Disasters are constant threats to humankind, and beyond losses in lives, they cause many implicit yet profound societal issues such as wealth disparity and digital divide . Among those recovery measures in the aftermath of disasters, restoring and improving communication services is of vital importance . Although existing works have proposed many architectural and protocol designs, none of them have taken human factors and social equality into consideration . Recent sociological studies have shown that people from marginalized groups (e.g., minority, low income, and poor education) are more vulnerable to communication outages . In this work, we take pioneering efforts in integrating human factors into an empirical optimization model to determine strategies for post-disaster communication restoration . We cast the design into a mix-integer non-linear programming problem, which is proven too complex to be solved . Through a suite of convex relaxations, we then develop heuristic algorithms to efficiently solve the transformed optimization problem . Based on a collected dataset, we further evaluate and demonstrate how our design will prioritize communication services for vulnerable people and promote social equality compared with an existing modeling benchmark.