Predictive Modeling of Carbon Monoxide with MOS Sensors and Machine Learning: A Potential Tool for Process Safety Improvement
##plugins.themes.bootstrap3.article.main##
Keywords
CO Detection, MOS Sensor, Feature Selection, Machine Learning, Process Safety Monitoring
Abstract
Carbon monoxide (CO) is a toxic, odorless gas commonly present in industrial processes and poses serious risks to occupational safety and health. This study proposes an optimized machine-learning-based approach to predict CO concentration using metal-oxide semiconductor (MOS) sensor arrays. The model was trained and evaluated on a public dataset comprising 650 time-series measurements from 14 thermally modulated MOS sensors, tested across CO concentrations ranging from 0 to 8.9 ppm under dynamic relative humidity (15%–75%). To optimize computational efficiency and mitigate multicollinearity, a multi-method feature selection strategy that combines Random Forest importance, Recursive Feature Elimination (RFE), and Mutual Information (MI) was implemented, successfully isolating sensors R10, R11, and R13 as the most robust predictors. A Random Forest Regression model, optimized via grid search and validated through five-fold cross-validation, was subsequently developed. The proposed framework demonstrated high predictive accuracy, achieving an R² of 0.884, Root Mean Square Error (RMSE) of 2.189 ppm, Mean Absolute Error (MAE) of 1.215 ppm, and Symmetric Mean Absolute Percentage Error (SMAPE) of 34.27%. These results highlight the potential of combining low-cost, feature-optimized MOS sensor arrays with ensemble machine learning for accurate, real-time gas monitoring. The framework provides a computationally efficient decision-support tool for the early detection of hazardous CO levels, contributing to safer process environments.
References
L. Xue, J. Dang, X. Li, F. Liu, Z. Qin, and Q. Wang, “Trace detection of carbon monoxide by AACVD of Ni-doped ZnO nanosheets,” Sensors and Actuators B: Chemical, vol. 442, p. 138126, Nov. 2025, doi: 10.1016/j.snb.2025.138126.
Z. Zhang et al., “Electronic nose based on metal oxide semiconductor sensors for medical diagnosis,” Progress in Natural Science: Materials International, vol. 34, no. 1, pp. 74–88, Feb. 2024, doi: 10.1016/j.pnsc.2024.01.018.
F. Raspagliesi, G. Bogani, S. Benedetti, S. Grassi, S. Ferla, and S. Buratti, “Detection of Ovarian Cancer through Exhaled Breath by Electronic Nose: A Prospective Study,” Cancers, vol. 12, no. 9, p. 2408, Aug. 2020, doi: 10.3390/cancers12092408.
K. Liu and C. Zhang, “Volatile organic compounds gas sensor based on quartz crystal microbalance for fruit freshness detection: A review,” Food Chemistry, vol. 334, p. 127615, Jan. 2021, doi: 10.1016/j.foodchem.2020.127615.
N. M. Zetola et al., “Diagnosis of pulmonary tuberculosis and assessment of treatment response through analyses of volatile compound patterns in exhaled breath samples,” Journal of Infection, vol. 74, no. 4, pp. 367–376, Apr. 2017, doi: 10.1016/j.jinf.2016.12.006.
Pradyumn, P. B. Barman, A. Sil, and S. K. Hazra, “Recent advancement in selective gas sensors and role of machine learning,” Journal of Alloys and Compounds, vol. 1030, p. 180757, May 2025, doi: 10.1016/j.jallcom.2025.180757.
J. Burgués, J. M. Jiménez-Soto, and S. Marco, “Estimation of the limit of detection in semiconductor gas sensors through linearized calibration models,” Analytica Chimica Acta, vol. 1013, pp. 13–25, Jul. 2018, doi: 10.1016/j.aca.2018.01.062.
J. Burgués and S. Marco, “Multivariate estimation of the limit of detection by orthogonal partial least squares in temperature-modulated MOX sensors,” Analytica Chimica Acta, vol. 1019, pp. 49–64, Aug. 2018, doi: 10.1016/j.aca.2018.03.005.
L. Fernandez, J. Yan, J. Fonollosa, J. Burgués, A. Gutierrez, and S. Marco, “A Practical Method to Estimate the Resolving Power of a Chemical Sensor Array: Application to Feature Selection,” Front. Chem., vol. 6, p. 209, Jun. 2018, doi: 10.3389/fchem.2018.00209.
M. Al-Hashem, S. Akbar, and P. Morris, “Role of Oxygen Vacancies in Nanostructured Metal-Oxide Gas Sensors: A Review,” Sensors and Actuators B: Chemical, vol. 301, p. 126845, Dec. 2019, doi: 10.1016/j.snb.2019.126845.
S. T. Araya et al., “Performance assessment of machine learning techniques in electronic nose systems for power transformer fault detection,” Energy and AI, vol. 20, p. 100497, May 2025, doi: 10.1016/j.egyai.2025.100497.
J. Xie, M. Sage, and Y. F. Zhao, “Feature selection and feature learning in machine learning applications for gas turbines: A review,” Engineering Applications of Artificial Intelligence, vol. 117, p. 105591, Jan. 2023, doi: 10.1016/j.engappai.2022.105591.
A. S. AlSalehy and M. Bailey, “Improving Time Series Data Quality: Identifying Outliers and Handling Missing Values in a Multilocation Gas and Weather Dataset,” Smart Cities, vol. 8, no. 3, Art. no. 3, Jun. 2025, doi: 10.3390/smartcities8030082.
M. G. L. Brown, M. G. Peterson, I. K. Tezaur, K. J. .Peterson, and D. L. Bull, “Random forest regression feature importance for climate impact pathway detection,” Journal of Computational and Applied Mathematics, vol. 464, p. 116479, Aug. 2025, doi: 10.1016/j.cam.2024.116479.
S. Sin, E. Kang, J. Kim, S. Oh, and J. Baek, “Random forest-based small training dataset complementation and features selection for capacity estimation of lithium-ion batteries in electric-powered application,” Applied Soft Computing, vol. 181, p. 113526, Sep. 2025, doi: 10.1016/j.asoc.2025.113526.
M. Heidari, M. H. Moattar, and H. Ghaffari, “Forward propagation dropout in deep neural networks using Jensen–Shannon and random forest feature importance ranking,” Neural Networks, vol. 165, pp. 238–247, Aug. 2023, doi: 10.1016/j.neunet.2023.05.044.
J. Xiong, T. Ma, Z. Lian, and R. de Dear, “Perceptual and physiological responses of elderly subjects to moderate temperatures,” Building and Environment, vol. 156, pp. 117–122, Jun. 2019, doi: 10.1016/j.buildenv.2019.04.012.
O. Kornyo et al., “Botnet attacks classification in AMI networks with recursive feature elimination (RFE) and machine learning algorithms,” Computers & Security, vol. 135, p. 103456, Dec. 2023, doi: 10.1016/j.cose.2023.103456.
F. Tian, S. Chen, X. Ji, J. Xu, M. Yang, and R. Xiong, “Robust lithium-ion battery state of health estimation based on recursive feature elimination-deep Bidirectional long short-term memory model using partial charging data,” International Journal of Electrochemical Science, vol. 20, no. 1, p. 100891, Jan. 2025, doi: 10.1016/j.ijoes.2024.100891.
N. S. Santhosh, K. M, K. K, and A. S, “AGXLG: Enhancing Hypothyroidism Prediction with Recursive Feature Elimination and Boosting Techniques,” Procedia Computer Science, vol. 258, pp. 3457–3467, 2025, doi: 10.1016/j.procs.2025.04.602.
R. Ahmed et al., “A novel integrated logistic regression model enhanced with recursive feature elimination and explainable artificial intelligence for dementia prediction,” Healthcare Analytics, vol. 6, p. 100362, Dec. 2024, doi: 10.1016/j.health.2024.100362.
L. Sun, J. Guo, X. Wu, and J. Xu, “Fuzzy C-means clustering-based multi-label feature selection via weighted neighborhood mutual information,” Information Sciences, vol. 718, p. 122389, Nov. 2025, doi: 10.1016/j.ins.2025.122389.
J. Chen, F. Fan, C. Wei, K. Polat, and F. Alenezi, “Decoding driving states based on normalized mutual information features and hyperparameter self-optimized Gaussian kernel-based radial basis function extreme learning machine,” Chaos, Solitons & Fractals, vol. 199, p. 116751, Oct. 2025, doi: 10.1016/j.chaos.2025.116751.
N. N. Viet, P. H. Phuoc, L. V. Thong, N. V. Chien, and N. Van Hieu, “A comparative study of machine learning models for identifying noxious gases through thermal fingerprint measurements and MOS sensors,” Sensors and Actuators A: Physical, vol. 375, p. 115510, Sep. 2024, doi: 10.1016/j.sna.2024.115510.
##plugins.themes.bootstrap3.article.sidebar##
##plugins.themes.bootstrap3.article.details##
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
The authors declare that:
1. This paper has not been published in the same form elsewhere.
2. It will not be submitted anywhere else for publication prior to acceptance/rejection by this Journal.
3. A copyright permission is obtained for materials published elsewhere and which require this permission for reproduction.