Evaluation of ARIMA, ARIMA-LSTM, LSTM, and CNN-LSTM Models for Daily Air Quality Index Forecasting in Jakarta

Mushliha; Nisrinah

doi:10.36080/idealis.v9i2.3799

Authors

Mushliha Teknologi Informasi, Teknik & Informatika, Universitas Bina Sarana Informatika, DKI Jakarta
Nisrinah Kesehatan Masyarakat, Ilmu – ilmu Kesehatan, Universitas Esa Unggul, DKI Jakarta

DOI:

https://doi.org/10.36080/idealis.v9i2.3799

Keywords:

ARIMA, CNN-LSTM, Air Quality Index (AQI), Deep Learning, Air Quality Forecasting

Abstract

Increasing air pollution in urban areas, particularly in DKI Jakarta, has made a reliable air quality prediction system increasingly essential for environmental control and public health risk management. The Air Quality Index (AQI) exhibits complex and fluctuating patterns, requiring forecasting methods capable of capturing both linear and non-linear. This study aims to conduct a comparative analysis of statistical, deep learning, and hybrid models for AQI forecasting using daily AQI data from Jakarta during the 2023–2025 period. The dataset includes polutant parameters such as , , , , , and CO. The proposed models consist of Autoregressive Integrated Moving Average (ARIMA), Long Short-Term Memory (LSTM), Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM), and a hybrid ARIMA-LSTM model. The research methodology includes data preprocessing, normalization using Min-Max Scaling, sequence generation using the sliding window approach, model training, and evaluation using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE). The experimental results show that the CNN-LSTM model achieved the best forecasting performance with MAE, RMSE, and MAPE values of 4.54, 6.03, and 13.83%, respectively, followed closely by the LSTM model. Meanwhile, the ARIMA model produced the lowest performance, and the hybrid ARIMA-LSTM model did not outperform the standalone deep learning models. These findings indicate that deep learning approaches, particularly CNN-LSTM, are more effective in capturing the complex dynamics of urban air pollution data and have strong potential to support air quality forecasting and pollution control systems in Jakarta.

Downloads

Download data is not yet available.

References

[1] R. Wangintan, M. Yani, H. Hardjomidjojo, and T. June, “Evaluation of Air Pollution Control Management in Jakarta by Crossed-Impact Matrix Multiplications Applied to Classification (MICMAC),” IOP Conf. Ser. Earth Environ. Sci., vol. 1358, no. 1, 2024, doi: 10.1088/1755-1315/1358/1/012044.

[2] Farhatun Haya, Khaira Nisa, Rio Febrian Ladipasa, Ari Suriani, and Afriza Media, “Dampak Polusi Udara terhadap Kesehatan Manusia,” WISSEN : Jurnal Ilmu Sosial dan Humaniora, vol. 3, no. 2, pp. 180–190, 2025, doi: 10.62383/wissen.v3i2.753.

[3] B. -, S. Listyarini, and L. Warlina, “Pengaruh Kebijakan Pencemaran Udara Sektor Transportasi Terhadap Nilai Indeks Kualitas Udara (Iku) Di Dki Jakarta,” Jurnal Ilmiah Pendidikan Lingkungan dan Pembangunan, vol. 24, no. 01, pp. 1–13, 2023, doi: 10.21009/plpb.v24i01.30798.

[4] M. Mushliha, “Implementasi CNN-BiLSTM untuk Prediksi Harga Saham Bank Syariah di Indonesia,” Jambura Journal of Mathematics, vol. 6, no. 2, pp. 195–203, 2024, doi: 10.37905/jjom.v6i2.26509.

[5] Z. Shahbazi, Z. Shahbazi, and S. Nowaczyk, “Enhancing Air Quality Forecasting Using Machine Learning Techniques,” IEEE Access, vol. 12, no. December, pp. 197290–197299, 2024, doi: 10.1109/ACCESS.2024.3516883.

[6] S. Deepan and M. Saravanan, “Air quality index prediction using seasonal autoregressive integrated moving average transductive long short-term memory,” ETRI Journal, vol. 46, no. 5, pp. 915–927, 2024, doi: 10.4218/etrij.2023-0283.

[7] K. Sekaran, M. Priyadharshini, and R. Sahay, “AirTFT: A Novel Transformer-Based Approach for Air Quality Prediction,” International Symposium on Advanced Networks and Telecommunication Systems, ANTS, pp. 114–119, 2024, doi: 10.1109/ANTS63515.2024.10898497.

[8] S. Tilahun, K. Abera, A. Habtie, B. Tilahun, M. Gebrehiwot, and T. N. Mengestie, “Predicting Air Quality for Sustainable Urban Living,” 2025 International Conference on Information and Communication Technology for Development for Africa (ICT4DA), no. Ml, pp. 55–59, 2025, doi: 10.1109/ict4da67218.2025.11282769.

[9] J. Xiao, Q. Wang, J. Cui, and J. Yu, “Multi-feature PM2.5 Prediction with ARIMA-LSTM,” Proceedings - 2022 International Conference on Machine Learning, Cloud Computing and Intelligent Mining, MLCCIM 2022, no. Mlccim, pp. 229–233, 2022, doi: 10.1109/MLCCIM55934.2022.00046.

[10] K. Kashif and R. Ślepaczuk, “LSTM-ARIMA as a hybrid approach in algorithmic investment strategies,” Knowl. Based. Syst., vol. 320, no. March, 2025, doi: 10.1016/j.knosys.2025.113563.

[11] J. Zhang and S. Li, “Air quality index forecast in Beijing based on CNN-LSTM multi-model,” Chemosphere, vol. 308, no. P1, p. 136180, 2022, doi: 10.1016/j.chemosphere.2022.136180.

[12] C. Shilengwe, K. Banda, and I. Nyambe, “Machine learning downscaling of GRACE/GRACE-FO data to capture spatial-temporal drought effects on groundwater storage at a local scale under data-scarcity,” Environmental Systems Research, vol. 13, no. 1, pp. 1–19, 2024, doi: 10.1186/s40068-024-00368-1.

[13] D. L. John, S. Binnewies, and B. Stantic, “Identifying Optimal Window Size Configurations for Big Data Time Series Forecasting,” Proceedings - 2024 IEEE International Conference on Big Data, BigData 2024, pp. 5138–5146, 2024, doi: 10.1109/BigData62323.2024.10825222.

[14] P. A. Faradiela, A. Wibowo, and Y. R. Husniyawati, “Forecasting the Number of Vertical Referrals for Bpjs Participants At Health Service Center in Universitas Airlangga Using Arima Model,” Jurnal Biometrika dan Kependudukan, vol. 11, no. 1, pp. 62–71, 2022, doi: 10.20473/jbk.v11i1.2022.62-71.

[15] F. A. Mahbubi, T. I. Hermanto, and C. D. Lestari, “Peramalan Penjualan Saham Nikel Menggunakan Algoritma Long Short Term Memory (Lstm),” IDEALIS : InDonEsiA journaL Information System, vol. 8, no. 1, pp. 138–149, 2025, doi: 10.36080/idealis.v8i1.3254.

[16] F. Aksan, Y. Li, V. Suresh, and P. Janik, “CNN-LSTM vs. LSTM-CNN to Predict Power Flow Direction: A Case Study of the High-Voltage Subnet of Northeast Germany,” Sensors, vol. 23, no. 2, 2023, doi: 10.3390/s23020901.

[17] T. A. Wahyudi et al., “Pengembangan E - Marketplace Ikan dengan Chatbot NLP berbasis LSTM di BBI Cangkiran Mijen,” vol. 14, pp. 2299–2317, 2025, doi: https://doi.org/10.32520/stmsi.v14i5.5170.

[18] C. Li, X. Fang, Z. Yan, Y. Huang, and M. Liang, “Research on Gas Concentration Prediction Based on the ARIMA-LSTM Combination Model,” Processes, vol. 11, no. 1, 2023, doi: 10.3390/pr11010174.

[19] R. Dadi, C. Narsingoju, K. Sridhar, K. Sravani, V. Mounika, and Shabana, “Temporal Fusion and Hybrid ARIMA-LSTM Models for Robust Time Series Forecasting: A Case Study on Ethereum Price Dynamics,” Proceedings of 2025 2nd International Conference on Cognitive Robotics and Intelligent Systems, ICC - ROBINS 2025, pp. 622–627, 2025, doi: 10.1109/ICC-ROBINS64345.2025.11086123.

[20] B. Bhagawanta and C. B. Santoso, “Deteksi penyakit daun kapas dengan deep learning berbasis Convolutional Neural Network (CNN),” IDEALIS InDonEsiA journaL Information System, vol. 8, no. 2, pp. 220–227, Jul. 2025, doi: 10.36080/idealis.v8i2.3517.

[21] Y. Ding, S. Liu, Z. Zhao, and S. Liu, “A CNN-LSTM Model for Electric Load Forecasting,” ICNSC 2024 - 21st International Conference on Networking, Sensing and Control: Artificial Intelligence for the Next Industrial Revolution, pp. 1–6, 2024, doi: 10.1109/ICNSC62968.2024.10760118.