Nathaniel D. Bastian
Instructor of Supply Chain & Information Systems
A leader, practitioner, researcher, and educator of mathematical, computational, analytical, data-driven, and decision-centric methods to support the improvement and enhancement of decision-making in cyber security, national defense, military operations, human resources and manpower, healthcare, logistics, energy and finance. My expertise lies in the scientific discovery and translation of actionable insights into effective decisions using algorithms, techniques, tools and technologies from operations research, data science, artificial intelligence, industrial engineering, and economics to design, develop, deploy and operationalize decision-support models for descriptive, predictive and prescriptive analytics. Teaching is enjoyable because it gives me the opportunity to interact with and engage students intellectually by exposing them to the data analytic process of transforming data into actionable insight for more informative decision-making. I thoroughly enjoy teaching at PSU because I can help students become more effective problem-solvers to tackle some of the most complex, challenging problems facing our nation.
- Multiple objective optimization, stochastic programming, and approximate dynamic programming
- Monte Carlo methods, deep reinforcement learning, pattern recognition, and artificial intelligence at scale
- Productivity analysis, efficiency/performance measurement, cost-effectiveness analysis, and econometric modeling
- Network science, graph mining and social network analysis in real-world, complex networks
Ph D, Industrial Engineering and Operations Research, Pennsylvania State University, 2016
ME, Industrial Engineering, Pennsylvania State University, 2014
MS, Econometrics and Operations Research, Maastricht University, 2009
BS, Engineering Management (Electrical Engineering) with Honors, U.S. Military Academy, 2008
DAAN 881 – Data Driv Dec Mkng (3)
Application & interpretation of analytics for real-life decision making. DAAN 881 Data-Driven Decision Making (3) The theory and application of several quantitative decision-making tools will be studied. The usefulness of these tools will be illustrated using projects and case studies throughout the course. Emphasis will be placed on the application of the tools and techniques and the results they generate. Finding patterns in data and appropriately grouping them are essential in the extraction of information in large datasets. This course will use Principal Component Analyses to transform highly correlated sets of data by means of orthogonal transformation. Cluster analysis will be used to properly group data when working with large datasets. When the outcomes involve categorical variables, Logistic regression techniques will be used to estimate the probabilistic values of the output. The decision space will be divided into smaller regions using Regression tree analyses. When factors are too numerous and highly collinear, Partial Least Square Regression methods will be performed.Public access datasets in the healthcare, transportation and finance industries will be used to demonstrate the applications and the limitations of these techniques.
BAN 888 – Implementing Analytics for Business (3)
This capstone course sets business analytics problem solving in a real-world context, including communication to non-statistically trained executives. Key topical areas are derived from the common activities of the business analyst and include business problem framing, analytics problem framing, data sourcing, cleaning and integration, analysis methodology selection, model building, model deployment and model lifecycle management including benefit assessment. Topics align with the body of knowledge in the Institute for Operations Research and the Management Sciences (INFORMS) Certified Analytics Professional Study Guide. Students explore each topic in a real-world context, by developing business analytics solutions to an ongoing course project in a team setting. Each team selects a project and works through all elements of the business analytics body of knowledge.