Computational

Mathematics

& Statistics

Computational

Mathematics

& Statistics

Solving problems with

statistics and mathematics

Solving problems with

statistics and mathematics

How can our nation predict the behavior of complex natural and man-made systems and find critical signatures amidst overwhelming data? PNNL helps decision makers be better informed when making choices in mission-critical domains.

PNNL has one of the largest and most accomplished teams of statisticians and computational scientists in the U.S. Department of Energy’s national laboratory complex. As leaders in mathematics research, mathematicians at PNNL develop novel computational modeling and data-analysis tools to predict and control complex systems and extract hidden features, anomalies, and signatures that support discovery and optimize data-gathering approaches through sampling and experimental design. Our applied mathematics and statistics capabilities complement research conducted throughout the laboratory, fueling fundamental understanding of physical, chemical, and biological principles using computational modeling, experimentation, and data evaluation.

Through rigorous analysis, we turn science and programmatic challenges into mathematical problems. An iterative process, our signature discovery spans exploratory data analysis, formal experimentation, and multiscale, multiphysics, and machine learning algorithms development and implementation. We build effective and efficient signature discovery strategies to rightsize the proper solution. When done right, effective and efficient signature discovery leads to understanding signatures and correctly identifies when a phenomenon is present.

PNNL is advancing mathematical theory to build strategically intelligent and resilient computational systems. We develop reasoning within artificial intelligence systems to sense application environments and adapt algorithms to optimally intersect with a dynamic system state.

Under a unifying theme of doing more with less, we design effective resource utilization strategies. PNNL researchers develop solutions for deploying limited resources, implementing effective supply chains, and analyzing concepts of operations to manage risks and achieve operational objectives in the face of uncertainties.

We design, implement, and validate software, tools, and statistical package products that allow deployable solutions within any research capability. From multiscale modeling for bioremediation to operations research and industrial engineering, all the way to climate research and human behavior modeling, our applied mathematics and statistics fuel fundamental understanding of physical, chemical, and biological principles using computational modeling, experimentation, and data evaluation. PNNL consistently delivers high-quality methods and applications to address critical scientific challenges in national security and energy resiliency.