Linda Nozick is Professor and Director of Civil and Environmental Engineering at Cornell University. She is co-founder and a past director of the College Program in Systems Engineering and has been the recipient of several awards, including a CAREER award from the National Science Foundation and a Presidential Early Career Award for Scientists and Engineers from President Clinton for “the development of innovative solutions to problems associated with the transportation of hazardous waste.” Dr. Nozick has authored over 60 peer-reviewed publications, many focused on transportation, the movement of hazardous materials, and the modeling of critical infrastructure systems. She has been an associate editor for Naval Research Logistics and a member of the editorial board of Transportation Research Part A. Dr. Nozick has served on two National Academy Committees to advise the U.S. Department of Energy on renewal of their infrastructure. During the 1998-1999 academic year, she was a Visiting Associate Professor in the Operations Research Department at the Naval Postgraduate School in Monterey, California. Dr. Nozick holds a B.S. in Systems Analysis and Engineering from the George Washington University and an MSE and Ph.D. in Systems Engineering from the University of Pennsylvania.
Statistics is about using data to estimate certain values and evaluate certain hypotheses; this makes perfect sense for passively studying how the world works (i.e., the scientific method). More often than not, however, we find ourselves wanting to use this statistical information to make decisions regarding the systems involved. Suppose we estimate that the demand for jet fuel next month will be greater than normal. How does this information affect the decision of an oil refinery to purchase crude oil from their various sources? How does an airline company decide how many flight crews to employ based on the current flight schedule? How does past sales information across the U.S. influence a company's decision over where to place its warehouses?
The quantification and mathematical solution of these types of decision-making problems are known broadly as optimization. The general features of an optimization problem are a set of quantifiable decisions that have a quantifiable effect that should be minimized or maximized (think cost or revenue) and a set of constraints on the possible values of those decisions. There are many different optimization branches, but the most prominent, due to its widespread applicability and computational efficiency, is linear programming, where the objective function and constraints are all linear.
In this course, you will explore the mathematics of linear programs, how to solve them, and how to evaluate your model. You will implement these techniques using packages in the free and open-source statistical programming language R to solve real-world logistical business problems. The focus will be on making these methods accessible for you in your own work.
You are required to have completed the following courses or have equivalent experience before taking this course:
- Understanding Data Analytics
- Finding Patterns in Data Using Association Rules, PCA, and Factor Analysis
- Finding Patterns in Data Using Cluster and Hotspot Analysis
- Regression Analysis and Discrete Choice Models
- Supervised Learning Techniques
- Neural Networks and Machine Learning
Key Course Takeaways
- Create linear programs by specifying decision variables, a linear objective function, and linear constraints
- Compute the optimal solution to a linear program using R
- Use linear programming to solve real-world logistical problems
- Integrate uncertainty into optimization models
How It Works
Who Should Enroll
- Current and aspiring data scientists
- Technical managers