This paper reports on research undertaken to determine the effectiveness of a 3D simulation environment used to train mining personnel in emergency evacuation procedures, designated the Fires in Underground Mines Evacuation Simulator (FUMES). Owing to the operational constraints of the mining facility, methods for measuring learning transfer were employed which did not require real world performance evaluation. Transfer measures that examined simulator performance relative to real world experience, fidelity assessment, and appraisal of the training value of the platform were utilised. Triangulation of results across all three measures indicated the presence of learning transfer, suggesting the viability of indirect measures in instances where real world performance testing is not possible. Furthermore, these indirect measures of learning transfer also provided some insight as to the strengths and weaknesses of the simulation design, which could be used to inform the development of future versions of the product.