BENCHMARKING ARIMAX DAN SUPPORT VECTOR REGRESSION (SVR) UNTUK PREDIKSI BIOMASSA IKAN NILA DALAM SISTEM AKUAKULTUR PINTAR
DOI:
https://doi.org/10.52453/technologic.v17i1.519Keywords:
Support Vector Regression, Ikan Nila, ARIMAX, Smart Aquaculture, Internet of ThingsAbstract
Estimasi biomassa Ikan Nila secara presisi merupakan elemen krusial dalam sistem Smart Aquaculture untuk mengoptimalkan Feed Conversion Ratio dan mencegah overfeeding. Tantangan utama pemodelan ini adalah fluktuasi data deret waktu yang non-linear akibat dinamika kualitas air, serta keterbatasan jumlah sampel observasi lapangan (dataset kecil). Penelitian ini membandingkan kinerja algoritma statistik AutoRegressive Integrated Moving Average (ARIMAX) dan Support Vector Regression (SVR) dengan kernel Radial Basis Function (RBF) untuk memprediksi bobot ikan harian. Dataset dikumpulkan selama 11 minggu dari kolam intensif, meliputi variabel bobot, pakan kumulatif, dan Oksigen Terlarut (Dissolved Oxygen, DO). Validasi model menerapkan skema Walk-Forward Validation. Hasil pengujian menunjukkan SVR secara signifikan mengungguli ARIMAX, menghasilkan Mean Absolute Percentage Error (MAPE) sebesar 6.08% dan R² Score 0.9191, dibandingkan ARIMAX dengan MAPE 11.97% dan R² Score 0.7569. SVR terbukti lebih responsif menangkap anomali perlambatan laju pertumbuhan ikan yang dipicu fase stres hipoksia. Kesimpulannya, model SVR memiliki komputasi inferensi efisien sehingga prospektif ditanamkan pada mikrokontroler IoT berbasis ESP32 guna mengendalikan aktuator pemberian pakan otomatis secara dinamis.
References
Q. Ma, S. Li, H. Qi, X. Yang, and M. Liu, “Rapid Prediction and Inversion of Pond Aquaculture Water Quality Based on Hyperspectral Imaging by Unmanned Aerial Vehicles,” Water (Switzerland), vol. 17, no. 4, Feb. 2025, doi: 10.3390/w17040517.
B. Shi, X. Jin, Y. Hu, J. Jiang, and Y. Sun, “Precision prediction of aquaculture water quality: a spatiotemporal model integrating optimized-LSTM and radial basis function neural networks,” PeerJ Comput. Sci., vol. 12, p. e3515, Jan. 2026, doi: 10.7717/peerj-cs.3515.
S. Akter et al., “Correction: Efficacy of using plant ingredients as partial substitute of fishmeal in formulated diet for a commercially cultured fish, Labeo rohita (Frontiers in Sustainable Food Systems, (2024), 8, (1376112), 10.3389/fsufs.2024.1376112),” 2025, Frontiers Media SA. doi: 10.3389/fsufs.2025.1649055.
M. C. B. Rodríguez, C. E. M. González, E. D. C. Almanza, C. G. Rodríguez, J. Á. Regino-Vergara, and A. López-Padilla, “Benefits and challenges of the internet of things in aquaculture production: a literature review,” 2025, Frontiers Media SA. doi: 10.3389/fsufs.2025.1590153.
Dhea Mayang Saputri, Seto Windarto, and Diana Chilmawati, “Effect of Feeding Frequency on Feed Utilization Efficiency and Growth of Asian seabass (Lates calcarifer) Fingerlings,” Journal of Aquaculture and Fish Health, vol. 14, no. 3, pp. 438–464, Sep. 2025, doi: 10.20473/jafh.v14i3.67912.
D. Cujbescu et al., “TECHNICAL SOLUTIONS FOR BIOMASS ESTIMATION ACCORDING TO THE CONCEPT OF AQUACULTURE 4.0,” INMATEH - Agricultural Engineering, vol. 72, no. 1, pp. 663–678, 2024, doi: 10.35633/inmateh-72-59.
M. H. Saleem et al., “A comprehensive review of machine learning and deep learning models for non-intrusive load monitoring: performance, analyses, practical insights, and emerging trends,” Applied Intelligence, vol. 55, no. 15, Oct. 2025, doi: 10.1007/s10489-025-06921-4.
I. Sutrisno and M. B. Rahmat, “Design and implementation of control system for remotely surface vehicle in aquaculture environmental monitoring,” Edelweiss Applied Science and Technology, vol. 9, no. 7, pp. 1519–1539, Jul. 2025, doi: 10.55214/2576-8484.v9i7.8965.
K. Al-Saeedi, A. Fish, D. Zhou, K. Tsakiri, and A. Marsellos, “Multi-Scale Decomposition and Autocorrelation Modeling for Classical and Machine Learning-Based Time Series Forecasting,” Mathematics, vol. 14, no. 2, p. 283, Jan. 2026, doi: 10.3390/math14020283.
H. Fang, T. Li, and H. Xian, “Comparative Analysis of Machine/Deep Learning Models for Single-Step and Multi-Step Forecasting in River Water Quality Time Series,” Water (Switzerland), vol. 17, no. 13, Jul. 2025, doi: 10.3390/w17131866.
J. S. Fandiño Pelayo, L. S. Mendoza Castellanos, R. Cazes Ortega, and L. G. Hernández-Rojas, “AI-Driven Monitoring for Fish Welfare in Aquaponics: A Predictive Approach,” Sensors, vol. 25, no. 19, Oct. 2025, doi: 10.3390/s25196107.
N. Tengtrairat, W. L. Woo, P. Parathai, D. Rinchumphu, and C. Chaichana, “Non-Intrusive Fish Weight Estimation in Turbid Water Using Deep Learning and Regression Models,” Sensors, vol. 22, no. 14, Jul. 2022, doi: 10.3390/s22145161.
L. P. Sari, A. Hamid, and H. Khaulasari, “Forecasting Zakat Potential in BAZNAZ East Java Using the ARIMAX Method with Calendar Variation Effects,” Euler : Jurnal Ilmiah Matematika, Sains dan Teknologi, vol. 13, no. 2, pp. 181–187, Jul. 2025, doi: 10.37905/euler.v13i2.31456.
P. A. Mendez-Santos, N. A. Chacón-Reino, L. F. Guerrero-Vásquez, J. O. Ordoñez-Ordoñez, and P. A. Chasi-Pesantez, “Estimation and Forecasting of the Average Unit Cost of Energy Supply in a Distribution System Using Multiple Linear Regression and ARIMAX Modeling in Ecuador,” Energies (Basel)., vol. 18, no. 14, Jul. 2025, doi: 10.3390/en18143659.
B. Shi, X. Jin, Y. Hu, J. Jiang, and Y. Sun, “Precision prediction of aquaculture water quality: a spatiotemporal model integrating optimized-LSTM and radial basis function neural networks,” PeerJ Comput. Sci., vol. 12, p. e3515, Jan. 2026, doi: 10.7717/peerj-cs.3515.
D. Chicco, M. J. Warrens, and G. Jurman, “The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation,” PeerJ Comput. Sci., vol. 7, pp. 1–24, 2021, doi: 10.7717/PEERJ-CS.623.
