SETYATMOKO, BAYU (2016) VISUALISASI VARIABEL CUACA MENGGUNAKAN METODE KRIGING BERBASIS BIG DATA. Masters thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
2213206712-Master_theses.pdf - Published Version Download (4MB) | Preview |
Abstract
Indonesia adalah salah satu negara di dunia yang rentan mengalami
bencana alam. Bencana banjir bahkan terjadi hampir diseluruh wilayah Indonesia.
Salah satu upaya "mitigasi bencana" yang perlu dilakukan adalah peningkatan
kualitas pemantauan kondisi cuaca menggunakan visualisasi geospasial variabel
cuaca yang dihasilkan oleh peralatan observasi cuaca. Peralatan observasi cuaca
yang sering digunakan di bandara udara dan pelabuhan laut di seluruh dunia
adalah Automatic Weather Station (AWS).
Beberapa permasalahan pengolahan data AWS adalah lokasi data tersebar
diseluruh wilayah Indonesia yang luas (107 lokasi), periode update data setiap 15
menit (velocity) dan bentuk data semi terstruktur (variety). Sehingga proses
pemindahan seluruh data ke server datawarehouse sulit untuk dilaksanakan karena
terbatasnya kapasitas bandwidth jaringan komputer dan kapasitas penyimpanan
data.
Proses pengolahan data semi terstruktur AWS menggunakan aplikasi Big
Data yang berjalan pada sistem data terdistribusi Hadoop Distributed File System
(HDFS) tanpa perlu memindahkan semua data AWS tapi hanya data yang
dibutuhkan. Sedangkan visualisasi variabel cuaca menggunakan analisa data
spasial dengan metode interpolasi kriging yang dapat digunakan sebagai alat
bantu pemantauan kondisi cuaca saat ini.
Analisa cross validation menunjukkan nilai error yang kecil yaitu Mean
Error (ME) sebesar -0,0281 dan Mean Squared Prediction Error (MSPE) sebesar
1,337 pada hasil visualisasi geospasial yang baik.
============================================================
Indonesia is one country in the world that is prone to natural disasters.
Flood disaster even occurs almost throughout the territory of Indonesia. One effort
"disaster mitigation" that needs to be done is to increase the quality monitoring of
geospatial visualization of weather conditions using weather variables generated
by the weather observation equipment. Weather observation equipment that are
often used in airports and seaports around the world is the Automatic Weather
Station (AWS).
Some of the problems of AWS data processing is the location of data
scattered throughout the vast territory of Indonesia (107 location), the period of
data updates every 15 minutes (velocity) and semi-structured data form (variety).
So that the process of moving all the data to the server datawarehouse difficult to
implement because of the limited capacity of the computer network bandwidth
and data storage capacity.
Semi structured AWS data processing using Big Data applications running
on distributed data system Hadoop Distributed File System (HDFS) without the
need to move all the data AWS but only the required data. While the visualization
of weather variables using spatial data analysis with kriging interpolation method
that can be used as a tool for monitoring current weather conditions.
Cross validation analysis shows that a small error values are Mean Error
(ME) of -0,0281 and Mean Squared Prediction Error (MSPE) of 1,337 on the
outcome of good geospatial visualization.
| Item Type: | Thesis (Masters) |
|---|---|
| Additional Information: | RTE 006.312 Set v |
| Uncontrolled Keywords: | big data, hadoop, kriging, visualisasi geospasial |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science. EDP |
| Divisions: | Faculty of Industrial Technology > Electrical Engineering > 20101-(S2) Master Thesis |
| Depositing User: | Anis Wulandari |
| Date Deposited: | 04 Jan 2017 07:21 |
| Last Modified: | 27 Dec 2018 04:53 |
| URI: | http://repository.its.ac.id/id/eprint/1297 |
Actions (login required)
 |
View Item |