A major challenge in present methods for risk control of major accidents in the petroleum sector is the inability to continuously monitor changes in risk level, thereby limiting their suitability to provide decision support during daily operations and planning. Research questions of particular importance include: How can operators continuously measure the status of critical safety barriers, how does the barrier status influence the risk level on the installation, and how should changes in risk level be visualised in order to facilitate improved recognition and evaluation of operational risks? Case studies have been performed to develop a method that better utilizes on-line data to assess the risk level, and use this information as decision support in the daily risk management. We propose a holistic risk-informed approach in which the starting point is the quantitative risk analysis (QRA) for a specific installation complemented with a barrier analysis. To visualise the risk, a risk barometer has been developed which presents the instantaneous risk level as well as trends over the last period. The risk barometer may support typical decisions such as the maximum number of work permits to be approved in a particular area and when to limit maintenance work in hydrocarbon areas