Design and validation of a reliable, low-cost IoT decision support system for proactive indoor air quality management

Indoor Air Quality has emerged as a critical determinant of health, productivity, and building performance, particularly as people spend over 90% of their time indoors. Continuous Indoor Air Quality monitoring is essential for preventing Sick Building Syndrome and optimizing HVAC energy consumption, yet the high cost and demanding calibration of reference-grade instruments hinder their large-scale deployment. This study presents MiQar, a low-cost, scalable IoT-based Decision Support System for real-time Indoor Air Quality-driven HVAC optimization in smart buildings. MiQar integrates heterogeneous Commercial Off-The-Shelf sensors for CO 2 , Total Volatile Organic Compounds, temperature, and humidity within a three-tier IoT architecture. The core innovation of MiQar is a multi-sensor redundancy mechanism that ensures cross-validation between sensors with different measurement principles, thus significantly enhancing data reliability without requiring complex, costly pre-deployment calibration. Laboratory benchmarking against an Aeroqual Series 500 reference confirmed low maximum deviations within ±38 ppm (CO 2,eq ) and ±32 ppb (Total Volatile Organic Compounds). Furthermore, field validation in a controlled pilot environment (university facility) demonstrated strong correlation R 2 = 0.86 for CO 2,eq ; R 2 = 0.73 for Total Volatile Organic Compounds) and reliable trend tracking under dynamic occupancy conditions. The rule-based Decision Support System effectively converted this validated data into proactive ventilation and maintenance actions, demonstrating MiQar's potential to significantly enhance Indoor Air Quality management rigor and promote energy efficiency in smart building environments.