Fatah, Beni (2025) OPTIMASI PENDETEKSIAN BEBAN DAN AKURASI KLASIFIKASI DIMENSI ANGKUTAN KENDARAAN MENGGUNAKAN WEIGH IN MOTION. S2 thesis, Universitas Mercu Buana Jakarta - Menteng.
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/other.png) |
Other (Cover)
55421110005-BENI FATAH-01 Cover - Beni Fatah.PDF Download (797kB) |
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
55421110005-BENI FATAH-02 BAB 1 - Beni Fatah.pdf Restricted to Registered users only Download (99kB) |
|
Text (BAB II)
55421110005-BENI FATAH-03 BAB 2 - Beni Fatah.pdf Restricted to Registered users only Download (586kB) |
|
Text (BAB III)
55421110005-BENI FATAH-04 BAB 3 - Beni Fatah.pdf Restricted to Registered users only Download (100kB) |
|
Text (BAB IV)
55421110005-BENI FATAH-05 BAB 4 - Beni Fatah.pdf Restricted to Registered users only Download (326kB) |
|
Text (BAB V)
55421110005-BENI FATAH-06 BAB 5 - Beni Fatah.pdf Restricted to Registered users only Download (94kB) |
|
Text (Daftar Pustaka)
55421110005-BENI FATAH-07 Daftar Pustaka - Beni Fatah.pdf Restricted to Registered users only Download (226kB) |
|
Text (Lampiran)
55421110005-BENI FATAH-08 Lampiran - Beni Fatah.pdf Restricted to Registered users only Download (434kB) |
Abstract
Peningkatan volume kendaraan dan beban angkutan di jalan raya telah menimbulkan berbagai masalah, seperti kerusakan infrastruktur jalan dan pelanggaran batas muatan kendaraan. Sistem Weigh in Motion (WIM) telah digunakan untuk mendeteksi beban kendaraan secara dinamis, namun akurasi klasifikasi dimensi kendaraan masih menjadi tantangan, terutama dalam kondisi lalu lintas yang padat dan bervariasi. Penelitian ini bertujuan untuk mengoptimasi pendeteksian beban dan meningkatkan akurasi klasifikasi dimensi angkutan kendaraan dengan memanfaatkan metode K-Nearest Neighbors (KNN). Tujuannya adalah menciptakan sistem yang mampu bekerja secara real-time dengan tingkat kesalahan minimal dan akurasi klasifikasi yang tinggi. Metode konvensional yang telah ada masih memiliki keterbatasan dalam menangani kompleksitas data serta variasi pola kendaraan. Oleh karena itu, dalam penelitian ini mengusulkan penerapan algoritma K-Nearest Neighbor (KNN) untuk mengatasi permasalahan tersebut. Data penelitian terdiri dari 6.646 kendaraan yang dibagi menjadi 70% data pelatihan dan 30% data uji. Data uji sebanyak 1.993 unit. Hasil evaluasi menunjukkan bahwa algoritma KNN mampu mengklasifikasikan golongan kendaraan dengan akurasi yang tinggi. Secara keseluruhan, nilai precision, recall, dan F1-score masing-masing mencapai 94,8%, 95,3%, dan 95,1%. Akurasi klasifikasi per jenis kendaraan menunjukkan 98,67% untuk long dump, 96,43% untuk bak buka samping, 94,44% untuk long wing box, sedangkan untuk unit box dan long box masing-masing mencapai 50%. Temuan ini vii mengindikasikan bahwa penerapan KNN dalam sistem WIM berpotensi meningkatkan keandalan pendeteksian beban dan akurasi klasifikasi dimensi kendaraan, yang selanjutnya dapat mendukung pengembangan solusi transportasi yang lebih efektif dan efisien. The increase in vehicle volume and transport loads on roads has caused various problems, such as damage to road infrastructure and vehicle load limit violations. The Weigh in Motion (WIM) system has been used to detect vehicle loads dynamically, but the accuracy of vehicle dimension classification remains a challenge, especially in heavy and varied traffic conditions. This study aims to optimise load detection and improve the accuracy of vehicle dimension classification by utilising the K-Nearest Neighbours (KNN) method. The goal is to create a system that is capable of working in real-time with minimal error and high classification accuracy. Conventional methods still have limitations in handling data complexity and vehicle pattern variations. Therefore, this study proposes the application of the K-Nearest Neighbour (KNN) algorithm to overcome these problems. The research data consisted of 6,646 vehicles, divided into 70% training data and 30% test data. There were 1,993 units of test data. The evaluation results show that the KNN algorithm is capable of classifying vehicle types with high accuracy. Overall, the precision, recall, and F1-score values reached 94.8%, 95.3%, and 95.1%, respectively. The classification accuracy per vehicle type was 98.67% for long dumps, 96.43% for side-opening trucks, 94.44% for long wing boxes, while for box units and long boxes, it reached 50% each. These findings indicate that the application of KNN in the WIM system has the potential to improve the reliability of load detection and the accuracy of vehicle ix dimension classification, which in turn can support the development of more effective and efficient transportation solutions.
| Item Type: | Thesis (S2) |
|---|---|
| NIM/NIDN Creators: | 554211110005 |
| Uncontrolled Keywords: | Weigh in Motion (WIM), K-Nearest Neighbors (KNN), klasifikasi dimensi kendaraan, deteksi beban kendaraan, optimasi akurasi. Weigh in Motion (WIM), K-Nearest Neighbours (KNN), vehicle dimension classification, vehicle load detection, accuracy optimisation. |
| Subjects: | 600 Technology/Teknologi > 620 Engineering and Applied Operations/Ilmu Teknik dan operasi Terapan > 621 Applied Physics/Fisika terapan > 621.3 Electrical Engineering, Lighting, Superconductivity, Magnetic Engineering, Applied Optics, Paraphotic Technology, Electronics Communications Engineering, Computers/Teknik Elektro, Pencahayaan, Superkonduktivitas, Teknik Magnetik, Optik Terapan, Tekn |
| Divisions: | Pascasarjana > Magister Teknik Elektro |
| Depositing User: | ZAIRA ELVISIA |
| Date Deposited: | 14 Oct 2025 01:57 |
| Last Modified: | 14 Oct 2025 01:57 |
| URI: | http://repository.mercubuana.ac.id/id/eprint/99189 |
Actions (login required)
 |
View Item |
