<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE root>
<article xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:ali="http://www.niso.org/schemas/ali/1.0/" article-type="research-article" dtd-version="1.2" xml:lang="en"><front><journal-meta><journal-id journal-id-type="publisher-id">RUDN Journal of Economics</journal-id><journal-title-group><journal-title xml:lang="en">RUDN Journal of Economics</journal-title><trans-title-group xml:lang="ru"><trans-title>Вестник Российского университета дружбы народов. Серия: Экономика</trans-title></trans-title-group></journal-title-group><issn publication-format="print">2313-2329</issn><issn publication-format="electronic">2408-8986</issn><publisher><publisher-name xml:lang="en">Peoples’ Friendship University of Russia named after Patrice Lumumba (RUDN University)</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="publisher-id">50624</article-id><article-id pub-id-type="doi">10.22363/2313-2329-2026-34-1-123-159</article-id><article-id pub-id-type="edn">TTTYIQ</article-id><article-categories><subj-group subj-group-type="toc-heading" xml:lang="en"><subject>Developed and developing countries economy</subject></subj-group><subj-group subj-group-type="toc-heading" xml:lang="ru"><subject>Экономика развитых и развивающихся стран</subject></subj-group><subj-group subj-group-type="article-type"><subject>Research Article</subject></subj-group></article-categories><title-group><article-title xml:lang="en">LSTM forecasting of Indonesia’s oil supply-demand deficit</article-title><trans-title-group xml:lang="ru"><trans-title>Прогнозирование дефицита спроса и предложения нефти в Индонезии</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0001-8679-7767</contrib-id><contrib-id contrib-id-type="researcherid">AAC-4800-2021</contrib-id><name-alternatives><name xml:lang="en"><surname>Muljono</surname><given-names>Wiryanta</given-names></name><name xml:lang="ru"><surname>Муджоно</surname><given-names>Вирьянта</given-names></name></name-alternatives><bio xml:lang="en"><p>PhD, senior researcher in digital economics, Ministry of Communication and Digital of the Republic of Indonesia; associate professor of energy economics, Universitas Sebelas Maret</p></bio><bio xml:lang="ru"><p>доктор философии, старший научный сотрудник в области цифровой экономики, Министерство связи и цифровых технологий Республики Индонезия; доцент кафедры экономики энергетики, Университет Себелас Марет</p></bio><email>wiryantamuljono@gmail.com</email><xref ref-type="aff" rid="aff1"/><xref ref-type="aff" rid="aff2"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0004-4639-0961</contrib-id><name-alternatives><name xml:lang="en"><surname>Setyanto</surname><given-names>Padmanabha Adyaksa</given-names></name><name xml:lang="ru"><surname>Сетьянто</surname><given-names>Падманабха Адьякша</given-names></name></name-alternatives><bio xml:lang="en"><p>student, School of Electrical Engineering and Informatics (Computing)</p></bio><bio xml:lang="ru"><p>студент Школы электротехники и информатики (Вычислительной техники), факультета электротехники и информатики (вычислительная техника)</p></bio><email>padmanabhaadyaksasetyanto@gmail.com</email><xref ref-type="aff" rid="aff3"/></contrib></contrib-group><aff-alternatives id="aff1"><aff><institution xml:lang="en">Ministry of Communication and Digital of the Republic of Indonesia</institution></aff><aff><institution xml:lang="ru">Министерство связи и цифровых технологий Республики Индонезии</institution></aff></aff-alternatives><aff-alternatives id="aff2"><aff><institution xml:lang="en">Universitas Sebelas Maret (UNS)</institution></aff><aff><institution xml:lang="ru">Университет Себелас Марет (UNS)</institution></aff></aff-alternatives><aff-alternatives id="aff3"><aff><institution xml:lang="en">Institut Teknologi Bandung</institution></aff><aff><institution xml:lang="ru">Бандунгский технологический институт (ITB)</institution></aff></aff-alternatives><pub-date date-type="pub" iso-8601-date="2026-06-15" publication-format="electronic"><day>15</day><month>06</month><year>2026</year></pub-date><volume>34</volume><issue>1</issue><issue-title xml:lang="en">NEW VECTORS OF TRADE AND INVESTMENT WITHIN BRICS+</issue-title><issue-title xml:lang="ru">НОВЫЕ ВЕКТОРЫ ТОРГОВЛИ И ИНВЕСТИЦИЙ В РАМКАХ БРИКС+</issue-title><fpage>123</fpage><lpage>159</lpage><history><date date-type="received" iso-8601-date="2026-06-15"><day>15</day><month>06</month><year>2026</year></date></history><permissions><copyright-statement xml:lang="en">Copyright ©; 2026, Muljono W., Setyanto P.A.</copyright-statement><copyright-statement xml:lang="ru">Copyright ©; 2026, Муджоно В., Сетьянто П.А.</copyright-statement><copyright-year>2026</copyright-year><copyright-holder xml:lang="en">Muljono W., Setyanto P.A.</copyright-holder><copyright-holder xml:lang="ru">Муджоно В., Сетьянто П.А.</copyright-holder><ali:free_to_read xmlns:ali="http://www.niso.org/schemas/ali/1.0/"/><license><ali:license_ref xmlns:ali="http://www.niso.org/schemas/ali/1.0/">https://creativecommons.org/licenses/by-nc/4.0</ali:license_ref></license></permissions><self-uri xlink:href="https://journals.rudn.ru/economics/article/view/50624">https://journals.rudn.ru/economics/article/view/50624</self-uri><abstract xml:lang="en"><p>Analysis of historical market data from 2020 to 2024 reveals a profound structural shift in Indonesia’s energy landscape: domestic crude production plummeted by 17.96%, while oil demand simultaneously surged by 14.57%. This widening supply-demand gap has severely intensified national energy security concerns, pushing Indonesia’s net deficit to approximately 1.0 million barrels per day (bpd). This critical gap directly pressures fiscal stability due to escalating fuel subsidy costs and exposes the macroeconomy to global price shocks. Given the market’s inherent non-linearity, driven by WTI price volatility and frequent, policy-led structural shifts (e.g., the deployment of AI for subsidized fuel control and the B40 biodiesel mandate), traditional linear models like ARIMA are severely limited in their predictive accuracy. We propose the Long Short-Term Memory (LSTM) deep learning network as a methodologically superior approach. The LSTM’s recurrent architecture, with its specialized gate mechanisms, is uniquely suited to capture the non-linear dynamics and long-term temporal dependencies of complex energy time series data. To ensure the reliability of our findings, model robustness was explicitly ensured via k-fold cross-validation and a thorough discussion of inherent dataset size limitations was provided, directly addressing methodological concerns. Empirical findings confirm the LSTM model’s significant superiority over the conventional benchmark, achieving a 57.24% reduction in Mean Absolute Percentage Error (MAPE) and significantly lower Root Mean Square Error (RMSE) compared to the ARIMA baseline. This high-precision forecast provides critical foresight for Indonesian policymakers, enabling proactive management of fiscal risk, targeted adjustments to foreign exchange reserves, and the successful acceleration of the national Indonesia Oil and Gas (IOG) 4.0 strategy toward long-term energy resilience.</p></abstract><trans-abstract xml:lang="ru"><p>Анализ исторических рыночных данных с 2020 по 2024 г. выявляет глубокий структурный сдвиг в энергетическом ландшафте Индонезии: внутренняя добыча сырой нефти упала на 17,96 %, в то время как спрос на нефть одновременно вырос на 14,57 %. Этот увеличивающийся разрыв между спросом и предложением серьезно обострил проблемы национальной энергетической безопасности, доведя чистый дефицит Индонезии примерно до 1 миллиона баррелей в сутки (б/с). Этот критический разрыв напрямую влияет на фискальную стабильность из-за роста расходов на субсидирование топлива и подвергает макроэкономику воздействию глобальных ценовых шоков. Учитывая присущую рынку нелинейность, обусловленную волатильностью цен на нефть марки WTI и частыми структурными сдвигами из-за политических решений (например, внедрение искусственного интеллекта для контроля за субсидируемым топливом и введение обязательного спроса на биодизельное топливо класса B40), традиционные линейные модели, такие как ARIMA, существенно ограничены в точности прогнозирования. Предложена сеть глубокого обучения с долговременной краткосрочной памятью (LSTM) в качестве методологически более совершенного подхода. Рекуррентная архитектура LSTM с ее специализированными механизмами вентилей идеально подходит для учета нелинейной динамики и долгосрочных зависимостей сложных динамических рядов данных по энергетике. Для обеспечения надежности результатов устойчивость модели была явно обеспечена с помощью k-кратной кросс-валидации, проведено подробное обсуждение присущих модели ограничений размера набора данных, что напрямую затрагивало методологические вопросы. Эмпирические результаты подтверждают значительное превосходство модели LSTM над традиционным эталонным тестом, достигая 57,24 % снижения средней абсолютной процентной ошибки (MAPE) и значительно более низкой среднеквадратической ошибки (RMSE) по сравнению с базовым тестом ARIMA. Этот высокоточный прогноз предоставляет индонезийским политикам критически важную информацию для прогнозирования, позволяя проактивно управлять фискальными рисками, целенаправленно корректировать валютные резервы и успешно ускорять реализацию национальной стратегии Индонезии «Нефть и газ (IOG) 4.0» в целях обеспечения долгосрочной энергетической устойчивости.</p></trans-abstract><kwd-group xml:lang="en"><kwd>relational sovereignty</kwd><kwd>neural network applications</kwd><kwd>fossil fuel consumption trends</kwd><kwd>energy security modeling</kwd><kwd>B40 biodiesel impact</kwd><kwd>geoeconomic resilience</kwd></kwd-group><kwd-group xml:lang="ru"><kwd>структурный дефицит</kwd><kwd>LSTM прогнозирование</kwd><kwd>дисбаланс спроса и предложения</kwd><kwd>анализ динамики временных рядов</kwd><kwd>энергетическая политика</kwd><kwd>индонезий-</kwd></kwd-group><funding-group/></article-meta><fn-group/></front><body></body><back><ref-list><ref id="B1"><label>1.</label><mixed-citation>Aditya, I.A., Wijayanto, T., &amp; Hakam, D.F. (2025). Advancing renewable energy in Indonesia: A comprehensive analysis of challenges, opportunities, and strategic solutions. Sustainability, 17(5), 2216. https://doi.org/10.3390/su17052216 EDN: OODZQX</mixed-citation></ref><ref id="B2"><label>2.</label><mixed-citation>Ahmad, T., Khan, M.J., &amp; Javed, I. (2022). Artificial intelligence applications in the oil and gas industry: A review. Journal of Petroleum Science and Engineering, 212, 110292. https://doi.org/10.1016/j.petrol.2022.110292</mixed-citation></ref><ref id="B3"><label>3.</label><mixed-citation>Arab, I., &amp; Benitez, R. (2025). Benchmarking time series forecasting models: From statistical techniques to foundation models in real-world applications [Preprint]. https://doi.org/10.48550/arXiv.2502.03395</mixed-citation></ref><ref id="B4"><label>4.</label><mixed-citation>Box, G.E.P., Jenkins, G.M., Reinsel, G.C., &amp; Ljung, G.M. (2015). Time series analysis: Forecasting and control. Wiley.</mixed-citation></ref><ref id="B5"><label>5.</label><mixed-citation>Dave, E., Leonardo, A., Jeanice, M., &amp; Hanafiah, N. (2021). Forecasting Indonesia exports using a hybrid model ARIMA-LSTM. Procedia Computer Science, 179, 480–487. https://doi.org/10.1016/j.procs.2021.01.031</mixed-citation></ref><ref id="B6"><label>6.</label><mixed-citation>Fattah, J., Ezzine, L., Aman, Z., El Moussami, H., &amp; Lachhab, A. (2018). Forecasting of demand using ARIMA model. International Journal of Engineering Business Management, 10, 1–9. https://doi.org/10.1177/1847979018808673</mixed-citation></ref><ref id="B7"><label>7.</label><mixed-citation>Fathin, M.R., Widhiyasana, Y., &amp; Syakrani, N. (2021). Model for predicting electrical energy consumption using ARIMA method. Advances in Engineering Research, 207, 298–303. https://doi.org/10.2991/aer.k.211106.047</mixed-citation></ref><ref id="B8"><label>8.</label><mixed-citation>Goodfellow, I., Bengio, Y., &amp; Courville, A. (2016). Deep learning. MIT Press.</mixed-citation></ref><ref id="B9"><label>9.</label><mixed-citation>Jo, Y., Sorkhabi, R.B., &amp; Panja, P. (2026). Machine learning for forecasting production in tight oil reservoirs: Application to the Bakken formation of Williston basin. Unconventional Resources, 6, Article 100335. https://doi.org/10.1016/j.uncres.2026.100335</mixed-citation></ref><ref id="B10"><label>10.</label><mixed-citation>Hanan, A., Swastika, M.B., &amp; Rohmawati, H.M. (2024). Effective downstream policy strategies for oil, gas, and mining sub-sectors in Indonesia (Policy Brief No. 5). Purnomo Yusgiantoro Center. https://doi.org/10.33116/pyc-br-5</mixed-citation></ref><ref id="B11"><label>11.</label><mixed-citation>Hermansah, M., &amp; Canas Rodrigues, P. (2024). Indonesian inflation forecasting with recurrent neural network long short-term memory (RNN-LSTM). Enthusiastic: International Journal of Applied Statistics and Data Science, 4(2), 132–142. https://doi.org/10.20885/enthusiastic.vol4.iss2.art5</mixed-citation></ref><ref id="B12"><label>12.</label><mixed-citation>Hochreiter, S., &amp; Schmidhuber, J. (1997). Long Short-Term Memory. Neural Computation, 9(8), 1735–1780. https://www.cri.minesparis.psl.eu/people/silber/cours/2025/nlp/bib/HochreiterS1997.pdf</mixed-citation></ref><ref id="B13"><label>13.</label><mixed-citation>Izudin, N.E.M., Sokkalingam, R., Daud, H., Mardesci, H., &amp; Husin, A. (2021). Forecasting electricity consumption in Malaysia by hybrid ARIMA-ANN. In S.A. Abdul Karim et al. (Eds.), Proceedings of the 6th International Conference on Fundamental and Applied Sciences (pp. 749–760). Springer. https://doi.org/10.1007/978-981-16-4513-6_66</mixed-citation></ref><ref id="B14"><label>14.</label><mixed-citation>Kumoro, M.N.P., Ainin, A.Q., Addinie, I.A., &amp; Nengkoda, A. (2022). A strategy to achieve 1 MMbopd national oil production target by implementation of artificial intelligence algorithm to accelerate the exploration process. In Proceedings of the Indonesian Petroleum Association, 46th Annual Convention &amp; Exhibition 2022. Indonesian Petroleum Association. https://doi.org/10.29118/IPA22-SE-157</mixed-citation></ref><ref id="B15"><label>15.</label><mixed-citation>Lai, G., Chang, W.-C., Yang, Y., &amp; Liu, H. (2018). Modeling long- and short-term temporal patterns with deep neural networks. In Proceedings of the 41st International ACM SIGIR Conference on Research &amp; Development in Information Retrieval (pp. 95–104). https://doi.org/10.1145/3209978.3210006</mixed-citation></ref><ref id="B16"><label>16.</label><mixed-citation>Mayasari, F., &amp; Dalimi, R. (2019). Fuel oil supply demand projection and planning in Indonesia using system dynamics modeling. International Journal of Smart Grid and Clean Energy, 8(1), 11–21. https://doi.org/10.12720/sgce.8.1.11-21</mixed-citation></ref><ref id="B17"><label>17.</label><mixed-citation>Mukhacheva, A.V. (2025). Econometric modeling of digital development of the economy and social sphere of Russian regions. π-Economy, 18(4), 7–31. (In Russ.). https://doi.org/10.18721/JE.18401 EDN: GYXBSF</mixed-citation></ref><ref id="B18"><label>18.</label><mixed-citation>Muljono, W., &amp; Setiyawati, S. (2022). Digital economy: The main power for digital industry in Indonesia. International Journal of Technological Globalisation, 15(4), 423–444. https://doi.org/10.1504/IJTGM.2022.125908 EDN: NYMZBZ</mixed-citation></ref><ref id="B19"><label>19.</label><mixed-citation>Nikitins, I. (2022). Innovative approaches to measure the business environment uncertainty and tools to combat it. Russian Journal of Innovation Economics, 12(1), 223–236. (In Russ.). https://doi.org/10.18334/vinec.12.1.114193 EDN: CWSFHI</mixed-citation></ref><ref id="B20"><label>20.</label><mixed-citation>Odimarha, A.C., Ayodeji, S.A., &amp; Abaku, E.A. (2024). Machine learning’s influence on supply chain and logistics optimization in the oil and gas sector: A comprehensive analysis. Computer Science &amp; IT Research Journal, 5(3), 725–740. https://doi.org/10.51594/csitrj.v5i3.976</mixed-citation></ref><ref id="B21"><label>21.</label><mixed-citation>Patria, N., Irawanto, B., &amp; Abrar, A.N. (2024). “(Don’t) stop the rising oil price”: Mediatization, digital discourse, and fuel price controversies in Indonesian online media. Journalism and Media, 6(3), 124. https://doi.org/10.3390/journalmedia6030124 EDN: GLAPIA</mixed-citation></ref><ref id="B22"><label>22.</label><mixed-citation>Resosudarmo, B.P. (2011). Energy security in Indonesia. In L. Anceschi &amp; J. Symons (Eds.), Energy security in the era of climate change (pp. 161–179). Palgrave Macmillan. https://doi.org/10.1057/9780230355361.0019</mixed-citation></ref><ref id="B23"><label>23.</label><mixed-citation>Ribeiro, A.M.N.C., do Carmo, P.R.X., Rodrigues, I.R., Sadok, D., Lynn, T., &amp; Endo, P.T. (2020). Short-term firm-level energy-consumption forecasting for energy-intensive manufacturing: A comparison of machine learning and deep learning models. Algorithms, 13(11), Article 274. https://doi.org/10.3390/a13110274</mixed-citation></ref><ref id="B24"><label>24.</label><mixed-citation>Rizvi, M.F. (2024). ARIMA model time series forecasting. International Journal for Research in Applied Science &amp; Engineering Technology (IJRASET), 12(5), 3782–3785. https://doi.org/10.22214/ijraset.2024.62416 EDN: UMYIEF</mixed-citation></ref><ref id="B25"><label>25.</label><mixed-citation>Safrina, R., Kilmi, A., Pradiva, Z.A., &amp; Sohirin, A. (2025). Enabling a resilient and sustainable ASEAN energy future through digitalisation (Policy Brief No. 12). ASEAN Centre for Energy (ACE). https://aseanenergy.org/enabling-a-resilient-and-sustainable-asean-energy-future-through-digitalisation/</mixed-citation></ref><ref id="B26"><label>26.</label><mixed-citation>Sahara, Dermawan, A., Amaliah, S., Irawan, T., &amp; Dilla, S. (2022). Economic impacts of biodiesel policy in Indonesia: A computable general equilibrium approach. Journal of Economic Structures, 11(1), 22. https://doi.org/10.1186/s40008-022-00281-9 EDN: RZPQPO</mixed-citation></ref><ref id="B27"><label>27.</label><mixed-citation>Setiabudi, P. (2026, January 10). B40 biodiesel consumption exceeds target in 2025, cuts diesel imports. Petromindo. https://www.petromindo.com/news/article/b40-biodiesel-consumption-exceeds-target-in-2025-cuts-diesel-imports</mixed-citation></ref><ref id="B28"><label>28.</label><mixed-citation>Smagulova, K., &amp; James, A.P. (2019). A survey on LSTM memristive neural network architectures and applications. The European Physical Journal Special Topics, 228, 2313–2324. https://doi.org/10.1140/epjst/e2019-900046-x EDN: HRKIKB</mixed-citation></ref><ref id="B29"><label>29.</label><mixed-citation>Syarovy, M., Nugroho, A.P., Sutiarso, L., Suwardi, Muna, M.S., Wiratmoko, A., Sukarman, &amp; Primananda, S. (2023). Prediction of oil palm production using recurrent neural network long short-term memory (RNN-LSTM). In J. Sumantyo et al. (Eds.), Proceedings of the International Conference on Smart and Innovative Agriculture (ICoSIA 2022) (pp. 55–66). Atlantis Press (Springer Nature). https://doi.org/10.2991/978-94-6463-122-7_6</mixed-citation></ref><ref id="B30"><label>30.</label><mixed-citation>Wibowo, A., Astuti, W., Wibowo, T.P., &amp; Purnomo, H. (2022). Estimation of energy demand in Indonesia using artificial neural network. International Journal of Energy Economics and Policy, 12(6), 261–271. https://doi.org/10.32479/ijeep.11390 EDN: ZFVHWU</mixed-citation></ref><ref id="B31"><label>31.</label><mixed-citation>Wijaya, I.G.P.O.I., Kristi, M.S., Maharani, N.S., &amp; Raharjo, J. (2025). Data adjustment for ARIMA method to investigate the impact of pandemic on electricity consumption prediction in Indonesia. Emitor: Jurnal Teknik Elektro, 25(1), 47–53. https://doi.org/10.23917/emitor.v25i1.8676</mixed-citation></ref><ref id="B32"><label>32.</label><mixed-citation>Zega, A.S.R., Hidayat, A.K., Jannah, N.T., &amp; Kartiasih, F. (2024). Selecting the best model for forecasting Indonesia’s oil and gas import value using ARIMAX and ARIMAX-LSTM. Dynamic Management Journal, 8(4), 912–941. https://doi.org/10.31000/dmj.v8i4 EDN: DLITGE</mixed-citation></ref><ref id="B33"><label>33.</label><mixed-citation>Zeinula, A., Khan, Q., Kalam, S., &amp; Riazi, M. (2025). Hybrid machine learning for oil production forecasting using random forest and long short-term memory models [Paper presentation]. SPE Annual Caspian Technical Conference and Exhibition, Baku, Azerbaijan. https://doi.org/10.2118/230427-MS</mixed-citation></ref><ref id="B34"><label>34.</label><mixed-citation>Zhuang, X.Y., Wang, W.D., Su, Y.L., Dai, Z.X., &amp; Yan, B.C. (2025). Deep learning-assisted optimization for enhanced oil recovery and CO2 sequestration considering gas channeling constraints. Petroleum Science, 22(8), 3397–3417. https://doi.org/10.1016/j.petsci.2025.04.028</mixed-citation></ref></ref-list></back></article>
