Robust decision making

We are living in a world with lots of uncertainties. Weather conditions, prices, geopolitical situations, and traffic jams are only a few uncertainties that we are encountering in our daily life. Societal and industrial parties are also facing uncertainties in demands, costs, and many more parameters on which their revenues are based. So, there are two main questions: 
(i) how sensitive is the solution obtained using the customary modeling approaches? 
(ii) how can we obtain a solution that is not sensitive (or robust) against uncertainties?

The goal of this course is to teach students how to answer these questions by studying a new concept that does not depend on likelihoods but want to find solutions that are safe given the existence uncertainty. A typical approach that is known to students is the topic of probability theory and likelihoods, meaning that the students already got familiar with knowing how to find the best actions if we know some scenarios have particular probability. However, in practice, such probabilities are not there, or are not exact. Therefore, we are focusing on an alternative method here.

We are going to discuss the following topics

• Preknowledge on Duality theories in Linear Optimization problems
• Sensitivity Analysis based on dual solutions
• Static Robust Optimization (SRO), as a method to obtain robust solutions
• Adjustable Robust Optimization (ARO), as a method to obtain robust policies
• Applications of SRO and ARO in Operations Management problems

For this course, we use two structures: the traditional lecturing format and the "flipped classroom" format. While for the traditional lectures, we will get together for a lecture, for the flipped classroom, there are no physical lectures, and all the material are prerecorded so the students can watch them at their own pace.

The goal of the course is to teach students how to deal with uncertainties in an Industrial Engineering problem. Students who pass this course are able to:

• Identify existence of an uncertainty in an Industrial Engineering problem
• Formulate dual problems of a linear optimization problem
• Identify sensitivities of a solution to different uncertainties
• Recognize different methods to obtain robust policies for practical problems
• Formulate proper Robust Optimization model based on a problem description
• Implement Robust Optimization models into Python
• Solve Robust Optimization models to obtain robust policies

You must meet one of the following collections of requirements

• Collection 1
• Completed Final examination Bsc program
• Completed none of the course modules listed below
• Robust policies for OM (1CM220)
• Collection 2
• Completed Pre-Master
• Completed none of the course modules listed below
• Robust policies for OM (1CM220)

• Books/articles are available via Canvas
• Books/articles are available via Canvas