Air and Water Quality Index Development for Environmental Assessment in Industrial Context

Abstract

Considering the importance of air and water to human existence, air and water pollution are critical issues that require collective effort for prevention and control. Different types of anthropogenic activities have resulted in environmental ruin. One of the tools that can be used for such a awareness campaignis Air Quality Index(AQI). The AQI was based on the concentrations of different types of pollutants: We are also familiar with the Water Quality Index (WQI), which in simply tells what the quality of drinking water is from a drinking water supply. There is a need for constant and real-time monitoring of air quality and water quality for the development of AQI and WQI, which in turn will enable clear communication of how unclean or unhealthy the air and water in the study area is. Similar systems have been developed utilizing IoT technologies, as discerned in [1], [2].

Introduction

Air and water pollution pose urgent global challenges, worsened by industrialization and urbanization, impacting health, ecosystems, and climate. Environmental monitoring uses indices like Air Quality Index (AQI) and Water Quality Index (WQI) to quantify pollution levels. This paper presents an IoT-based system combining sensors and machine learning to predict AQI and WQI in industrial environments, enabling real-time monitoring, forecasting, and actionable insights.

AQI Prediction System:
Uses sensors (e.g., MQ-135, MQ-7, PM2.5) connected to TTGO LoRa32 microcontrollers to collect air pollutant data. Machine learning models—Decision Tree and Random Forest regressors—analyze data to predict air quality. Data is transmitted wirelessly to cloud storage and visualized on dashboards with alerts for poor air quality.

WQI Prediction System:
Monitors water parameters like pH, turbidity, TDS, and temperature with appropriate sensors. Uses machine learning models (OLS regression, SVR, logistic regression, decision trees, neural networks) to predict water quality, providing classifications such as Excellent or Poor. Data is visualized on user-friendly dashboards with alert notifications.

Implementation:
Includes sensor calibration, edge processing, cloud integration (MongoDB on Google Cloud), and development of interactive web dashboards with heatmaps and temporal trends. Alerts are sent via SMS and email when pollution thresholds are exceeded. System performance was validated through testing and pilot deployments.

Results:

• AQI prediction achieved ~90% accuracy, identifying pollution hotspots.

• WQI prediction showed over 92% reliability, with machine learning models like OLS and SVR achieving 100% accuracy in tests.

• Real-time dashboards and alerts enhanced environmental awareness and response.

• The IoT-cloud approach offered a cost-effective solution supporting better environmental management and policy-making.

Conclusion

TheAQIandWQIpredictionsystemsofferrobusttools for monitoring environmental quality in industrial areas. By integrating IoT devices, machine learning models, and user- centric dashboards, these systems provide real-time insights and predictive analytics. Future enhancements include: 1) Incorporatingadditionalsensorsforbroaderpollutant coverage. 2) LeveragingadvancedAItechniqueslikedeeplearningfor improved accuracy. 3) Deployingrenewableenergysolutionsforsustainable sensor operation. 4) Expandingsystemdeploymenttoruralandurbanareas for comprehensive coverage.

References

[1] Verma,R.,Ahuja,L.,&Khatri,S.(2018).WaterQuality Index Using IoT. IEEE Xplore. [2] Esfahani, S., Rollins, P., & Specht, J. (2020). SmartCity Battery Operated IoT Based Indoor Air Quality Monitoring System. IEEE Sensors. [3] Ramadani, M. E., & Zain, A. T. (2021). Monitoring System on Carp Farming Ponds as IoT-Based Water Quality Control. ICRACOS. [4] Jha,R.K.(2020).AirQualitySensingandReporting System Using IoT. ICIRCA.

Copyright

Copyright © 2025 Devang Vartak, Tanmay Pal, Pranjal Srivastava, Prof. Govind Haldankar. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.