This paper illustrates the challenges of safety management in Academia, which are difficult to overcome using conventional methods of risk management. The positive aspects and disadvantages of the risk management tools specifically designed for the laboratory collection are addressed. We propose a different approach to hazard detection that can ease the burden for Academia safety experts of tedious time-consuming risk assessment. Based on the observations of Academia’s safety management, we are attempting to build a semi-quantitative risk analysis model that will conform to the concept of safety-II. This paper suggests an approach that allows to integrate this model into human oriented performance driven decision making.