Decision Analysis with Artificial Intelligence Support (MOD500)
Immerse yourself in the intricacies of decision modeling and uncertainty analysis, exploring its vast applications. Built on the solid foundation of normative decision theory, this course combines prescriptive tools with state-of-the-art computational techniques, including ML and AI. By harnessing the power of modern software, stakeholders can adeptly navigate complex and uncertain decision terrains, culminating in insightful actions.
Course description for study year 2025-2026. Please note that changes may occur.
Course code
MOD500
Version
1
Credits (ECTS)
10
Semester tution start
Autumn
Number of semesters
1
Exam semester
Autumn
Language of instruction
English
Content
- Introduction: Embarking on decision insights and the pivotal role of data and AI.
- Uncertainty Modeling: Utilizing data analytics and AI tools to encapsulate the nuances of uncertain outcomes.
- Decision Theory: Theoretical underpinnings of informed decision-making.
- Basic Decision Analysis: Integrate Probabilistic Modeling and AI in Sensitivity Analysis, Value of Information Analysis, Risk Profiles, and Stochastic Dominance Analysis.
- Decision Modeling using Influence Diagrams: Structuring decision processes visually with AI support.
- Modeling Risk Preferences: Understanding Risk Aversion, Risk Attitudes, and Utility Functions.
- Assessment of Probability Distributions: Leveraging data analytics and AI for deeper insights into potential outcomes' probabilities.
- The Decision Analysis Cycle/Process: A systematic approach to structured decision-making.
- Multiple Objective Decision Making: How to juggle and prioritize contrasting objectives.
- ChatGPT in Decision Insights: Hands-on exploration of how ChatGPT can augment decision modeling and analysis. Learn how to integrate AI tools in your decision workflow.
Learning outcome
Knowledge:
- An understanding of modeling's role in strategic decision-making and of what constitutes a good decision model
- A thorough understanding of essential elements of good modeling principles to strive for clarity in complex and uncertain decision-making situations.
- Be able to recognize and account for the human biases and errors that most often affect decision making.
- Develop models, tools, and mental frameworks that will allow you to deal effectively with uncertainty
- Revise your beliefs after gathering additional information using Bayesian methods
- Examine and quantify the value created by gathering additional information
- Quantify your appetite for risk and how to factor this into your decision making
- Use Bayesian Networks to help structure a decision model
- Design models with a parametric approach to maximize insights
- Anticipate decision-makers' questions and design in features to answer them
- Build a decision model for a typical engineering or corporate decision situation
- Optimization between multiple strategic alternatives
- Understand sensitivity analysis and the information it provides
- Conduct probabilistic analysis to general additional insights and understand risk
- Identify how to effectively communicate the insights derived from your model
Skills:
- Skills needed to build a good basic decision model and to use it in generating powerful insights into the decision situation
- Be able to apply and construct decision models and to use the most important elements in decision analysis relevant to engineering type decision-making in the face of uncertainty.
General qualifications:
Students should understand fundamental logical principles and analyses and be able to communicate their choices and recommendations clearly.
Required prerequisite knowledge
Recommended prerequisites
Exam
Project assignment and oral exam
| Form of assessment | Weight | Duration | Marks | Aid |
|---|---|---|---|---|
| Project assignment in groups | 1/2 | 6 Weeks | Letter grades | All |
| Oral exam | 1/2 | 30 Minutes | Letter grades | None permitted |
The final grade is composed 50% by a group project and 50% by a oral exam. You must pass all parts to get a passing grade in the course. Continuous evaluation.No re-sit opportunities are offered for the assessments. Students who do not pass or wish to improve their grade, must retake the assessment parts the next time the course has regular instruction.