Process Safety Progress, Vol.26, No.2, 129-139, 2007
A model for estimating the probability of missile impact: Missiles originating from bursting horizontal cylindrical vessels
Past explosion events in process facilities have centered attention on the fact that missiles generated as the result of vessel fragmentation pose significant risk to personnel and process equipment and can trigger knock-on or domino effects in industrial accidents. To promote the design of inherently safer facilities-and to enable more effective mitigation and control measures with respect to missile risks-it is necessary to perform missile risk analysis studies at the early design phase. To aid in such an analysis and to predict domino scenarios, it is essential to have models that can quantify (1) the probability of a missile impact on a target and (2) the consequences of the probable impact. In the present work, we propose a model to quantify the probability of missiles that originate from bursting horizontal cylinders, having an impact on spherical target vessels located in a process area. Although previous investigations on the quantification of missile impact probability were based on a single-scenario approach, the current model is based on two credible scenarios. The model is built on the concept of a vulnerable area (VA), defined as the probable impact zone sketched around the target object. The concept of an effective range interval (ERI) was adopted and was extended by introducing an effective trajectory interval (ETI) and effective orientation interval (EOI). The uncertainties in the model parameters were addressed by means of Monte Carlo sampling approach. A case study is used to evaluate the performance of the proposed model. The case study uses a grid-based approach (GBA) to provide interactive contour plotting of the missile impact probabilities over a defined process area. (c) 2006 American Institute of Chemical Engineers Process.
Keywords:missile risk;impact probability;impact consequences;explosion modeling;probabilistic analysis;Monte Carlo analysis;domino effect analysis