Influence Diag And Decision Trees - Guillaume Nicaise

4m ago
4 Views
1 Downloads
883.50 KB
26 Pages
Last View : 2m ago
Last Download : 3m ago
Upload by : Lucca Devoe
Transcription

Decision & Risk Analysis Influence Diagrams, Decision Trees Reference: Clemen & Reilly. Making Hard Decisions, 2nd ed. Chapter 3. Duxbury, 2001 NOTE: Some materials for this presentation courtesy of Dr. Dan Maxwell

Decision & Risk Analysis The Elements of a Decision Objectives and means Alternatives to choose between Uncertainty in Events and Outcomes Consequences of the Decision How do we structure the problem? Common ways of structuring the decision problem: Influence Diagrams Decision Trees

Decision & Risk Analysis Influence Diagrams Three Levels of Specification Chance (outcome) Influence Decision Value Probabilistic Relevance Information Arc Chance (Outcome) Deterministic Chance Functional Relevance Courtesy of Dr. Dan Maxwell

Decision & Risk Analysis A Simple Influence Diagram Figure 3.5

Decision & Risk Analysis Showing Influence with Arrows

Decision & Risk Analysis Objectives Hierarchy into Influence Diagram Maximize Overall Satisfaction Invest in Computer Industry Return on Investment Venture Succeeds or Fails Invest? Return on Investment Computer Industry Growth Overall Satisfaction Figure 3.6

Decision & Risk Analysis Multiple Objectives Note: The Oval Shape can be Used as well as the Diamond Shape for the Final Satisfaction Node Figure 3.7

Decision & Risk Analysis Basic Diagram for Investment Wild Success Flop Savings Business Figure 3.9 Savings Success Flop Business Success Flop 2200 2200 5000 0

Decision & Risk Analysis An Influence Diagram for Evacuation FIGURE 3.11

Decision & Risk Analysis Sequential Decision Problems FIGURE 3.13 Note: Payoff Cash Flow; Total Payoff Net Present Value

Decision & Risk Analysis A Simple Influence Diagram for New Product NOTE alternative method of representing payoff (consequence) node FIGURE 3.14

Decision & Risk Analysis New Product - Additional Detail Profit FIGURE 3.15

Decision & Risk Analysis Deterministic Nodes Units Sold Price? Introduce Product? Fixed Cost Revenue Unit Variable Cost Cost Profit Figure 3.16

Decision & Risk Analysis Another Representation of Deterministic Nodes FIGURE 3.16

Decision & Risk Analysis Starting an Influence Diagram FIGURE 3.18

Decision & Risk Analysis Intermediate Step FIGURE 3.19

Decision & Risk Analysis A Completed Influence Diagram FIGURE 3.20

Decision & Risk Analysis Decision Tree Representation Influence Diagram Representation FIGURE 3.21

Decision & Risk Analysis Decision Trees – Dealing with Uncertainty Probabilities replace the weights – Account for uncertainty – Used to evaluate expected values Example - Venture Capital Problem Large return p Uncertain outcome represented by circle Decision represented by box est v in Do in no ve t st success failure 1-p R Loss L 0 Small return savings r

Decision & Risk Analysis Decision Trees (continued) Expected return on investment: – If investment is made E(I) pR (1 - p)L – If investment not made E(N) r Decision: – Invest if pR (1 - p)L r – Don’t invest if r pR (1 - p)L What would you do if r pR (1 - p)L? Decision Trees evaluated left to right decision must be made before uncertain event takes place decision is conditional on the known outcome of the uncertain event

Decision & Risk Analysis Example Do I take my umbrella or not? – If I don’t and it is sunny, that is best - value 1.00 – If I do I won’t get wet but it’s inconvenient - value 0.80 – If I don’t and it rains I ruin my suit - value 0.00 – Probability of sunshine is p Payoff Payoff Calculation: sunshine ke ella a T br Um Do Um n’t t br ake ell a p rain 1-p sunshine p rain 1-p 0.80 0.80 1.00 0.00 – Take umbrella 0.8p 0.8(1-p) 0.8 – Don’t take umbrella 1.0p 0(1-p) p Therefore: Take umbrella if p 0.8

Decision & Risk Analysis Basic Decision Tree FIGURE 3.25

Decision & Risk Analysis Sequential Decisions Evacuate Now Safety, High Cost Stay Forecast: “Will Hit Miami Wait for Forecast Stay Forecast: “Will Miss Miami Figure 3.29 Hurricane Hits Miami Danger, Low Cost Hurricane Misses Miami Safety, Low Cost Evacuate Safety, High Cost Hurricane Hits Miami Danger, Low Cost Hurricane Misses Miami Safety, Low Cost Evacuate Safety, High Cost

Decision & Risk Analysis Sequential Decision Tree FIGURE 3.30

Decision & Risk Analysis Influence Diagram or Decision Tree Influence Diagram Decision Trees 1. Gives basic information 2. Less messy 1.Gives detailed info 2.More messy due to greater details 3.Not so appealing 3. Graphically more appealing when presented to upper management Must be viewed as complementary techniques. One strategy is to start with influence diagram and fill in the details to develop a decision tree.

Decision & Risk Analysis Read Chapter 3. Next class: Solving Influence Diagram and decision trees (Chapter 4)

Influence Diagram or Decision Tree Influence Diagram Decision Trees 1. Gives basic information 1.Gives detailed info 2. Less messy 2.More messy due to greater details 3. Graphically more appealing 3.Not so appealing when presented to upper management Must be viewed as complementary techniques. One strategy is to start with influence diagram and .

Related Documents:

Influence Diagram or Decision Tree Influence Diagram Decision Trees 1. Gives basic information 1.Gives detailed info 2. Less messy 2.More messy due to greater details 3. Graphically more appealing 3.Not so appealing when presented to upper management Must be viewed as complementary techniques. One strategy is to start with influence diagram and .

Decision & Risk Analysis Influence Diagram or Decision Tree Influence Diagram Decision Trees 1. Gives basic information 1.Gives detailed info 2. Less messy 2.More messy due to greater details 3. Graphically more appealing 3.Not so appealing when presented to upper management Must be viewed as complementary techniques. One strategy is toFile Size: 871KB

The Man Who Planted Trees Curriculum Guide ABOUT TREES, pg. 1 ABOUT TREES, cont. pg. 12 Author Jean Giono once said that he wrote The Man Who Planted Trees because he wanted to “to make people love trees, or more precisely to make people love planting trees.” Read on to learn more about trees and the many

deletion operations, importance of balancing, AVL trees, searching insertion and deletions in AVL trees, redblack trees, comparison - with AVL trees, search insert and delete operations. 4 7. Multiway Trees: Issues in large dictionaries, m-way search trees, B-trees, search insert and delete operations, height of B-tree

Compare and contrast the Net and Applet Java packages 4. To develop Java application using Servlet . Trees - Binary trees - Binary tree representation and traversals - Threaded binary trees - Expression Trees -Binary Search Tree - Applications of trees. Balanced trees: AVL trees. Priority queue - Binary heap - Heap operations - Applications .

Trees Removed - Rural 60 50 70 50 75 125 100* Trees Planted 323 242 375** 245 162 80 200* *Forecast ** The 2019 "Trees Planted" amount reflects 225 City trees, plus 150 Ballpark Commons trees *** With more developments, we will need to plant more development trees and replacement trees

Decision Trees for Decision-Making under the Predict-then-Optimize Framework Adam N. Elmachtoub1 Jason Cheuk Nam Liang2 Ryan McNellis1 3 Abstract We consider the use of decision trees for decision-making problems under the predict-then-optimize framework. That is, we would like to first us

ASME A17.1-2019/CSA B44:19 (Revision of ASME A17.1-2016/CSA B44-16) Safety Code for Elevators and Escalators. Includes Requirements for Elevators, Escalators, Dumbwaiters, Moving Walks, Material Lifts, and Dumbwaiters With Automatic Transfer Devices. AN AMERICAN NATIONAL STANDARD. ASME A17.1-2019/CSA B44:19 (Revision of ASME A17.1-2016/CSA B44-16) Safety Code for Elevators and Escalators x .