Pemodelan Produksi Minyak Dan Gas Bumi Pada Platform "MK" Di PT. X Menggunakan Metode ARIMA, Neural Network, Dan Hibrida ARIMA-Neural Network

Prameswari, Windia Cinde (2016) Pemodelan Produksi Minyak Dan Gas Bumi Pada Platform "MK" Di PT. X Menggunakan Metode ARIMA, Neural Network, Dan Hibrida ARIMA-Neural Network. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
1314105010-Undergraduate_Thesis.pdf - Accepted Version

Download (1MB) | Preview
[img]
Preview
Text
1314105010-Paper.pdf - Accepted Version

Download (240kB) | Preview
[img]
Preview
Text
1314105010-Presentation.pdf - Presentation

Download (915kB) | Preview

Abstract

Minyak dan gas bumi dapat diambil secara langsung melalui sumursumur yang dibuat, namun sumur-sumur tersebut tidak akan menghasilkan jumlah minyak dan gas bumi yang konstan setiap hari. Ketika kandungan minyak dan gas mulai turun maka yang harus dilakukan adalah memberikan treatment terhadap sumur tersebut, sehingga minyak dan gas yang masih terkandung di dasar bumi bisa naik dengan jumlah yang lebih banyak. Tujuan dilakukannya penelitian ini adalah untuk membantu perusahaan dalam menganalisis jumlah produksi minyak dan gas bumi selama periode 14 hari selanjutnya, sehingga dapat diketahui apakah selama periode 14 hari selanjutnya diperlukan treatment terhadap sumur. Data yang digunakan adalah jumlah produksi minyak dan gas bumi pada platform “MK” pada tahun 2015. Pemodelan jumlah produksi minyak dan gas bumi dilakukan menggunakan tiga metode, yaitu ARIMA, neural network, dan Hibrida ARIMA-neural network. Hasil yang diperoleh berdasarkan analisis ketiga metode tersebut adalah pada jumlah produksi minyak bumi model terbaik diperoleh dari metode hibrida ARIMA-neural network, dengan hasil ramalan yang cenderung sama selama 14 hari yaitu 1961 barel. Sedangkan jumlah produksi gas bumi model terbaik diperoleh dari metode neural network, dengan ramalan produksi untuk 14 hari selanjutnya cenderung meningkat. ==================================================================================================== Petroleum and natural gas can be taken directly through the wells that were made, but the wells will not produce the amount of oil and gas is a constant every day. When petroleum and natural gas begin to fall then that should be done is to provide treatment to those wells, so that oil and gas are still contained in the bottom of the earth could rise with higher numbers. The purpose of this study is to assist companies in analyzing the amount of petroleum and natural gas production over the next 14 day period, so it can be known whether during the period of 14 d ays is required subsequent treatment of the wells. The data used is the number of petroleum and natural gas production on the platform "MK" in 2015. Modelling the number of oil and gas production is done using three methods, namely ARIMA, Neural Network, and Hybrid ARIMA-NN. The results obtained by the analysis of these three methods is the best model of petroleum production is obtained from a hybrid method ARIMA-NN, with production that tends same forecast for 14 days, about 1961 barel. Meanwhile, natural gas production is obtained from the best models Neural Network methods, with production forecast production for the next 14 days is likely to increase.

Item Type: Thesis (Undergraduate)
Additional Information: RSSt 519.535 Pra p 3100016065898
Uncontrolled Keywords: ARIMA, Hibrida ARIMA-Neural Network, Minyak dan Gas Bumi, Neural Network
Subjects: H Social Sciences > HB Economic Theory > Economic forecasting--Mathematical models.
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions: Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User: - Davi Wah
Date Deposited: 25 Feb 2020 07:42
Last Modified: 25 Feb 2020 07:42
URI: https://repository.its.ac.id/id/eprint/75122

Actions (login required)

View Item View Item