Academics

Engineering faculty receives NSF CAREER Award for simulation analysis research

UNIVERSITY PARK, Pa. — Eunhye Song, Harold and Inge Marcus Early Career Assistant Professor of Industrial and Manufacturing Engineering at Penn State, received a National Science Foundation (NSF) Faculty Early Career Development (CAREER) Award to support her work in simulation analysis and quantifying model risk in simulations. 

Eunhye Song, Harold and Inge Marcus Early Career Assistant Professor of Industrial and Manufacturing Engineering at Penn State, received a National Science Foundation (NSF) Faculty Early Career Development (CAREER) Award. Credit: Photo provided by Eunhye SongAll Rights Reserved.

The NSF CAREER Awards are given to early-career faculty who have potential to serve as academic role models in research and education and lead advances in the mission of their department or organization, according to the NSF website.

“Writing the proposal for the CAREER grant was a great way to reflect on my research contributions and plan for my mid-term and long-term career goals,” Song said. “I am incredibly excited to have been awarded the grant and to move forward to achieve my career goals.”

Song’s research focuses on analyzing system performances using stochastic simulation. Many industries have adopted stochastic simulation models to emulate real-world system performances, predict trends and support operational decisions. Researchers consider the models “stochastic” due to the randomness in the simulation inputs that the models generate to mimic real-world systems’ unpredictability. Because it is impossible to perfectly characterize the real-world systems’ unpredictability, these models are subject to model risk, which refers to the risk of making an incorrect statistical inference because the model does not accurately emulate the target real-world system. Song works to quantify this model risk and to reduce the risk in decisions made using the models. 

Song and her research team specifically investigate when the input distributions of the simulator are estimated from real-world data. When estimation errors occur, the simulators will less accurately predict the real-world systems’ performances, potentially causing profit loss, low service quality or safety hazards. 

“Our solution involves designing efficient simulation experiments, selecting and fitting a statistical model that provides good inference as well as creating an optimization algorithm to find a solution to the decision-making problem the simulator is built to support,” Song said.

Song also plans to use the CAREER Award to expand the educational aspect of her research. Her goal is to train current and next-generation STEM workers on the importance of model risk and making it the focus of simulation analysis. Song also plans to collaborate with Simio, a market leading discrete-event simulation software company, to train the current simulation workforce on model risk. In 2020, Song created a pilot program, called Discover Operations Research (DiscovOR), that established learning modules for high school students. Plans to provide guest lectures in high school engineering classes were put on hold because of the ongoing pandemic, but Song said she is excited to use the CAREER Award to expand the DiscovOR program and revisit the lectures in the future. 

“I feel passionate about the education plans of the award,” Song said. “I'm looking forward to expanding and running the DiscovOR program once we all go back to normalcy.”

Last Updated April 22, 2021

Contact