Optimasi Rute Unmanned Aerial Vehicle Dalam Penanganan Stressed Region Pada Precision Agriculture Dengan Menggunakan Algoritma Hibrida Iterated Local Search Dan Lovebird

Agung, Aryathama Raditya (2026) Optimasi Rute Unmanned Aerial Vehicle Dalam Penanganan Stressed Region Pada Precision Agriculture Dengan Menggunakan Algoritma Hibrida Iterated Local Search Dan Lovebird. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5026221182-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Pertumbuhan kebutuhan pangan di Indonesia menuntut peningkatan efisiensi produksi pertanian, di mana precision agriculture menawarkan solusi melalui pemanfaatan teknologi termasuk penggunaan Unmanned Aerial Vehicle (UAV) untuk penanganan stressed region. Permasalahan penentuan rute UAV dalam penyemprotan agrokimia dimodelkan sebagai Vehicle Routing Problem (VRP) dengan fungsi tujuan minimasi makespan, yaitu jarak tempuh UAV terlama dalam satu siklus operasi. Penelitian ini mengusulkan algoritma Hibrida ILS dan Lovebird (HILS-L) yang menggabungkan kerangka Iterated Local Search dengan mekanisme inisialisasi berbasis pool solusi algoritma Lovebird, serta varian HILS-L dengan perturbasi adaptif yang mengombinasikan Double Bridge Move dan Ruin and Recreate secara dinamis berdasarkan kondisi stagnasi pencarian. Pengujian pada tiga set data menggunakan 20 kali run per algoritma menunjukkan bahwa HILS-L standar mencatatkan rata-rata makespan 1097,27 m, 1666,67 m, dan 2484,87 m pada Dataset 1, Dataset 2, dan Dataset 3 sehingga mengungguli ILS dan Genetic Algorithm, sedangkan HILS-L Adaptif lebih lanjut menekan rata-rata makespan menjadi 976,50 m, 1512,88 m, dan 2392,77 m dengan keunggulan terhadap OR-Tools pada Dataset 2. Kelemahan utama varian adaptif terletak pada stabilitas hasil yang lebih rendah sebagaimana tercermin dari standar deviasi yang lebih tinggi dibandingkan HILS-L standar.
=================================================================================================================================
The growing food demand in Indonesia necessitates increased agricultural efficiency, where precision agriculture offers a solution through technology utilization including the use of Unmanned Aerial Vehicles (UAVs) for managing stressed regions. The UAV route planning problem for agrochemical application is modelled as a Vehicle Routing Problem (VRP) with the objective of minimizing makespan, defined as the longest travel distance among all UAVs in a single operational cycle. This study proposes a Hybrid ILS and Lovebird (HILS-L) algorithm combining the Iterated Local Search framework with a Lovebird-based solution pool initialization mechanism, along with an adaptive perturbation variant that dynamically combines Double Bridge Move and Ruin and Recreate based on search stagnation conditions. Experiments on three datasets using 20 runs per algorithm show that standard HILS-L achieves average makespans of 1097.27 m, 1666.67 m, and 2484.87 m on Dataset 1, Dataset 2, and Dataset 3 respectively, outperforming ILS and Genetic Algorithm, while the adaptive variant further reduces average makespans to 976.50 m, 1512.88 m, and 2392.77 m with superiority over OR-Tools on Dataset 2. The primary limitation of the adaptive variant lies in its lower solution stability reflected by a higher standard deviation compared to standard HILS-L.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Precision agriculture, stressed region, Unmanned Aerial Vehicle, Vehicle Routing Problem, Lovebird algorithm, Iterated Local Search algorithm
Subjects:	T Technology > T Technology (General) > T57.84 Heuristic algorithms.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis
Depositing User:	Aryathama Raditya Agung
Date Deposited:	20 Jul 2026 01:31
Last Modified:	20 Jul 2026 01:31
URI:	http://repository.its.ac.id/id/eprint/135641

Actions (login required)

View Item