rules based on gender, Predicate, teacher teaching methods, student learning interest, and level of understanding. Based on the results of the study, it was found that if the teacher's teaching method is good, the predicate value is B, the student's learning interest is less interested, and the gender is male, then the student's level of understanding in mathematics lessons is not understood.

As a developing city, Binjai has a variety of business potential that can be exploited by micro entrepreneurs. However, in identifying and exploiting these opportunities, they are often faced with various obstacles, such as lack of access to market information, intense competition, and changes in consumer needs. Therefore, determining effective business opportunities is the key to the growth and sustainability of micro businesses in Binjai City. Determining business opportunities for micro businesses in Binjai City includes an understanding of the complexities and challenges faced by the MSME sector in identifying and exploiting business opportunities. Determining business opportunities requires alternative types of business in the TOPSIS method to compare various business opportunities based on important factors so that you can choose the one with the most potential and profit. In this context, the use of the TOPSIS method is important to assist in making more informed and effective decisions for authorities such as the Department of Cooperatives and MSMEs. This method will provide a systematic framework for evaluating various existing business opportunities, enabling a more objective and accurate assessment to support the development of MSMEs in Binjai City.

Technological advancements in the era of globalization demand improvements in the quality of academic services and educational facilities in institutions. STMIK Kaputama is committed to creating a conducive academic environment by providing optimal facilities. This study aims to determine student satisfaction with campus facilities using the K-Means Clustering method. Data were obtained from recapitulated survey reports and questionnaires filled out by students in 2024. The K-Means Clustering method was chosen for its ability to group students based on their similar preferences for campus facilities. The results show that, in general, students are fairly satisfied, though their preferences for specific facilities vary. These findings can be used to make recommendations for the improvement and development of campus facilities, help STMIK Kaputama allocate resources more efficiently, and plan strategies to enhance the quality of facilities to meet student expectations.

The advancement of information technology and globalization has transformed shopping behaviors, with social media becoming the primary platform for online shopping. This study aims to analyze the online shopping preferences of residents in Binjai City through social media using clustering methods, specifically the K-Means algorithm. Data were collected via a questionnaire targeting 523 respondents in Binjai City, focusing on variables such as gender, age, and the social media platforms used. Clustering methods are employed to group online shopping data into representative clusters, helping identify community preferences for specific social media platforms for shopping. Matlab is used to process the data and generate relevant insights into online shopping patterns, facilitating decision-making regarding the selection of the most suitable social media platform for transactions.The findings of this study are expected to provide valuable insights for both sellers and buyers in determining the most effective social media platforms for online shopping. Additionally, the results will be useful for residents of Binjai City to understand and choose the social media platforms that best meet their online shopping needs.