A novel, fuzzy-based air quality index (FAQI) for air quality assessment

The ever increasing level of air pollution in most areas of the world has led to development of a variety of air quality indices for estimation of health effects of air pollution, though the indices have their own limitations such as high levels of subjectivity. Present study, therefore, aimed at developing a novel, fuzzy-based air quality index (FAQI1) to handle such limitations. The index developed by present study is based on fuzzy logic that is considered as one of the most common computational methods of artificial intelligence. In addition to criteria air pollutants (i.e. CO, SO2, PM10, O3, NO2), benzene, toluene, ethylbenzene, xylene, and 1,3-butadiene were also taken into account in the index proposed, because of their considerable health effects. Different weighting factors were then assigned to each pollutant according to its priority. Trapezoidal membership functions were employed for classifications and the final index consisted of 72 inference rules. To assess the performance of the index, a case study was carried out employing air quality data at five different sampling stations in Tehran, Iran, from January 2008 to December 2009, results of which were then compared to the results obtained from USEPA air quality index (AQI). According to the results from present study, fuzzy-based air quality index is a comprehensive tool for classification of air quality and tends to produce accurate results. Therefore, it can be considered useful, reliable, and suitable for consideration by local authorities in air quality assessment and management schemes.

Research highlights
► FAQI minimizes subjectivity of conventional indices like USEPA AQI.
► The index can consider the interactions between parameters included.
► Both criteria air pollutants and hazardous pollutants like BTEX are included.
► More accurate results can be obtained using this index, compared to USEPA AQI.