OPIM101 - Spring 2012 - Lecture 6

OPIM101 - Spring 2012 - Lecture 6 - Structured Decision...

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
Slide 1 OPIM 101, Spring 12, Profs Cachon/Savin, Lecture #6 Structured Decision Making Lecture 6 February 21, 2012
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Slide 2 OPIM 101, Spring 12, Profs Cachon/Savin, Lecture #6 Complex decisions § Research and invest in a compound to treat malaria? § Develop a track of land in the Gulf of Mexico to produce oil and natural gas? § Hire an applicant for a consulting position? § Finance the development of a movie script? § Settle out of court or go to trial? § Perform a medical test or not?
Background image of page 2
Slide 3 OPIM 101, Spring 12, Profs Cachon/Savin, Lecture #6 Investing in a new compound to develop a new drug § Developing the drug will require additional investment: It will cost $80 million to move the compound through clinical trials. § Developing the drug is risky: Clinical trials will judge the compound on two dimensions: u Efficacy – is it an effective treatment? u Toxicity – how harmful are the side effects? There is a 25% chance it will pass clinical trials. 25% is called a prior probability – your initial probability assessment. § Developing a drug is potentially profitable: § If it passes clinical trials, it will generate $360 million in future profit, (not including the development cost) § If it fails, future profit is $0.
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Slide 4 OPIM 101, Spring 12, Profs Cachon/Savin, Lecture #6 Some data on drug development Indication/disease Phase 1 Phase 2 Phase 3 Overall Alzheimer's disease 0.08 0.24 0.37 0.01 Anxiety disorder 0.01 0.07 0.17 0.00 Breast cancer 0.41 0.48 0.89 0.18 Congestive heart failure 0.27 0.47 0.64 0.08 Epilepsy 0.17 0.39 0.65 0.04 Erectile dysfunction 0.48 0.60 0.80 0.23 Hepatitis B 0.71 0.77 0.97 0.53 Migrane 0.27 0.45 0.72 0.09 Stroke 0.07 0.23 0.34 0.01 Historical probability of success at each clinical milestone for a sample of compounds
Background image of page 4
Slide 5 OPIM 101, Spring 12, Profs Cachon/Savin, Lecture #6 A simple decision tree Invest Abandon Fail (0.75) Pass (0.25) $280 = $360 - $80 -$80 $0 = action or decision node = event node probability payoff outcomes Outcome description Action description
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Slide 6 OPIM 101, Spring 12, Profs Cachon/Savin, Lecture #6 A couple of rules about events and probabilities § Each event node must have an outbound arrow for all possible outcome of that event. § Each possible outcome for that event must have a probability associated with it. § The sum of the probabilities must equal 1.00: Pr(Pass) + Pr(Fail) = 1.00 0.25 + 0.75 = 1.00 Invest Abandon Fail (0.75) Pass (0.25) $28 0 -$80 $0
Background image of page 6
Slide 7 OPIM 101, Spring 12, Profs Cachon/Savin, Lecture #6 What should we do? One option is Maximin § The maximin decision criterion: For each action evaluate the minimum payoff that could occur. Choose the action that yields the maximum minimum payoff. § In our example:
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 8
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 34

OPIM101 - Spring 2012 - Lecture 6 - Structured Decision...

This preview shows document pages 1 - 8. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online