Course Detail (Course Description By Faculty)

Decision making plays a critical and catalytic role in the modern business world, as managers are often forced to prioritize among feasible actions in complex situations; In such scenarios, making an informed quantitative decision --versus an educated guess-- can drastically change the outcome and impact the business. However, successful quantitative decision making requires the ability to structure complex problems, to analyze available options in an uncertain world, and to finally make the best decision given the information available.

In this course, we learn basic frameworks, methodologies, and analytical tools required for quantitative managerial decision making in a wide range of business areas including operations, marketing, and finance. Our goal is learning the foundations and applying these tools to several managerial questions related to resource allocation, revenue management, market design, risk analysis, and data analysis. On our journey towards this goal, we learn how to "structure a decision problem" by identifying objectives, decision alternatives, input parameters, and sources of uncertainty. We then learn how to "build mathematical models to formalize decision problems" by applying analytical tools such as optimization, simulation, and decision trees. We then switch gears to "analyzing model solutions" by applying basic statistics, probability, and stochastic processes. If time permits, we will cover some basics in machine learning that can be helpful in quantitative decision making.

[New addition] Given the transformative nature of AI, this technology has drastically changed the way managers think about mathematical and spreadsheet modeling. We will explore ideas and tools related to AI that can help with managerial decision modeling. In particular, we will learn how to use off-the-shelf frontier LLMs---in particular ChatGPT/Codex, Claude Code/Co-work, and Cursor---to interpret business cases, translate them into mathematical and spreadsheet models, and eventually solving them. We will also explore ways we can use AI to verify models and run sensitivity analysis. 

We mainly use Microsoft Excel as a platform for model building, solution, and analysis: In addition to standard Excel tools such as Goal Seek and Data Table, we will learn to use important Excel add ons such as Sensitivity Toolkit, Solver, SolverTable, Precision Tree, @RISK, and RiskOptimizer. These tools can also be used in other Booth classes. If time permits, we will also try to have a crash tutorial on simulation and modeling in Python, which will be helpful to prepare for more advanced decision modeling courses such as Bus 36109.

Any previous or concurrent exposure to statistics at the level of 41000, financial accounting at the level of 30000, and microeconomics at the level of 33001 will be helpful, but not strictly required. Although the example models discussed in this class cross many functions of business, little or no prior background in those areas is required.

It is assumed that students have some familiarity with Excel. However, one does not have to be an Excel expert to benefit from the course. Knowing how to enter and copy simple formulas involving relative and absolute cell addresses, how to use general-purpose Excel functions (for example, the If() function) and how to draw different types of graphs in Excel should be sufficient. There will be an Excel review session during the first week to help set the expectations.

This course involves in-class software demonstrations and "hands-on" practices. Students will be expected to bring their laptop to class each week.

Cannot enroll in BUSN 36106 if BUSN 20510 taken previously.

Based on exams, homework assignments, which include spreadsheet model building and case analyses, and class participation. We have take home Midterm and Final exam, both open book/open notes and mandatory. Computers will be needed. Cannot be taken pass/fail. No auditors.

Grades

Restrictions