Claim Missing Document
Check
Articles

Found 6 Documents
Search

Rekomendasi Penentuan Pemasangan Beton Pada Titik Longsor Menggunakan Algoritma K-Nearest Neighbor (K-NN) Fadly, Andi Nur; Ahmad, Andani; Wardi, Wardi
Patria Artha Technological Journal Vol 7, No 2 (2023): Patria Artha Technological Journal
Publisher : Department of Electrical Engineering, University of Patria Artha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33857/patj.v7i2.728

Abstract

Rekomendasi penentuan pemasangan beton pada titik longsor menggunakan algoritma KNearest Neighbor (Dibimbing oleh Andani Ahmad dan Wardi) Provinsi Sulawesi Barat (Sulbar) merupakan salah satu provinsi di Indonesia yang rawan terhadap kejadian bencana. Kondisi alam yang kompleks dengan topografi wilayah yang bervariasi membuat Provinsi Sulbar memiliki ancaman bencana akibat fenomena seperti banjir yang paling sering terjadi. Tujuan dari penelitian ini Untuk mengetahui Rekomendasi penentuan pemasangan beton pada titik longsor menggunakan Algoritma KNearestNeighboord (KNN) dan Untuk mengimplementasikan Algoritma K-Nearest Neighboord (KNN) untuk rekomendasi penentuan penentuan pemasangan beton pada titik longsor di Sulawesi Barat. Algoritma yang digunakan dalampenelitian ini adalah algoritma K-NN, algoritma K-NN digunakan untuk mengklasifikasikan data yang menghasilkan prediksi dengan tingkat akurasi berdasarkan sampel data training dan data testing. Hasil dari penelitian ini adalah penerapan Algoritma K-Nearest Neighbor (K-NN) dapat memetakan titik - titik kelongsoran yang ada di provinsi sulawesi barat sedangkan hasil evaluasi yang dilakukan dengan melakukan pengujian akurasi yaitu dengan menguji akurasi metode berdasarkan persentase data training 80% dengan data testing 20% menghasilkan tingkat akurasi 75.53% dengan tingkat resiko sedang.
Deteksi Kalori Makanan Tradisional Indonesia Menggunakan Metode Single Shot Multibox Detector (SSD): Calorie Detection of Traditional Indonesian Food Using the Single Shot Multibox Detector (SSD) Method Riswanto, Riswanto; Ahmad, Andani; Hazriani, Hazriani; Tribuana, Dhimas
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 4 No. 3 (2024): MALCOM July 2024
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v4i3.1332

Abstract

Tujuan penelitian untuk mengembangkan sistem pendeteksi kalori makanan dengan menggunakan metode Single Shot Multibox Detector (SSD). Juga, bertujuan untuk mengatasi masalah manusia yang kesulitan dalam mengestimasi jumlah kalori yang dikonsumsi dari makanan. Dengan menggunakan model kecerdasan buatan dan bantuan kamera pada perangkat ponsel, pada penelitian ini memungkinkan pengguna untuk melakukan estimasi kalori yang lebih akurat. Sistem ini dirancang secara otomatis untuk mengidentifikasi dan memperkirakan jumlah kalori dalam makanan berdasarkan citra visual. Pemilihan metode SSD didasarkan pada keunggulannya dalam mendeteksi objek dengan tingkat akurasi yang tinggi dan kecepatan pengolahan yang cepat. Proses penelitian melibatkan beberapa tahap, termasuk pengumpulan dataset citra makanan, pelatihan model SSD dengan konfigurasi Hyperparameter pada 40.000 langkah, menggunakan data training sebanyak 90%, validasi 10%, dan testing 10%, serta menggunakan batch size 16 dan learning rate 0.007943453. Hasil eksperimen menunjukkan total loss sebesar 0.1670681 dan mean average precision (mAP) sebesar 65.09%. Jenis makanan berhasil dideteksi dengan baik, dan aplikasi mobile terkait mampu mengestimasi kalori makanan setelah deteksi jenis makanan. Meskipun demikian, penelitian mengidentifikasi beberapa tantangan, terutama dalam meningkatkan akurasi deteksi pada makanan dengan struktur kompleks atau variasi presentasi yang ekstensif. Temuan dari penelitian ini diharapkan dapat memberikan kontribusi penting dalam pengembangan sistem pendukung keputusan untuk pemantauan otomatis asupan kalori. 
Pemanfaatan Solar Cell pada Sistem Pertanian Hidroponik Cerdas di Samata Green House Group Indonesia, Kabupaten Gowa Muslimin, Zaenab; Ahmad, Andani; Areni, Intan Sari; ., Dewiani; Akil, Yusri Syam; Salam, Ejah Umraeni; ., Wardi; Anshar, Muh; Mayasari, Fitriyanti; Achmad, Andini Dani
JURNAL TEPAT : Teknologi Terapan untuk Pengabdian Masyarakat Vol 7 No 2 (2024): Kolaborasi yang Kuat untuk Kekuatan Kemasyarakatan
Publisher : Faculty of Engineering UNHAS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25042/jurnal_tepat.v7i2.509

Abstract

Hydroponics is a planting technique using water-based nutrient solutions, not through soil media, like conventional farming. This hydroponic system solves the problems arising from soil-based farming that requires plowing and weeding and is susceptible to pests, climate, and the need for large land areas. However, to produce quality soil, the hydroponic system requires monitoring of nutrients, temperature, humidity, water content, pH, and light. Developing a smart hydroponic system or a system based on Artificial Intelligence of Things (IoT) can guarantee these parameters online or in real-time monitoring. The next problem is the need for a continuous and reliable supply of electrical energy so that the built AIoT-based hydroponic system can work optimally. Disruption of electrical continuity will affect the crop yields from this system. Therefore, the Community Service (PkM) activity from the Department of Electrical Engineering, Hasanuddin University, conducted socialization of the use of solar cells (solar panels) in the AIoT-based smart hydroponic system, which is a solution to the problem of the continuity of electricity supply at the Samata Green House (SGH) Group, as a partner of the activity, which still fully uses electricity from PLN. Based on the results of quantitative analysis of pre-test (before socialization) and post-test (after socialization) measurements by activity participants, an increase in the level of understanding of participants was obtained, which was initially 11-55% "quite understanding" to "understanding" of the components of the Solar Power Plant (PLTS) system, as well as how to operate and maintain PLTS, to 100% of participants "quite understanding" to "very understanding" regarding these parameters. This indicates that this activity has achieved the expected targets.
Sistem Absensi Dosen Berbasis Radio Frequency Identification (RFID) Sebagai Perhitungan Honor Dosen Ikhwan Mardin, Muhammad; Ahmad, Andani; Wardi, Wardi
Jurnal JEETech Vol. 5 No. 1 (2024): Nomor 1 May
Publisher : Universitas Darul Ulum

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32492/jeetech.v5i1.5102

Abstract

This research designed a sense to find out the number of lecturers who enter and finish teaching lectures. Fikom, UMI still uses a manual system in managing lecturers' teaching salaries, so that it has several battles that usually occur in manual systems, including the attendance process and attendance recapitulation inefficient, less transparent because lecturers cannot access the issue of honorarium that lecturers will receive, and what will happen is that the salary calculation is not accurate so that it often results in disputes between lecturers and the finance department. on this occasion the author tries to design a working system of a tool that can recap lecturer attendance that can function to increase efficiency, transparency, and accuracy in the management of lecturer teaching honoraria. The system designed in this study was made using the RFID module and Raspberry Pi as the primary component of the system. The original way of working of this system comes from reading identification data from the RFID card carried by the lecturer. The data read by the RFID will be sent to the Raspberry Pi as input data. Raspberry pi will send RFID data readings to the server computer into input data. personal computer server will get data from Raspberry pi and will check the data, when the data received is synchronized with the ongoing course schedule, the server computer will send the data to Raspberry and will be displayed on the LCD. which will occur from this research shares that the process of computerized honorarium management tends to be easier and management has a better level of accuracy than using manual methods. in terms of news, reports that are designed tend to be simpler and faster because they are automatically in accordance with the input data entered previously.
Desain dan Analisis Quality of Service Pada Sistem Otomasi Rumah Menggunakan Wireless Sensor Network Raharjo, Muhammad Agung; Ahmad, Andani; Wabula, Yuyun
Journal of Information System Research (JOSH) Vol 6 No 3 (2025): April 2025
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v6i3.7156

Abstract

The problem raised in the study is how to control lights and air conditioners automatically via a smartphone with reliability and power efficiency. A home automation system for the use of lights and air conditioners works according to weather and temperature conditions using LDR sensors and DHT11 sensors from anywhere via the internet network so that it can identify and monitor the device in an on or off-state. For this reason, a design of a home control system is needed that can help users identify and monitor electronic devices remotely through a user interface, or automatically. This study aims to design a home automation system based on a Wireless Sensor Network (WSN) so that it can be used to control lights and air conditioners automatically and flexibly. In addition, this study also measures and analyzes Quality of Service (QoS) using delay parameters, packet loss, and throughput according to the TIPHON standard. From this study, the results of the tests carried out for the control function of electronic equipment connected to the home automation system using ESP8266, LDR, and Relay Node can function properly, while the network quality test according to the Quality of Service (QoS) Analysis using the TIPHON standard obtained results for Delay is 154.6 ms can be categorized as Good, while for the quality of the Loss packet it is 0.0033% with an outstanding category, while for the Throughput value is 2.2996 kbps which is categorized as bad.
Teknologi Irigasi Berbasis IoT: Integrasi Sensor Nirkabel (SN), Energi Surya dan Logika Fuzzy Sahibu, Supriadi; Surahman; Ahmad, Andani
JURNAL FASILKOM Vol. 15 No. 1 (2025): Jurnal FASILKOM (teknologi inFormASi dan ILmu KOMputer)
Publisher : Unversitas Muhammadiyah Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37859/jf.v15i1.8899

Abstract

Agriculture is a vital sector in the global economy that faces various challenges, especially in the efficient use of water resources. Irrigation is an important aspect of agriculture, especially in rice fields, because optimal water supply directly affects plant productivity. However, in Pao Village, Malangke District, and North Luwu Regency, water availability is increasingly limited due to climate change, urbanization, and inefficient exploitation of resources. Most farmers in the area do not have irrigation networks so during the dry season their rice fields experience drought and are at risk of crop failure. This study aims to design and develop an automatic irrigation system based on the Internet of Things (IoT) supported by wireless sensors and solar energy to improve the efficiency of water use in rice fields. The research method used is a prototype approach consisting of three main stages, namely preparation, prototype development, and system implementation. In the preparation stage, field observations and literature studies were carried out to determine system needs. The development stage includes design, implementation, and internal evaluation of the prototype. Furthermore, the system was implemented and tested in the field to assess its effectiveness in controlling irrigation automatically. The results of the study showed that the developed system was able to detect water levels using ultrasonic sensors and humidity sensors connected to Arduino. Data from the sensors is processed using fuzzy logic to automatically control the water pump based on the water conditions in the farmland. The energy used comes from solar panels, so the system can operate independently and is environmentally friendly. Tests have shown that the system works well in regulating the water supply to the rice fields efficiently and can help farmers overcome drought and increase agricultural productivity.