Konstruksi Struktur Jaringan Pipa Air dengan Metode DBSCAN dan Node Embedding

Pardamean, Nathanael (2025) Konstruksi Struktur Jaringan Pipa Air dengan Metode DBSCAN dan Node Embedding. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5002201056_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (5MB) | Request a copy

Abstract

Perencanaan yang efisien dari jaringan distribusi air merupakan aspek penting dalam manajemen sumber daya air. Penelitian ini bertujuan untuk meningkatkan efisiensi konstruksi jaringan pipa air dengan mempertimbangkan keterbatasan fisik dan sumber daya. Penelitian dilakukan dengan data jaringan pipa air PDAM Surabaya, memodelkan struktur topologi jaringan ke dalam ruang embedding. Pendekatan yang digunakan menggabungkan Density-based Spatial Clustering of Application with Noise untuk mengelompokkan node pelanggan berdasarkan kedekatan spasial, node embedding untuk merepresentasikan jaringan sebagai vektor dalam ruang berdimensi rendah, dan optimasi dengan algoritma Prim untuk membangun jaringan pipa air dengan mempertimbangkan batasan seperti node sumber tunggal dan pembangunan hanya pada bidang datar. Hasil penelitian menunjukkan bahwa pendekatan ini mampu melakukan konstruksi jaringan pipa air secara efisien dalam hal waktu komputasi, sekaligus menghasilkan rute pipa yang optimal dari segi panjang total jaringan yang lebih pendek. Penggunaan kombinasi node embedding, algoritma Prim, dan DBSCAN terbukti efektif dalam mengoptimasi jaringan pipa air, dengan hasil yang menunjukkan pengurangan panjang jaringan dibandingkan metode manual. Disarankan untuk mengeksplorasi algoritma clustering lain.
======================================================================================================================================
Efficient planning of water distribution networks is a crucial aspect of water resource management. This study aims to enhance the efficiency of water pipeline network construction by considering physical and resource constraints. The research utilizes pipeline network data from PDAM Surabaya, modeling the topological structure of the network into an embedding space. The approach integrates Density-based Spatial Clustering of Applications with Noise to cluster customer nodes based on spatial proximity, node embedding to represent the network as vectors in a low-dimensional space, and optimization using Prim’s algorithm to construct water pipeline networks while considering constraints such as a single source node and construction limited to flat terrains. The results indicate that this approach efficiently constructs water pipeline networks in terms of computational time while also producing optimal pipeline routes with shorter total network lengths. The combination of node embedding, Prim’s algorithm, and DBSCAN has proven effective in optimizing water pipeline networks, with results demonstrating a reduction in network length compared to manual methods. It is recommended to explore other clustering algorithms.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Algoritma Prim, DBSCAN, Node Embedding, Prim's Algorithm, DBSCAN, Node Embedding
Subjects:	Q Science > QA Mathematics > QA166 Graph theory Q Science > QA Mathematics > QA278.55 Cluster analysis Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Nathanael Pardamean
Date Deposited:	04 Feb 2025 07:35
Last Modified:	04 Feb 2025 07:35
URI:	http://repository.its.ac.id/id/eprint/117745

Actions (login required)

View Item