In this study, a multi-objective method for allocating the number and configuration of an air quality monitoring network based on non-dominated sorting genetic algorithm II has been presented. The multiple cell approach based on the solution of an Eulerian Model built on K-theory was used to predict the dispersion of emitted pollutants (SO2, CO, NO x ) from different emission sources. The multi-objective optimization method proposed in this study utilized two objectives: (1) maximum coverage area with respect to continuity of covered area and minimum overlap among coverage areas and (2) detection of violations over ambient standards. The concept of sphere of influence was used to determine the spatial area coverage of the monitoring station, and a weighing function was employed to measure the capability of a designed network to detect violations of air quality standards. The results show that three stations are suitable for the study region with coverage efficiency of 80 %. Analyzing the effect of cutoff correlation coefficient r c shows that, when the r c increases, although the coverage area decreases, the covered region will be well represented and overlap region will decrease.