CD Skripsi

Klasifikasi Dan Monitoring Performa Sumur Menggunakan Algoritma Random Forest Pada Pt. Pertamina Hulu Rokan

---

Manual monitoring of oil well performance has limitations in efficiency and accuracy when dealing with field condition changes. This study applies the Random Forest algorithm to build a well performance classification model based on historical data. The dataset consists of 259,923 records from well compliance tests in the X Field of WK Rokan, containing technical parameters such as total oil, fluid, water ratio, and production efficiency. The model's performance was evaluated using accuracy, precision, recall, and f1-score metrics to classify well conditions into three categories: declining, stable, and improving. The best experiment was achieved using a 90:10 data split, with an accuracy of 92.57%, precision of 96.09% (stable), 89.61% (improving), and 86.17% (declining), recall of 93.84% (stable), 93.16% (improving), and 88.46% (declining), and f1-score of 94.95% (stable), 91.35% (improving), and 87.30% (declining). The model was also implemented to predict new data and generate technical recommendations based on the classification results.

Keywords: Classification, Historical Data, Monitoring, Oil Well Performance, Random Forest

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

2103111393

Penerbit

Pekanbaru : Universitas Riau FMIPA Sistem Informasi., 2025

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

2103111393

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

SISTEM INFORMASI

Info Detail Spesifik

-

Pernyataan Tanggungjawab

Mutia

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar