<p>Mine ventilation monitoring systems are essential for the early detection of malfunctions in ventilation networks, the timely identification of emerging hazardous conditions, and for providing valuable data to improve the energy efficiency of mine ventilation. One of the key challenges in this context is reconstructing airflow distribution throughout the ventilation network using data from a limited number of measurement points. This requires solving the so-called inverse ventilation problem (IVP) — adjusting the airflow resistances of mine workings to minimize the discrepancy between calculated and measured airflows at sensor locations. To date, IVP is typically solved manually. However, automated solutions are a crucial component for developing digital twins of mine ventilation systems. Achieving this requires a solid theoretical foundation — an efficient method for solving IVP for networks with arbitrary topology and sensor placement. This study addresses this challenge by proposing a novel IVP solution method based on Tikhonov regularization. The proposed algorithm has been numerically implemented and its effectiveness demonstrated on a model of a real mine ventilation network. For the first time, it is demonstrated that when airflow sensors are sparsely distributed, there exists an optimal non-zero regularization parameter. In contrast, for dense sensor networks, the regularization parameter can be effectively set to zero.</p>