Propeller is part of a key component in fishing boat propulsion. Propeller can provide momentum to the fluid which can be a thrust on the ship. However, The failure of the propeller found prematurely. The failure of the propeller maybe because of overload on the propeller model so the fatigue life of the propeller becomes low. On the other hand, the rapid development of technology can simulate a design model to look for failures that occur. Finite Element Analysis is one of the designer solutions to determine the age of failure of a model and failure-prone areas in a model. This study uses propeller model data from fishing boat with engine 24HP in Dumai City TPI that always fail prematurely. The material used is copper alloy. While the drawing model uses Autodesk Inventor and Finite Element Analysis simulation using ANSYS R17.2 software with the number of model nodes is 51108 and the number of elements of the model is 26268. The results obtained from this study are Von Mises stress on the simulation model that is equal to 613.33 MPa to 0.01164 MPa. While the deformation value due to the effect of loading on the model is 5,3657 mm to 0 mm. These results affect the age of fatigue (fatigue life) on the model with the highest value 109 and the lowest 0. The results of the fatigue life value on the model affect the results of the level of damage and the safety number of the model with successive values of 1032 to 1 and 15 to 0.32446. The conclusion of the result is the propeller will fail prematurely.

This research focuses on corncob waste as activated carbon. The matrix of tapioca (starch) is used as adhesive. The objective of this research is to compare the results of engine exhaust emission, power, and torque of vehicle with and without AC made of corncobs. The utilization of AC classified into two types of composition namely corncob and starch with 80:20 and 70:30. It arranged on the intake air filter bound and neatly to absorb nitrogen from the air so that pure oxygen burned in the combustion chamber. The results show that the AC reduces the percentage of CO, HC, CO2, O2, and AFR at various engine rotation. It also influences power and torque of vehicle performance. The result shows that AC slightly effect the performance. It decreases the torque at 70:30 composition. The main conclusion can be drawn that AC from corncobs greatly affects exhaust emissions in vehicles.

Abstrak: Pencarian topik skripsi seringkali menjadi tantangan bagi mahasiswa, terutama mahasiswa kependidikan keteknikan untuk menentukan topik yang tepat sesuai dengan kebutuhan industri dan pendidikan. Tujuan dari penelitian ini adalah untuk mengetahui apakah ChatGPT mampu membantu mahasiswa dalam menemukan topik skripsi yang sesuai dengan minat dan bidang ilmunya. Penelitian ini menggunakan metode studi eksploratif yang bertujuan untuk memahami fenomena atau menjelaskan hubungan antara variabel-variabel dalam suatu populasi. Penelitian ini dilakukan pada delapan prodi kependidikan di Fakultas Teknik Universitas Negeri Medan dengan jumlah populasi sebanyak 540 mahasiswa tahun masuk 2019. Pengambilan sampel menggunakan teknik convenience sampling dengan masing-masing sampel 10 mahasiswa setiap prodi, total sampel adalah 80 mahasiswa. Berdasarkan hasil survei, mahasiswa yang telah memiliki topik skripsi berjumlah 63 orang atau 79% dan mahasiswa yang belum memiliki topik skripsi berjumlah 17 orang atau 21%. Selanjutnya fokus terhadap hasil survei 17 mahasiswa yang belum memiliki topik skripsi. Dimana 17 mahasiswa telah memilih topik skripsi yang dihasilkan oleh ChatGPT, dengan delapan orang memilih karena latar belakang, enam orang karena sumber referensi, dan tiga orang karena metode penelitian. Berdasarkan tanggapan dari 14 responden mengungkapkan bahwa tidak pernah menemukan topik skripsi yang dipilih dan sisanya tiga orang pernah menemukan topik skripsi pada penelitian terdahulu. Dan 17 mahasiswa yang belum memiliki topik skripsi memberi tanggapan bahwa 100% topik yang dihasilkan oleh ChatGPT mempunyai nilai kebaharuan. Oleh karena itu, ChatGPT mempunyai inovasi untuk menemukan topik skripsi yang mempunyai nilai kebaharuan. Bahkan ChatGPT mampu merekomendasikan topik-topik skripsi yang diminta sesuai dengan keinginan dan keilmuan yang dimiliki. Kata Kunci: ChatGPT, Topik Skripsi, Kependidikan Abstract: Finding thesis topics can often be a challenge for students, especially engineering education students, as they strive to identify the most appropriate topic that aligns with industry and educational requirements. The purpose of this study is to determine the ability of ChatGPT to assist students in discovering thesis topics that match their interests and areas of expertise. This research uses an exploratory study method, which aims to understand phenomena or explain the relationship between variables within a population. The study was conducted in eight education programs at the Faculty of Engineering, Universitas Negeri Medan, with a total population of 540 students from the 2019 batch. Sampling was done using the convenience sampling technique with each sample of 10 students from each program of study, the total sample size was 80 students. Based on the survey results, students who already have a thesis topic 63 people or 79%, and students who do not have a thesis topic 17 people or 21%. In addition, focus on the survey results of 17 students who do not have a thesis topic. Seventeen students chose the thesis topic generated by ChatGPT, with eight people choosing it because of the background, six because of the reference source, and three because of the research method. Based on the responses of 14 respondents, it was revealed that they had never found the chosen thesis topic, and the remaining three people had found thesis topics in previous research. In addition, 17 students who did not have a thesis topic responded that 100% of the topics generated by ChatGPT had a novelty value. Therefore, ChatGPT is innovative in finding thesis topics with novelty value. Even ChatGPT can recommend the requested thesis topics according to one’s wishes and knowledge. Keywords: ChatGPT, Thesis Topic, Education

This study aims to improve learning outcomes of corrosion and coating techniques through the application of a module-based blended learning model. The subjects of this study were 31 students of mechanical engineering education, Faculty of Engineering, Universitas Negeri Medan. Based on the research results can be described that the learning outcomes of 31 students got a B = 3 students (9,68%), got a C = 25 students (80.64%), got an E = 3 students (9,68%). Average student assessment of lecturers' teaching activities before applying the module (x̄) = 2.26. The average observer's assessment of student activities before applying the module (x̄) = 2,20. The average student assessment of the module display (x̄) = 3,36. The average student assessment of the application of the module (x̄) = 3,31. The average student assessment of the lecturer's teaching activities (x̄) = 3,62. The observer's average assessment of student activities in class (x̄) = 3,68.Student learning outcomes from 31 students, got an A = 13 students (41,94 %), got a B = 17 students (54,84%), got an E = 1 student (3,22%). Based on this research activity, it can be concluded that the use of corrosion modules and coating techniques with blended learning models can improve learning outcomes of corrosion and coating techniques.

The problem-based learning method is one of the methods that can be applied in EFL classes, especially for students' writing skills, so that it can create a pleasant learning atmosphere and make learning more active in classroom learning activities. The problem-based learning model is one of the learning models that can shape scientific behavior, social actors, and develop students' curiosity. Learners are initially faced with a problem, then followed by a student centered information search process. Seeing the importance of this method, this study aims to see the extent to which the PBL method can improve students' ability to write Narrative text. With Classroom Action Research design, this study was applied to 20 students of class X. In collecting data, tests were used to obtain an overview of student achievement, each of which was given in each cycle. The results of this study show that PBL can improve students' narrative text writing skills, this can be seen from the percentage of student results in each cycle. In addition to making the classroom atmosphere active and participatory, PBL also makes students think more critically in solving their problems. Students are required to be able to solve their problems both individually and in groups. Seeing the important role of this PBL method, the results of this study can be used as starting points or corrections for further research