Implementing advanced Monte Carlo simulation under spreadsheet environment

This paper presents a spreadsheet computational framework for implementing an advanced Monte Carlo method called Subset Simulation for uncertainty propagation that can provide better resolution for low failure probability level at the same time retaining some robustness features of direct Monte Carlo. While the efficiency of Subset Simulation has been demonstrated by numerous studies, attention in this work is devoted to application robustness of the spreadsheet framework. This concern is relevant because advanced Monte Carlo algorithms, or in general variance reduction techniques, gain their efficiency by exploiting information about the problem, which may require intrusive exchange of information with the system analysis model during the simulation process. To explore and authenticate implementation issues, a prototype Visual Basic Application (VBA) package is developed that can perform efficient uncertainty propagation by plugging as an Add-In into a spreadsheet that performs deterministic analysis. The resulting uncertainty propagation process is non-intrusive, requiring immaterial modification of the deterministic analysis spreadsheet. Operationally the proposed framework divides the whole process into system modeling (deterministic analysis), uncertainty modeling (generation of random variables) and uncertainty propagation (Subset Simulation). It is hoped that the development work can promote the use of advanced Monte Carlo simulation tools for uncertainty propagation in the decision-making process.