Claim Missing Document
Check
Articles

Enhancing Rainfall Forecasting Performance in Bandung City Using Bi-LSTM with Grid Search Optimization on Gregorian and Lunar Calendar Data Yunizar, Mahdayani Putri; Talakua, Andrew Hosea; Darmawan, Gumgum
Parameter: Jurnal Matematika, Statistika dan Terapannya Vol 4 No 3 (2025): Parameter: Jurnal Matematika, Statistika dan Terapannya
Publisher : Jurusan Matematika FMIPA Universitas Pattimura

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30598/parameterv4i3pp595-601

Abstract

Rainfall is a climatic factor that strongly influences human activities and plays a crucial role in decision making related to water resources, mobility, and disaster preparedness. High rainfall intensity may escalate into hydrometeorological hazards, underscoring the importance of accurate rainfall forecasting to support early warning and mitigation efforts. This study aims to compare the forecasting accuracy of monthly rainfall predictions between the Gregorian and lunar calendars using the Bidirectional Long Short-Term Memory (Bi-LSTM) model optimized through a grid search approach. The method is designed to capture temporal patterns arising from the distinct structures of two asynchronous calendars. Daily rainfall data from Bandung City, Indonesia, covering the period from 2000 to 2025, were converted into monthly series in both calendar systems. The results reveal that the Gregorian calendar provides significantly better forecasting performance, achieving the lowest MAPE value of 11.60 percent at the three-month horizon. In contrast, the lunar calendar shows higher variability and reaches its best MAPE of 31.43 percent at the same horizon. These findings indicate that the Gregorian calendar offers a more stable temporal representation for rainfall forecasting in Bandung and supports improved predictive modeling for climate-related decision making.
Mathematical Modeling and Regime Dynamics of Ammonium Chloride Imports in Indonesia Using Threshold Vector Autoregressive Integrated (TVARI) Approach Widiantoro, Carissa Egytia; Arisanti, Restu; Darmawan, Gumgum
CAUCHY: Jurnal Matematika Murni dan Aplikasi Vol 11, No 1 (2026): CAUCHY: JURNAL MATEMATIKA MURNI DAN APLIKASI
Publisher : Mathematics Department, Universitas Islam Negeri Maulana Malik Ibrahim Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.18860/cauchy.v11i1.40052

Abstract

Ammonium chloride plays an important role in Indonesia as a key raw material for NPK fertilizers and chemical industries. Despite its importance, domestic production remains limited, and potential supply from by-product sources has not been utilized effectively. Consequently, Indonesia depends heavily on imports sourced mainly from a single country. This situation creates vulnerabilities in industrial supply chain and highlights the need for a clearer understanding of import volume and value dynamics. This study employs a Threshold Vector Autoregressive Integrated (TVARI) model to capture nonlinear and regime-dependent adjustments in the joint dynamics of import volume and value, with import volume growth as the threshold variable. The approach's novelty lies in its ability to accommodate structural changes that cannot be adequately represented by a single linear specification. Empirical results identify two statistically distinct regimes defined by whether import growth lies below or above an estimated threshold. In the first regime, where growth is below the threshold, short-run dynamics are primarily driven by changes in import value, indicating price-related adjustments. In the second regime, import volume exhibits stronger responses, reflecting quantity adjustments associated with supply-side conditions. These findings demonstrate that linear models are insufficient to capture asymmetric adjustment mechanisms in import behavior. By providing a formal mathematical description of regime-dependent dynamics, this study contributes to a deeper understanding of Indonesia’s industrial import structure and offers insights for data-informed supply chain planning. The results support policy discussions related to Sustainable Development Goals 8 (economic stability), 9 (industry resilience), and 12 (responsible consumption and production).
Co-Authors Achmad Bachrudin Akbar, Muhammad Faizal Alamanda Putri, Fariza Aldi Anugerah Sitepu Alfarisi, Widi Wildani Alifia, Wanda Aliya Auliyazhafira, Shabira Amanah Dwiadi, Qurnia Angga Pratama Anindya Apriliyanti Pravitasari Apriliana, Linda Aribah, Rana Asrirawan Aurilia Pratiwi, Dhanti Azka Larissa Rahayu Bertho Tantular Budhi Handoko Budi Nurani Ruchjana Budianti, Laila Clarissa Clorinda, Chrysentia Dedi Rosadi Defi Yusti Faidah Deltha Airuzsh Lubis Dina Prariesa Eko Yulian eko yulian, eko Ery Sadewo, Ery Fajar Indrayatna Farhan Bagus Prakoso Ferdian Agustiana Fitriani Azuri, Dila Hadi, Juandi Haura, Zhafira Hirlan Khaeri I Gede Nyoman Mindra Jaya Indriani , Ayu Intan Nurma Yulita Ismatilah, Nuzila Janatin, Janatin Karin, Nabila Khaeri, Hirlan Kiki Amelia, Kiki Kusuma Putri, Aisha Muhamad Budiman Johra Muhammad Faizal Akbar Mulya Nurmansyah Ardisasmita Mulya, Callista Audrey Najwa, Sandrina Neneng Sunengsih Neneng Sunengsih Novianti Indah Putri Nurhapilah, Hani Nurul Gusriani Pian Widianingsih Puteri, Dian Islamiaty Putri Syallya, Najma Rafifah Putri, Salma Azzahra Rafidah, Raihanah Rahman Al Madan, Aulia Resa Septiani Pontoh Restu Arisanti Rhafi Ahdian, Muhammad Rina Sri Kalsum Siregar Rini Luciani Rahayu Rizal Amegia Saputra Ruchjana, Budi N Ruslan Ruslan Samaria Nauli, Theresia Sangrila, Ayu Sastradipraja, C K Setialaksana, Wirawan - Sitepu, Aldi Anugerah Sitohang, Yosep Oktavianus Sri Sutjiningtyas Sri Winarni Sri Yuliana Sudartianto, Sudartianto Talakua, Andrew Hosea Tri Wulanda Fitri Triyani Hendrawati Utami, Yosi Febria Widiantoro, Carissa Egytia Widodo, Valeno Glenedias Wildani Alfarisi, Widi Yasyfi Avicenna, Muhammad Yeny Krista Franty Yogo Aryo Jatmiko Yosep Oktavianus Sitohang Yunizar, Mahdayani Putri Yusep Suparman Yuyun Hidayat Zen Munawar Zulhanif Zulhanif