Related Experiment Video
Updated: Sep 13, 2025
05:16
Flying Insect Detection and Classification with Inexpensive Sensors
Published on: October 15, 2014
25.3K
A Low-Cost IoT Sensor and Preliminary Machine-Learning Feasibility Study for Monitoring In-Cabin Air Quality: A Pilot
Nurdaulet Tasmurzayev1,2,3, Bibars Amangeldy1,2,3, Gaukhar Smagulova1,4,5
1Institute of Combustion Problems, Almaty 050012, Kazakhstan.
Sensors (Basel, Switzerland)
|July 30, 2025
Summary
Passenger density significantly impacts indoor air quality on Almaty
Area of Science:
- Environmental Science
- Public Health
- Data Science
Background:
- Urban public transport air quality is crucial for passenger health.
- Poor ventilation in Almaty's public transport leads to high CO2 and PM2.5 levels.
- Elevated pollutant concentrations pose risks for respiratory and cardiovascular diseases.
Purpose of the Study:
- Investigate air quality (CO2, PM2.5, PM10, temperature, humidity) in Almaty's public transport.
- Analyze pollutant fluctuations based on passenger density and time of day.
- Evaluate machine learning models for predicting in-cabin air quality.
Main Methods:
- Continuous air quality monitoring using Tynys mobile IoT devices.
- Bench-calibration of IoT devices against reference sensors.
- Training and validation of machine learning models (logistic regression, decision tree, XGBoost, random forest) on environmental and occupancy data.
Main Results:
- Passenger occupancy is the primary driver of in-cabin air pollution.
- XGBoost model achieved the highest predictive accuracy (91.25%).
- High-temporal-resolution data revealed pollution spikes during peak ridership.
Conclusions:
- Improving ventilation in Almaty's public transport is essential for public health.
- Machine learning models effectively capture complex environmental variable relationships.
- Low-cost IoT and data analytics offer practical solutions for safeguarding urban mobility health.
