The prevention of fatal risks is a key objective for the global resources industry, especially considering poor performance with regard to fatality rate reduction and the continued occurrence of mine disasters globally. However, the industry has a credible history, over several decades, of the development and deployment management systems to control potential Serious Injury, Fatal and Catastrophic (P-SIFC) risks. While these systems have had a significant impact on the thinking and approaches to the prevention of fatal risks, fatal accident rates have not shown the desired reduction. Recent fatal and serious injury events in Australia have led to the establishment of several enquiries to further identify improvements that could be considered. Globally, fatal accident rates remain stubbornly high or plateaued in many jurisdictions, and several catastrophic events have occurred, such as tailing dams disasters in Brazil and South Africa, and the recent coalmine disaster in Turkey. Apart from the apparent lack of progress, it is evident that the deployment of critical risk systems has introduced more bureaucracy, slower response systems and additional burdens on already overloaded middle and supervisory levels of management. Organisations are rarely able to effectively tap into the risk source of risk knowledge of the frontline worker and supervisor, for various reasons discussed in this paper. To establish a valid and valuable channel of critical risk information flow from the frontline employee, Debmarine Namibia instigated the development and deployment of a system called SafeSENTRY, where the focus is on the ‘dynamic identification of dynamic risks’. This paper outlines the system, the principles of the design, the range of risk metrics that can be derived from the observations of the frontline operators, and demonstrates how a range of posited shortcomings in critical risk management is addressed how features of the High Reliability Organization (HRO) have been developed.