Claim Missing Document
Check
Articles

Found 2 Documents
Search

Analisis Perbandingan Kompleksitas Waktu Algoritma Mst Dalam Penyusunan Jaringan Pipa Air Bersih Wirakusuma, Kadek Ardy; Diatmika, Kadek Doni
Jurnal Teknologi Dan Sistem Informasi Bisnis Vol 7 No 1 (2025): Januari 2025
Publisher : Prodi Sistem Informasi Universitas Dharma Andalas

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47233/jteksis.v7i1.1813

Abstract

The availability of clean water is the top priority for the survival of the community. Therefore, efforts are needed to distribute clean water evenly, including by building a pipeline network that reaches every house in an area. The Drinking Water Supply and Sanitation System Management Group (KPSPAMS) of Ulakan Village, Karangasem Regency, built a water pipeline network to overcome the problem of lack of clean water experienced by local residents, especially during the dry season. However, development is still often not optimal because there is a pipe rotation (circuit) that causes the length of the pipe not to be optimized and the overall cost to be higher. Thus, the preparation of an efficient and effective pipeline network is the main focus. The location studied is part of Ulakan Village, Manggis District, Karangasem Regency. This study aims to find the most optimal pipeline network arrangement through a comparison of three Minimum Spanning Tree (MST) algorithms, namely the Kruskal, Prim and Sollin algorithms. This study represents the pipeline graph into an edge list. Furthermore, with the three algorithms, the list is analyzed based on the complexity of time and computer programs to find the Minimum Spanning Tree (MST). This research seeks to provide a practical solution of the comparison of the three algorithms with a focus on theoretical understanding and experimental results.
OPTIMASI MODEL PREDIKSI EXTREME GRADIENT BOOSTING DENGAN GENETIC ALGORITHM UNTUK PRODUKSI DAN PRODUKTIVITAS PADI Wirakusuma, Kadek Ardy; Ni Putu Novita Puspa Dewi; Kadek Yota Ernanda Aryanto
STORAGE: Jurnal Ilmiah Teknik dan Ilmu Komputer Vol. 4 No. 4 (2025): November
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/storage.v4i4.6633

Abstract

Produksi padi di Kabupaten Buleleng meningkat dari tahun 2021 hingga 2023, namun produktivitas justru menurun hingga 8%. Kondisi ini berpotensi mengganggu stabilitas pasokan beras di tengah pertumbuhan penduduk sebesar 4,59% per tahun. Diperlukan pendekatan prediktif berbasis kecerdasan buatan untuk memodelkan hubungan kompleks antar variabel pertanian. XGBoost dipilih karena kemampuannya dalam menangkap pola non-linear dan sering digunakan dalam analisis data pertanian, sementara Genetic Algorithm (GA) digunakan untuk menentukan kombinasi hiperparameter optimal guna meningkatkan performa model. Model XGBoost tanpa optimasi diterapkan sebagai pembanding untuk mengevaluasi efektivitas pendekatan hybrid. Hasil analisis menunjukkan bahwa optimasi hiperparameter berpengaruh signifikan terhadap hasil prediksi. Model GA-XGBoost menghasilkan tingkat kesalahan lebih rendah, dengan penurunan nilai MAPE sekitar 2,98% untuk prediksi produksi padi dan 0.21% untuk produktivitas dibandingkan dengan model standar atau default.