Smart Tech Meets Clean Air: New AI-Based Model Enhances Indoor Air Quality Monitoring
Shah Alam, Malaysia — In a landmark development for environmental health and smart building technologies, researchers from Management & Science University and University of Baghdad have designed a cutting-edge system to assess Indoor Air Quality (IAQ) using a Fuzzy Inference System (FIS)—a form of artificial intelligence that mimics human decision-making.
Published in ICT Express: Conference Proceedings (Elsevier), the research—led by Brainvendra Widi Dionova —aims to help governments, businesses, and building owners ensure healthier indoor environments amid rising concerns about pollution, climate control, and public health risks.
Why Indoor Air Quality Matters
With people spending over 90% of their time indoors, air quality in enclosed environments directly impacts human health and productivity. Contaminants like carbon dioxide (CO₂), carbon monoxide (CO), volatile organic compounds (VOCs), particulate matter (PM2.5), and temperature/humidity imbalances are linked to respiratory issues, allergies, and even cognitive impairment.
Traditional air quality measurement systems often rely on hard thresholds and limited sensor data—leading to delayed or inaccurate risk detection.
Introducing Fuzzy Logic for Smart IAQ Assessment
The study proposes the use of a Fuzzy Inference System (FIS)—a decision-making framework that mimics how humans make judgments based on approximate data, rather than relying solely on binary logic.
The researchers designed an IAQ model using FIS with key input parameters:
- Air Temperature
- Relative Humidity
- Carbon Dioxide Levels
- Carbon Monoxide Levels
- Particulate Matter (PM2.5) Concentration
Each of these inputs is processed using fuzzy rules—developed through expert knowledge—to evaluate indoor air conditions as Good, Moderate, or Unhealthy.
System Design and Implementation
The model comprises three major components:
- Fuzzification Module
Translates raw sensor data into fuzzy variables (e.g., “Low,” “Medium,” “High”).
- Inference Engine
Applies a set of fuzzy IF-THEN rules to analyze inputs and determine air quality levels. For example:
- IF CO₂ is High AND Temperature is High → THEN IAQ is Unhealthy.
- Defuzzification Module
Converts fuzzy output into a clear numerical IAQ score for display and analysis.
The system can be integrated with IoT-based environmental sensors, enabling real-time monitoring, automated HVAC system adjustments, and proactive alerts via smartphone apps.
Simulation & Results
Using MATLAB, the team ran extensive simulations across multiple environmental scenarios. The FIS model successfully provided dynamic, real-time assessments, outperforming traditional threshold-based systems by adapting to multi-factorial input combinations.
The results confirm that fuzzy logic enhances IAQ prediction accuracy, especially in fluctuating indoor environments such as:
- Offices
- Hospitals
- Educational institutions
- Smart homes and factories
Broader Implications for Smart Cities
The paper emphasizes the model’s applicability to smart buildings and urban infrastructure, where integrated air quality monitoring is critical to sustainable development and public safety.
“Our FIS model provides a reliable and intelligent framework for real-time environmental health monitoring. It’s a step toward smarter, healthier living spaces,” said co-lead author Dr. Mohammed N. Abdulrazaq, currently working as Associate Professor, Gulf University, Bahrain.
Future Directions
The authors suggest expanding the system to include:
- VOCs and bioaerosol detection,
- Machine learning enhancements,
- Cloud-based data sharing for regional air quality mapping.
This research not only bridges AI and environmental science but also opens new pathways for deploying affordable, scalable, and autonomous IAQ solutions worldwide.