Estimation of Maximum Loss - A comparison study

Övrigt konferensbidrag, 2010

Petroleum refineries are at risk due to the flammability of both the raw material and the products. In addition to the threat to the working force there is also an apparent economical risk associated with the processing of flammable compounds. This economical risk is not only due to the direct impact of a fire or explosion but also includes the cost of business interruption in case of a shutdown due to an accident. As with any situation that involves economical risks refineries may share their risk with insurers. The premium is then based on the size of the financial risk. Thus a decision has to be made by the operator how to share the risk of economical losses with insurance companies at a fair price. However, the decision process is not easy and it generally includes modelling of various scenarios to determine to which extent the process area can be damaged if, for example, a pipe rupture occurs. On the extreme end of modelling the so called Estimated Maximum Loss (EML) scenarios are found. These scenarios try to predict the maximum loss a particular installation can sustain due to an accident. Unfortunately a standard model for estimation of maximum loss does not exist. It means that brokers reach different results on the same scenario due to applying different models and different assumptions. Thus an operator may face uncertainty during the decision process. Hence improvements should be made not only to the models used but also to the concept of EML itself. Therefore, a study has been conducted on a case "Preem Refinery" where several scenarios previously had been modelled by two different brokers using two different softwares, ExTool and SLAM. The aim of this paper is to review the concept of EML and to analyse the used models to see which parameters that influenced the results. The results of the study show that: Overpressure damage threshold values, cloud weights, and releasable inventory are the main sources of deviation in the modelled scenarios; Clear cut-off values for the probabilities of an accident should be used to avoid the "not plausible" argument sometimes heard. As for improving the models themselves, no clear reason for working with threshold values when it comes to overpressure damage can be found. A continuous curve seems more fitting in the age of computers. The possibility to shift such a curve to account for the difference in overpressure sustainability between different types of process equipment could also be explored; and, there are many aspects to investigate further in order to make potential loss predictions more reliable, and this should be well worthwhile since a lot of money is at stake when plant owners and insurers decide on insurance limits and premiums.