Leveraging Large Language Models for Discrepancy Value Prediction in Custody Transfer Systems: A Comparative Analysis of Probabilistic and Point Forecasting Approaches

Discrepancies in custody transfer systems in the oil and gas industry pose significant financial, regulatory, and operational risks. Accurate prediction of these discrepancies is critical to optimizing operations and minimizing potential losses. This study evaluates the effectiveness of Large Langua...

Full description

Saved in:
Bibliographic Details
Main Authors: Fiki Hidayat, Arbi Haza Nasution, Fajril Ambia, Dike Fitriansyah Putra, Mulyandri
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10964215/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850148363108876288
author Fiki Hidayat
Arbi Haza Nasution
Fajril Ambia
Dike Fitriansyah Putra
Mulyandri
author_facet Fiki Hidayat
Arbi Haza Nasution
Fajril Ambia
Dike Fitriansyah Putra
Mulyandri
author_sort Fiki Hidayat
collection DOAJ
description Discrepancies in custody transfer systems in the oil and gas industry pose significant financial, regulatory, and operational risks. Accurate prediction of these discrepancies is critical to optimizing operations and minimizing potential losses. This study evaluates the effectiveness of Large Language Models (LLMs), specifically the Chronos-FineTuning Amazon Chronos T5 Small model, alongside statistical, machine learning, and deep learning models, in both probabilistic and point forecasting tasks. The evaluation covers metrics such as Weighted Quantile Loss (WQL), Scaled Quantile Loss (SQL), Mean Absolute Error (MAE), Symmetric Mean Absolute Percentage Error (SMAPE), and Root Mean Square Error (RMSE). The results highlight the superior performance of the Chronos model in both forecasting paradigms, demonstrating its ability to capture uncertainty and deliver precise predictions. This research offers valuable insights into selecting forecasting methodologies to improve custody transfer operations, underscoring the transformative potential of LLMs in industrial applications.
format Article
id doaj-art-e16f99061ed3415a96f75a63b731ba09
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-e16f99061ed3415a96f75a63b731ba092025-08-20T02:27:16ZengIEEEIEEE Access2169-35362025-01-0113656436565810.1109/ACCESS.2025.356025410964215Leveraging Large Language Models for Discrepancy Value Prediction in Custody Transfer Systems: A Comparative Analysis of Probabilistic and Point Forecasting ApproachesFiki Hidayat0https://orcid.org/0000-0003-1407-8952Arbi Haza Nasution1https://orcid.org/0000-0001-6283-3217Fajril Ambia2https://orcid.org/0000-0002-3905-4300Dike Fitriansyah Putra3https://orcid.org/0000-0002-6772-130X Mulyandri4Department of Petroleum Engineering, Universitas Islam Riau, Pekanbaru, Riau, IndonesiaDepartment of Informatics Engineering, Universitas Islam Riau, Pekanbaru, Riau, IndonesiaDepartment of Petroleum Engineering, Universitas Islam Riau, Pekanbaru, Riau, IndonesiaDepartment of Petroleum Engineering, Universitas Islam Riau, Pekanbaru, Riau, IndonesiaPT Pertamina Hulu Rokan, Pekanbaru, Riau, IndonesiaDiscrepancies in custody transfer systems in the oil and gas industry pose significant financial, regulatory, and operational risks. Accurate prediction of these discrepancies is critical to optimizing operations and minimizing potential losses. This study evaluates the effectiveness of Large Language Models (LLMs), specifically the Chronos-FineTuning Amazon Chronos T5 Small model, alongside statistical, machine learning, and deep learning models, in both probabilistic and point forecasting tasks. The evaluation covers metrics such as Weighted Quantile Loss (WQL), Scaled Quantile Loss (SQL), Mean Absolute Error (MAE), Symmetric Mean Absolute Percentage Error (SMAPE), and Root Mean Square Error (RMSE). The results highlight the superior performance of the Chronos model in both forecasting paradigms, demonstrating its ability to capture uncertainty and deliver precise predictions. This research offers valuable insights into selecting forecasting methodologies to improve custody transfer operations, underscoring the transformative potential of LLMs in industrial applications.https://ieeexplore.ieee.org/document/10964215/Probabilistic time-series forecastinglarge language modelsdiscrepancycustody transfer system
spellingShingle Fiki Hidayat
Arbi Haza Nasution
Fajril Ambia
Dike Fitriansyah Putra
Mulyandri
Leveraging Large Language Models for Discrepancy Value Prediction in Custody Transfer Systems: A Comparative Analysis of Probabilistic and Point Forecasting Approaches
IEEE Access
Probabilistic time-series forecasting
large language models
discrepancy
custody transfer system
title Leveraging Large Language Models for Discrepancy Value Prediction in Custody Transfer Systems: A Comparative Analysis of Probabilistic and Point Forecasting Approaches
title_full Leveraging Large Language Models for Discrepancy Value Prediction in Custody Transfer Systems: A Comparative Analysis of Probabilistic and Point Forecasting Approaches
title_fullStr Leveraging Large Language Models for Discrepancy Value Prediction in Custody Transfer Systems: A Comparative Analysis of Probabilistic and Point Forecasting Approaches
title_full_unstemmed Leveraging Large Language Models for Discrepancy Value Prediction in Custody Transfer Systems: A Comparative Analysis of Probabilistic and Point Forecasting Approaches
title_short Leveraging Large Language Models for Discrepancy Value Prediction in Custody Transfer Systems: A Comparative Analysis of Probabilistic and Point Forecasting Approaches
title_sort leveraging large language models for discrepancy value prediction in custody transfer systems a comparative analysis of probabilistic and point forecasting approaches
topic Probabilistic time-series forecasting
large language models
discrepancy
custody transfer system
url https://ieeexplore.ieee.org/document/10964215/
work_keys_str_mv AT fikihidayat leveraginglargelanguagemodelsfordiscrepancyvaluepredictionincustodytransfersystemsacomparativeanalysisofprobabilisticandpointforecastingapproaches
AT arbihazanasution leveraginglargelanguagemodelsfordiscrepancyvaluepredictionincustodytransfersystemsacomparativeanalysisofprobabilisticandpointforecastingapproaches
AT fajrilambia leveraginglargelanguagemodelsfordiscrepancyvaluepredictionincustodytransfersystemsacomparativeanalysisofprobabilisticandpointforecastingapproaches
AT dikefitriansyahputra leveraginglargelanguagemodelsfordiscrepancyvaluepredictionincustodytransfersystemsacomparativeanalysisofprobabilisticandpointforecastingapproaches
AT mulyandri leveraginglargelanguagemodelsfordiscrepancyvaluepredictionincustodytransfersystemsacomparativeanalysisofprobabilisticandpointforecastingapproaches