Occupational health and safety (OHS) in mining has changed drastically in terms of hazard and risk reduction due to the uptake of new technologies, namely digital and automation solutions. Technology-driven solutions not only allow for more informed decisions based on data and optimized processes, but also reduce personnel exposure to hazards. This becomes especially attractive for underground hard rock mines, needing to adjust to current trends such as deep mining. Notwithstanding, the implementation of technology has brought about issues and challenges stemming from interaction inadequacies between humans and machines, as it happened in the transition to mechanized mining or as it is the case nowadays amidst the digital transformation. This lack of human-systems integration poses a project risk, both from the operational and health and safety (H&S) perspective. Therefore, a methodology able to standardize a process for identification, selection and evaluation of meaningful Safety Key Performance Indicators (KPIs) for underground mining opens an opportunity for small or mid-size mining companies with intentions of improving their corporate H&S goals. This research seeks to identify trends among selected technologies deployed in underground hard rock mines and develop, test and validate Safety KPIs supporting the decision-making and implementation process of new technologies and tools in the mine 4.0, while also being applicable to other technology solutions or scenarios. For conducting this research, a set of diverse methodologies and techniques were merged and followed. Firstly, an extensive literature review on the topic of technology and H&S was conducted focusing on technologies with capability to enhance H&S in operation environments. Secondly, for the development of a unique methodology, a research design was tailored to potential H&S aspects with a leading behavior or risk prevention potential, where the three main impact factors, human, technology and mine environment, were present; and efforts could be diversified across departments. In addition, a process for identification of metrics was created. To conclude, qualitative data was gleaned for testing the design through interviews with experts from the mining equipment and automation industry. The proposed methodology enabled the compilation of 48 Safety Performance Indicators, 13 of which were targeted as Safety KPIs. The methodology was validated in a comparative case study (traditional mechanized vs. modern mining) concerning the loss of control over a LHD by making use of a bow-tie analysis focusing on preventive controls. Furthermore, qualitative and semi-quantitative analysis with modified and adapted risks matrixes was conducted to measure the efficiency of the Safety KPIs in each scenario. The results obtained denote the importance of the human component among the selected Safety KPIs in addition to the integration of the system human-technology for both scenarios. All things considered, the proposed methodology proved effective in the case study and coherent with the results obtained from the overall research.