Uncertainty Modelling in Metamodels for Fire Risk Analysis

In risk-related research of fire safety engineering, metamodels are often applied to approximate the results complex and evacuation simulations. This approximation may cause epistemic uncertainties, inherent uncertainties simulations lead aleatory uncertainties. However, neither ‘metamodel uncertainty’ nor ‘inherent have been included in for engineering. For this reason, paper presents a metamodel that includes uncertainty risk analysis. is based on moving least squares; derived from prediction interval. The modelled with an original approach, directly using all replications scenarios without assumption specific probability distribution. generic was case study analysis road tunnel showed high accuracy. It found clear effects analysis, which makes their consideration important.