Decision-Making, Design, and Strategy Under Uncertainty

Lead Instructor(s): 

Richard de Neufville
Mort Webster

Date: 

TBD

Course Fee: 

$4,100

CEUs: 

2.9

Status: 

  • Closed

It is highly recommended that you apply for a course at least 6-8 weeks before the start date to guarantee there will be space available. After that date you may be placed on a waitlist. Courses with low enrollment may be cancelled up to 4 weeks before start date if sufficient enrollments are not met. If you are able to access the online application form, then registration for that particular course is still open.

COURSE SUMMARY

An all too common practice in industrial or policy planning is to use a best-guess forecast, and optimize the design, strategy, operation, or policy for that forecast. Unfortunately, this practice systematically leads to inefficient and generally undesirable outcomes because it does not explicitly consider uncertainty during the design/planning stage.

This course is designed to achieve two critical objectives:

  1. To increase your awareness and appreciation for WHY uncertainty matters
  2. To give you the TOOLS to characterize uncertainty and to design flexible strategies that will be robust to uncertainty.

The primary focus of this course is on the concepts and the intuition. Numerous real-world examples will demonstrate the consequences of ignoring uncertainty and what can be done instead. Through hands-on exercises, the course will introduce you to existing software tools that can aid in designing effective flexible strategies, including Decision Analysis, Real Options, and Monte Carlo simulation. Finally, substantial time will be devoted to applying the tools and concepts that participants bring from their own professional contexts, with iteration and feedback from instructors throughout the week.

This short course will provide the tools and more importantly the conceptual framework for improving the flexibility and robustness of designs, strategies, and decisions. The degree of technological, market, and regulatory uncertainty is sufficiently large that designs and strategies that do not explicitly build in flexibility often perform poorly. Participants who complete this course will acquire the tools and the framework for improving their decisions and operations and getting significantly better performance.

Takeaways from this course include:

  • Developing awareness of the pervasiveness of uncertainty and its consequences for decision-making
  • Understanding how to use Monte Carlo simulation and Risk Analysis tools to characterize uncertainty
  • Understanding how to use Decision Trees to structure a design or decision problem, and how to use it to identify potential flexibility
  • Understanding how to use Lattice Models for problems with more decision points
  • Developing a framework for thinking about flexible design, strategy, or decision under uncertainty

Who should attend:

This course will be useful to anyone involved in designing, developing, or operating complex technological systems or businesses that operate in conditions of uncertainty and volatility. Past participants have included executives, managers, senior engineers, CFOs, strategy planning professionals, and research scientists. The tools and concepts of this course can be readily applied to many sectors, including energy, biotech, construction, financial, engineering, and manufacturing.

Computer Requirements:

Laptops are required for this course. Participants will need to have Microsoft Excel installed and have a working familiarity with the program, including familiarity with basic formulas. Class exercises will involve using the quantitative tools being taught.

Program Outline: 

This course will be organized each day around the following objectives:

  • Introduce key concepts for identifying and evaluating flexible alternatives
  • Introduce a quantitative/computational method for evaluating alternatives
  • Provide hands-on experience with the quantitative tool
  • Apply the concepts and tools to a problem that participants bring from their professional contexts

Each day will build incrementally on the previous days, resulting in a complete analysis of participants' sample problems by the end of the week.

Day One (Intro and Uncertainty)

  • Session 1—1.25 hours (9:30-10:45)
    • Introduction to course, decision-making under uncertainty, Flaw of Averages
  • Break (10:45-11:00)
  • Session 2—1.5 hours (11:00am-12:30pm)
    • Uncertainty: concepts, examples

Lunch (12:30-2:00)

  • Session 3—1.5 hours (2:00-3:30)
    • Uncertainty tools: probability distributions and Monte Carlo simulation

Break (3:30-3:45)

  • Session 4—1.25 hours (3:45-5:00)
    • Exercise 1: Uncertainty and Monte Carlo
  • Session 5—1.5 hours (5:00-6:30)
    • Networking reception

Day Two (Decisions under Uncertainty)

  • Session 6—1.75 hours (9:00-10:45)
    • Structuring Decisions, Decision Trees, Solving Decision Trees

Break (10:45-11:00)

  • Session 7—1.5 hours (11:00-12:30)
    • Sensitivity Analysis, value of information, risk preferences

Lunch (12:30-2:00)

  • Session 8—1.5 hours (2:00-3:30)
    • Exercise 2: Decision Trees

Break (3:30-3:45)

  • Session 9—1.25 hours (3:45-5:00)
    • Generating flexible options

Day Three (Lattices)

  • Session 10—1.75 hours (9:00-10:45)
    • Binomial Lattices

Break (10:45-11:00)

  • Session 11—1.5 hours (11:00-12:30)
    • When to use a lattice

Lunch (12:30-2:00)

  • Session 12—1.5 hours (2:00-3:30)
    • How to use a lattice: Copper Mine case study

Break (3:30-3:45)

  • Session 13—1.25 hours (3:45-5:00) 
    • Exercise 3: Lattice analysis

Day Four (Real Options and Advanced Topics)

  • Session 14--1.75 hours (9:00-10:45)
    • Real options I

Break (10:45-11:00)

  • Session 15—1.5 hours (11:00-12:30)
    • Garage case

Lunch (12:30-2:00)

  • Session 16—1.5 hours (2:00-3:30)
    • Real options II

Break (3:30-3:45)

  • Session 17—1.25 hours (3:45-5:00)
    • Exercise 4: Real options and simulation

Day Five (Advanced Topics and Wrap-Up)

  • Session 18—1.75 hours (9:00-10:45)
    • Large-scale problems: stochastic programming and dynamic programming

Break (10:45-11:00)

  • Session 19—1.5 hours (11:00-12:30)
    • Exercise 5: Application to participants' decision problems
  • Session 20—1.5 hours (12:30-2:00)
    • Course debrief, feedback, and awarding of certificates (Lunch provided)
    • Course concludes (2:00) 

Included in Tuition:
With course fee, participants will also receive:

  • The book Flexibility in Engineering Design MIT Press, 2011 (reprinted 2016) by de Neufville and Scholtes (signed copy will be sent upon payment for course)
  • A USB drive with numerous reference materials as PDF files
  • Networking reception for all participants and a guest on Monday evening
  • Lunch on Friday

Course Schedule: 

Class runs 9:30 am - 5:00 pm on Monday and 9:00 am - 5:00 pm the rest of the week except for Friday when it ends at 2:00 pm.

Instructors: 

Location: 

This course takes place on the MIT campus in Cambridge, Massachusetts. We can also offer this course for groups of employees at your location. Please complete the Custom Programs request form for further details.

Content: 

Fundamentals: Core concepts, understandings, and tools (60%) 60
Latest Developments: Recent advances and future trends (15%) 15
Industry Applications: Linking theory and real-world (25%) 25

Delivery Methods: 

Lecture: Delivery of material in a lecture format (40%) 40
Discussion or Groupwork: Participatory learning (30%) 30
Labs: Demonstrations, experiments, simulations (30%) 30

Levels: 

Introductory: Appropriate for a general audience (50%) 50
Specialized: Assumes experience in practice area or field (30%) 30
Advanced: In-depth explorations at the graduate level (20%) 20