Diesel particulate matter (DPM) is carcinogenic to humans. Underground miners have a high risk of over-exposure to high concentrations of DPM. To control DPM effectively, it is essential to understand the DPM dispersion characteristics. In this study, the DPM distributions of three key and representative mining activities, shotcreting, charging and loading activity, in an underground development face were studied. A computational model for the mining activities was developed using 3D imagery, onsite data and OpenFOAM. Tracer gas experiments were first conducted in the underground mine for the validation of CFD simulation. The simulations were carried out at a steady-state using the standard k-ε turbulence model, and the transport and dispersion of DPM were modelled using a segregated species transport model. DPM distribution characteristics for each mining activity were analysed, and the regions with high concentration (>0.1 mg/m3) were identified, and the reasons for the high concentrations were also discussed. At last, the efficiency of the current auxiliary ventilation system on DPM dilution was evaluated based on the simulation results. The results show that a broader region with high DPM concentration was identified in the downstream of the loader during the loading activity, and this issue could be solved by simply increasing the ventilation rate. The findings in this paper could be used for optimizing the auxiliary ventilation design for future mining activities in this development face.