Decision Analysis

Jun Zhuang,
SUNY at Buffalo
Author Profile

Summary

This course provides an overview of modeling techniques and methods used in decision analysis, including multiattribute utility models, decision trees, and Bayesian models. Psychological components of decision making are discussed. Elicitation techniques for model building are emphasized. Practical applications through real-world model building are described and conducted, including business management, supply chain and logistics, transportation, health care, and homeland security. Each group,which consists 1-3 students, will work on a project throughout the semester, including oral presentations and written reports.


Course Size:
71-150

Course Format:
Lecture only

Institution Type:
default

Course Context:

This is an elective course for senior undergraduate and graduate students, most of them are from the Department of Industrial and Systems Engineering, University at Buffalo. Some other students came from other departments in the School of Engineering and Applied Sciences; and some came from the School of Management.

Course Content:

The ability to make good decisions is a fundamental skill for engineers and managers in any organization. This course will teach the skills and concepts that you need to make better decisions. The ideas will be applicable in your personal life as well as your professional life. We will learn quantitative techniques for identifying good decisions in complex situations, but also general concepts that can help you even if you don't use those quantitative techniques. These ideas can help you deal with many of the things that can make decision-making difficult, such as uncertainty about future outcomes, tradeoffs between competing objectives ("comparing apples and oranges''), and nonlinearity of preferences (e.g., the fact that twice as much of something may not be twice as good).

Course Goals:

After successfully completing this course, the students should be able to: (a) Recognize the types of problems that decision analysis can and can't address; (b) Identify the values, objectives, attributes, decisions, uncertainties, consequences, and trade-offs in a real decision problem; (c) Apply the concepts learned in this class (expected value, value of information, risk aversion, and tradeoffs between attributes) to identify good decisions and strategies; (d) Represent a decision problem graphically and mathematically; (e) Determine the optimal decision mathematically; (f) Identify which parameters have the most impact on the results of an analysis; and (g) Explain the results of a decision analysis to managers and other non-specialists.

Course Features:

Several decision scenarios related to sustainability and resilience will be discussed in the class. The students are also welcome to work projects related to sustainability and resilience to disasters. The projects will be documented and presented in class.

Course Philosophy:

This course was designed following a campus-wide initiative to Extreme Events, and along with my research on optimization and decision making in homeland security problems.

Assessment:

Homework (30%), First mid-term exam (20%), Second mid-term exam (20%), Project written proposal (5%), Project oral proposal presentation (5%), Project oral final presentation (5%), and Project written final report (15%).

Syllabus:

Syllabus (Acrobat (PDF) 90kB May7 14)

References and Notes:

Making Hard Decisions with DecisionTools, By Robert T. Clemen and Terence Reilly, Cengage Learning, 3rd edition (updated May 13, 2013). ISBN-10: 0538797576. ISBN-13: 978-0538797573.
Smart Choices: A Practical Guide to Making Better Decisions, By
John S. Hammond, Ralph L. Keeney, and Howard Raiffa,
New York, NY: Broadway Books, 2002. ISBN: 978-0875848570.