Most accidents in chemical plants are caused by human errors. A common source of human error is lack of information. By integrating additional high quality information into the human machine interface, decision-making could be better supported and improved. Our approach strives to preserve the information gathered in the hazard and risk assessment for the operation phase. In this contribution,we present the developed formal models and tools to support transformation of the results of a conventional HAZOP study. That makes it possible to link the analysis to subsequent refinements and implementation steps of the Cause and Effect matrix. This transformation approach provides unique features.While not constraining human creativity, it provides a consistent way to integrate new and modified safety functions from repeated and updated hazard and risk assessments.