Project Grant 2432256

This $400,438 Project Grant awarded by the National Science Foundation (NSF) Engineering program (CFDA 47.041) will support research to improve methods for resource allocation decisions when problem parameters are highly uncertain and data is sparse. The project, conducted by the Regents of the University of Michigan - University of Michigan-Dearborn, will develop a statistical bootstrapping approach called the Average Percentile Upper Bound (APUB) to provide robust upper confidence bounds for population means. This will enhance the reliability, efficiency, and sustainability of optimization processes, particularly for applications like electric vehicle charging infrastructure deployment where demand data is limited. The research aims to advance the theoretical foundations of APUB, increase its computational efficiency, improve its adaptability to varying uncertainty levels, and validate its practical effectiveness through case studies. The project represents a collaboration with Argonne National Laboratory and will involve undergraduate and graduate student participation.