How the Human Factors Approach Improves Process Safety - Applying Fram (Functional Resonance Analysis Method) to Analyze Risks and Accidents

Since the first process plants of the First Industrial Revolution, loss of containment has been the precursor to the chain of events that leads to an accident. Therefore, the ability to identify, understand and correct containment losses, regardless of their size, is a fundamental necessity for process safety, since its inception. In particular for the Oil & Gas Industry, which at its beginning, in 1859, dealt with process variables of the order of 20 meters deep, 50 barrels per day and 1 kgf/cm2 of pressure, currently, in the Brazilian Pre-salt, the process variables are on the order of 7,000 meters deep, 2 million barrels per day and 600 kgf/cm2. The loss of process control, the loss of containment in this current scenario is critical, if not catastrophic. Thus, continuing to use essentially linear tools, methodologies, and concepts, such as FTA (Fault Tree Analysis) in a complex sociotechnical workplace with extremely critical process variables, the identification of failures and problems will be limited, restricted, if not inadequate. In the same way that process plants have evolved, it is also necessary, to the same extent, for the evolution of safety management systems, ensuring that projects, risk assessments and the investigation of process accidents are able to provide solutions appropriate to the complexity and criticality of process plants. In this sense, FRAM (Functional Resonance Analysis Method) is a methodology that allows non-linear connections in modeling the functioning of a process, identifying complex couplings and connections of current industrial process plants, such as the process plant of an FPSO (Floating Production Storage and Offloading). Taking this process plant as an example, this study presents the risk assessment of the oil treatment process onboard an FPSO, identifying the critical systemic elements and performance influencing factors that affect process safety. Likewise, for accident investigation, the analysis of the FPSO CSM accident was adopted, which in 2015, after the loss of containment of the primary process, an explosion occurred, causing the loss of the process plant and the fatality of nine workers. The results of these analyzes demonstrated that proximal causes, such as human errors and equipment failures, have a partial contribution to process safety. It is the organizational factors, such as culture, hierarchical levels, procedures, annual budgets, and mergers that have been identified as the major contributors to process safety, producing profound and permanent changes in the company’s organizational structures.