Stochastic simulation is among the most widely used tools in industry, government, and academia for analyzing the impact of changes to a system. Operations engineers use simulation models to investigate feasible system designs that do not yet exist in reality. Modern simulation problems often feature a massive number of feasible configurations, each of which may require substantial computational effort to evaluate their performance via simulation. However, state-of-the-art general-purpose methods are insufficient for the modern large-scale simulation problems. The proposed work will develop methods that achieve scalable simulation inference by exploiting more information than is currently employed by general-purpose algorithms. The idea of the work is to extract additional structural information from simulations using the grey-box nature of modern simulations. These novel methods will improve decision-making and deliver powerful inference in large-scale environments that are beyond the capabilities of existing simulation analysis tools.
|Effective start/end date||8/1/22 → 7/31/25|
- National Science Foundation (CMMI-2206973)
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.