Development of a framework for dynamic risk assessment of environmental impacts in chemicals warehouse using CFD-BN

Abstract

Chemical warehouses are one of the high-risk areas in the process industries due to the high diversity and quantity of stored chemicals. Risk assessment is a useful tool for developing appropriate strategies to prevent and control the risks. In this study, computational fluid dynamics (CFD) and Bayesian network (BN) approaches were proposed for dynamic risk assessment. Initially, bow tie (BT) method was used for identifying basic events and modeling the consequences. In order to determine the consequences intensity (heat flux and CO and CO2 concentration), fire dynamics simulator (FDS) and solid flame model were used. A total of 21 causes or failures were identified in the chemical spills, 13 cases of which were related to basic events. Out of the identified causes, the forklift and drum strike basic event had the most contribution of the chemical spills in the warehouse, and the probability of the major spill event scenario 1.25495E−11 was estimated by Bayesian networks. The estimated risk by CFD in combination with BN, after updating, is unacceptable compared to the UK risk criterion. Bayesian networks and CFD approach for dynamic assessment of environmental impact risk in chemical warehouses provide the capability to quantitatively and dynamically assess the consequences of chemical spills by modeling its cause and effect in the warehouse. Based on the results of this method, effective preventive measures can be taken to control the consequences of chemical spills in the warehouse.