Proceedings of the 11th International Applied Business and Engineering Conference, ABEC 2023, September 21st, 2023, Bengkalis, Riau, Indonesia

Research Article

Optimization of Oil Well Operations with Machine Learning Model to predict Well Production at Company "XYZ"

Download37 downloads
  • @INPROCEEDINGS{10.4108/eai.21-9-2023.2342918,
        author={Yasser  Arapat and Juni Nurma Sari and Satria Perdana Arifin and Yohana Dewi Lulu Widyasari},
        title={Optimization of Oil Well Operations with Machine Learning Model to predict Well Production at Company "XYZ"},
        proceedings={Proceedings of the 11th International Applied Business and Engineering Conference, ABEC 2023, September 21st, 2023, Bengkalis, Riau, Indonesia},
        publisher={EAI},
        proceedings_a={ABEC},
        year={2024},
        month={2},
        keywords={petroleum reservoir arima ann},
        doi={10.4108/eai.21-9-2023.2342918}
    }
    
  • Yasser Arapat
    Juni Nurma Sari
    Satria Perdana Arifin
    Yohana Dewi Lulu Widyasari
    Year: 2024
    Optimization of Oil Well Operations with Machine Learning Model to predict Well Production at Company "XYZ"
    ABEC
    EAI
    DOI: 10.4108/eai.21-9-2023.2342918
Yasser Arapat1,*, Juni Nurma Sari1, Satria Perdana Arifin1, Yohana Dewi Lulu Widyasari1
  • 1: Politeknik Caltex Riau
*Contact email: yasser22mttk@mahasiswa.pcr.ac.id

Abstract

PT XYZ is a national oil company located in Minas, Siak Regency, Riau, Indonesia. Over time, the oil production process has decreased due to several reasons, one of which is the lack of pressure from the production well. So that at this time the wells that have decreased production are treated by injecting produced water into the reservoir using injection wells with the aim of increasing reservoir pressure so that the oil contained at the bottom of the earth will rise to the surface of the earth. Water that is produced with oil will be treated and injected back into the reservoir. Produced water is treated in accordance with the characteristics of the water contained in the reservoir, which, if different, will damage the reservoir. This research will discuss the application of data mining methods in the well production prediction process. The methods used are ARIMA and ANN, and the testing is done using RMSE and MAE. The results of this study show that the ANN method has higher accuracy results. This can be seen from the average accuracy results of each production with several treatments.