ABSTRACT We present a new cosmologically motivated mock redshift survey of the dusty star-forming galaxy population. Our mock survey is based on the Bolshoi–Planck dark matter halo simulation and covers an area of 5.3 deg2. Using a semi-empirical approach, we generate a light-cone and populate the dark matter haloes with galaxies. Infrared properties are assigned to the galaxies based on theoretical and empirical relations from the literature. Additionally, background galaxies are gravitationally lensed by dark matter haloes along the line of sight assuming a point-mass model approximation. We characterize the mock survey by measuring the star formation rate density, integrated number counts, redshift distribution, and infrared luminosity function. When compared with single-dish and interferometric observations, the predictions from our mock survey closely follow the compiled results from the literature. We have also directed this study towards characterizing one of the extragalactic legacy surveys to be observed with the TolTEC camera at the Large Millimeter Telescope: the 0.8 sq. degree Ultra Deep Survey, with expected depths of 0.025, 0.018, and 0.012 mJy beam−1 at 1.1, 1.4, and 2.0 mm. Exploiting the clustering information in our mock survey, we investigate its impact on the effect of flux boosting by the fainter population of dusty galaxies, finding that clustering can increase the median boosting by 0.5 per cent at 1.1 mm, 0.8 per cent at 1.4 mm, and 2.0 per cent at 2.0 mm, and with higher dispersion.