Evaluasi Akurasi Geometri Model Building Information Modeling Gedung Daruslan Berdasarkan Data Point Cloud

Mif'Al Bagus Prasetyo, Ika Rahmawati Suyanto, S.T., M.Eng.

2025 | Tugas Akhir | D4 TEKNIK PENGELOLAAN DAN PEMELIHARAAN INFRASTRUKTUR SIPIL

Perkembangan teknologi konstruksi telah mendorong integrasi Building Information Modeling (BIM) dan point cloud sebagai metode representasi geometri bangunan secara presisi. Gedung Daruslan dipilih sebagai objek studi untuk mengevaluasi akurasi model BIM berbasis data point cloud sekunder. Penelitian ini bertujuan menilai sejauh mana model BIM dapat merepresentasikan kondisi fisik bangunan secara akurat melalui validasi statistik.

Penelitian ini menggunakan data point cloud sekunder hasil pemindaian Terrestrial Laser Scanner (TLS), serta data primer berupa pengukuran lapangan menggunakan distometer. Pemodelan dilakukan menggunakan Autodesk Revit, sedangkan analisis statistik dilakukan di Microsoft Excel. Evaluasi akurasi dilakukan dengan perhitungan Root Mean Square Error (RMSE) dan uji-t berpasangan terhadap 30 titik sampel yang dipilih melalui metode stratified random sampling berbasis spasial.

Hasil menunjukkan bahwa data point cloud memiliki presisi tinggi (points < 6>

The advancement of construction technology has encouraged the integration of Building Information Modeling (BIM) and point cloud as a precise method for representing building geometry. The Daruslan Building was selected as the case study to evaluate the geometric accuracy of a BIM model generated from secondary point cloud data. This research aims to assess how accurately the BIM model reflects the physical condition of the building through statistical validation.

This study utilizes secondary point cloud data acquired using a Terrestrial Laser Scanner (TLS), along with primary data obtained through field measurements using a distometer. The BIM model was created using Autodesk Revit, while statistical analysis was conducted in Microsoft Excel. Accuracy evaluation involved calculating the Root Mean Square Error (RMSE) and conducting a paired t-test on 30 sample points selected using a spatially based stratified random sampling method.

The results indicate that the point cloud data has high spatial precision (points < 6>

Kata Kunci : Building Information, Modeling, point cloud, validasi geometri, RMSE, uji statistik.