Model Spasial Adaptif untuk Prediksi Risiko Lingkungan pada Proyek Demolisi Bangunan di Perkotaan Berbasis Hybrid Kriging-Fuzzy

Atpriyanti, Dinar Nurina (2026) Model Spasial Adaptif untuk Prediksi Risiko Lingkungan pada Proyek Demolisi Bangunan di Perkotaan Berbasis Hybrid Kriging-Fuzzy. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6012241009-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (8MB) | Request a copy

Abstract

Kegiatan demolisi bangunan di kawasan perkotaan padat berpotensi menimbulkan risiko lingkungan yang signifikan, terutama berupa polusi debu (PM₁₀), kebisingan, dan getaran yang dapat berdampak pada keselamatan pekerja, kesehatan masyarakat, serta keberlanjutan proyek. Namun, praktik pemantauan risiko lingkungan pada proyek demolisi umumnya masih bersifat parsial, bergantung pada sensor titik, dan memiliki keterbatasan dalam merepresentasikan variasi spasial serta ketidakpastian kondisi lapangan secara memadai. Penelitian ini bertujuan mengembangkan model penilaian risiko lingkungan yang bersifat spatial-adaptive melalui integrasi algoritma Kriging dan Fuzzy Inference System (FIS) sebagai sistem pemantauan dan peringatan dini pada proyek demolisi bangunan. Kriging digunakan untuk merekonstruksi distribusi spasial polusi debu, kebisingan, dan getaran dari data pengukuran terbatas sehingga menghasilkan peta sebaran risiko yang kontinu dan representatif. Selanjutnya, FIS mengonversi keluaran numerik hasil interpolasi menjadi tingkat risiko lingkungan linguistik, yaitu aman, waspada, dan bahaya, berdasarkan ambang batas keselamatan yang ditetapkan. Model diterapkan pada studi kasus pembongkaran gedung tiga lantai di Universitas Muhammadiyah Surakarta menggunakan pendekatan kuantitatif berbasis data lingkungan. Hasil validasi menunjukkan bahwa model memiliki kinerja prediktif yang baik, dengan nilai Root Mean Square Standardized Error (RMSSE) masing-masing sebesar 0,8097 untuk debu, 0,7286 untuk kebisingan, dan 0,8152 untuk getaran. Nilai RMSSE yang mendekati satu menunjukkan bahwa model menghasilkan estimasi yang relatif tidak bias dan memiliki tingkat akurasi yang memadai dalam memprediksi distribusi spasial polutan. Penelitian ini berkontribusi dengan menghadirkan pendekatan prediksi dan penilaian lingkungan yang holistik, adaptif, dan berorientasi pada pengambilan keputusan, sehingga dapat mendukung identifikasi area berisiko serta penyusunan strategi mitigasi pada proyek demolisi di kawasan perkotaan padat
===================================================================================================================================
Building demolition activities in densely populated urban areas pose significant environmental risks, particularly in the form of dust (PM₁₀), noise, and vibration pollution, which can adversely affect worker safety, public health, and project sustainability. However, environmental risk monitoring practices in demolition projects are generally still fragmented, rely on point-based sensor measurements, and are limited in their ability to adequately represent spatial variability and on-site uncertainty. This study aims to develop a spatial-adaptive environmental risk assessment model by integrating the Kriging algorithm and a Fuzzy Inference System (FIS) as an environmental monitoring and early warning system for building demolition projects. Kriging is employed to reconstruct the spatial distribution of dust, noise, and vibration pollution from limited measurement data, generating continuous and representative environmental risk maps. Subsequently, the FIS converts the interpolated numerical outputs into linguistic environmental risk levels (safe, caution, and hazardous) based on established safety threshold values. The model is applied to a case study of a three-storey building demolition at Universitas Muhammadiyah Surakarta using a quantitative approach based on environmental data. Validation results indicate that the proposed model achieves good predictive performance, with Root Mean Square Standardized Error (RMSSE) values of 0.8097 for dust, 0.7286 for noise, and 0.8152 for vibration. These RMSSE values, which are close to one, indicate that the model provides relatively unbiased estimates with satisfactory predictive accuracy for the spatial distribution of environmental pollutants. This study contributes a holistic, adaptive, and decision-oriented approach to environmental risk monitoring that supports the identification of high-risk areas and the development of mitigation strategies for building demolition projects in densely populated urban environments

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	demolisi bangunan, kriging, fuzzy inference system, risiko lingkungan, debu, kebisingan, getaran, pemetaan spasial, building demolition, Kriging, fuzzy inference system, environmental risk, dust, noise, vibration, spatial mapping
Subjects:	T Technology > TD Environmental technology. Sanitary engineering
Divisions:	Faculty of Civil, Planning, and Geo Engineering (CIVPLAN) > Civil Engineering > 22101-(S2) Master Thesis
Depositing User:	Dinar Nurina Atpriyanti
Date Deposited:	24 Jul 2026 03:29
Last Modified:	24 Jul 2026 03:29
URI:	http://repository.its.ac.id/id/eprint/136891

Actions (login required)

View Item