Advanced Monte Carlo Method for atmospheric dispersion model in risk assessment
Résumé
In this paper, an advanced interval based Monte Carlo method is proposed, in order to take intoaccount uncertain inputs of an atmospheric dispersion model and to propagate these uncertainties for estimatingwith accuracy the geographical region in which the concentration of the considered toxic gas is upper than thethreshold of irreversible effects. The principle of the proposed method is to generate random interval supportsof model inputs instead of random values in order to increase accuracy and reduce the sampling size. Threesimulation approaches, Monte Carlo simulation, interval analysis method and advanced interval based MonteCarlo method, are applied and compared for computing the dangerous, safety and uncertainty areas. The proposedapproach leads to reduce the sampling size versus Monte Carlo simulation while retaining equivalent orbetter precision.