HASIBUAN, SARAH APRILIA (2026) Optimasi Parameter Spatio Temporal Dbscan Menggunakan Particle Swarm Optimization Untuk Identifikasi Zona Rawan Kebakaran Di Kota Surabaya. Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
22083010011-cover.pdf Download (1MB) |
|
Text (Bab 1)
22083010011-bab1.pdf Download (155kB) |
|
Text (Bab 2)
22083010011-bab2.pdf Restricted to Repository staff only until 16 September 2028. Download (732kB) |
|
Text (Bab 3)
22083010011-bab3.pdf Restricted to Repository staff only until 16 September 2028. Download (1MB) |
|
Text (Bab 4)
22083010011-bab4.pdf Restricted to Repository staff only until 16 September 2028. Download (1MB) |
|
Text (Bab 5)
22083010011-bab5.pdf Download (81kB) |
|
Text (Daftar pustaka)
22083010011-daftar pustaka.pdf Download (93kB) |
|
Text (Lampiran)
22083010011-lampiran.pdf Restricted to Repository staff only Download (88kB) |
Abstract
As a major metropolitan city with high population density and intense socioeconomic activity, Surabaya is highly vulnerable to fire disasters. Frequent fire incidents across various areas have serious consequences, including threats to public safety, environmental damage, and economic losses. According to data from the Surabaya City Regional Disaster Management Agency (BPBD), fire incidents remain one of the primary issues that need to be addressed through a data-driven analytical approach to identify patterns of occurrence more accurately and in greater depth. However, fire analysis that integrates spatio and temporal dimensions remains limited, resulting in suboptimal utilization of spatio-temporal information for mitigation. This study aims to analyze and optimize fire-prone zones in Surabaya using the Spatial-Temporal DBSCAN (ST-DBSCAN) method. Fire station data was collected from Google Maps using the Instant Data Scraper extension, then aggregated and cleaned for use as supporting data in an analysis of proximity to fire stations. This method is used to group fire incidents based on spatio and temporal proximity, thereby identifying cluster of areas that have the potential to become fire-prone zones. The optimization process employs Particle Swarm Optimization (PSO) to determine the optimal parameter combinations for ST-DBSCAN, namely spatio Epsilon (εs), temporal Epsilon (εt), and minimum points (MinPts) and successfully generated 4 clusters with 0 noise points, a Davies-Bouldin Index (DBI) value of 0.5730, a Silhouette score of 0.6324, and a Calinski-Harabasz Index (CHI) value of 3060.00. The objective function in the optimization is designed based on the Davies Bouldin Index (DBI) value, noise ratio, and cluster count penalty to ensure the clustering results are more representative. The research results are expected to provide more accurate mapping of fire-prone areas as a basis for developing mitigation strategies and supporting disaster management policy-making in the City of Surabaya.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76.6 Computer Programming |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Data Science | ||||||||||||
| Depositing User: | Sarah Aprilia Hasibuan | ||||||||||||
| Date Deposited: | 16 Sep 2026 06:27 | ||||||||||||
| Last Modified: | 16 Sep 2026 08:59 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/60271 |
Actions (login required)
 |
View Item |