A Sensitivity Analysis on the Air Quality Index Based on the Pollutants Concentrations in Tehran

As air pollution is one of the most serious global problems nowadays, this project tries to be a part of the studies in this field. Tehran as the capital city of Iran is one of the most polluted cities. This paper includes a sensitivity analysis on the calculation method of the Air Quality Index (AQI) for the whole city. The necessity to be able to analyze the air quality of the city needs a unique AQI for the whole city. Thus, three statistical indicators mean, median and maximum are compared in this paper to find out which of these indicators reports the number of days in each level of the AQI category more precisely in Tehran in the year 1395 (March/20/2016 - March/20/2017). Maximum is expected to be the best indicator but the results show that it is better to consider each pollutant separately for choosing the best indicator. However, maximum reports the air quality worse than what it really is, which leads to more caution. Simulation of the concentration of the pollutants was done by a hybrid model. The observed data of the concentration of the criteria pollutants were denoised by the wavelet transformation, then a neuro-fuzzy system using fuzzy clustering was used for modeling. The value of R2 for the models is almost 0.9 that is a sign of their accuracy.