OIDD 261 Lecture Notes - Lecture 3: Sensitivity Analysis
Notes from Slides 4:
Concept 1: Exposure to the Risk
• probabilistic --not clear person is increasing risk of cancer by being exposed to radon or
working with chemical
• Deterministic--certain that worker is being exposed to increased toxicity by being
exposed to radon or working with a chemical
Concept 2: Effect of Exposure
• Uncertain effect--no cancer/cancer
• Deterministic effect--cancer
Exposure and Effect translate into the standard elements of any decision problem
Actions to Reduce Risk of Cancer from Radon
• Modify the environment
• Modify exposure processes
• Modify effects processes
**Green Radon Problem Example**
Structuring Decision Problems
• Nature of the Problem
• Generate alternatives
• Specify objectives
• Impact of Different Alternatives
• Specify uncertainties
• Specify consequences
• Determine Preferences of Decision Makers
• Value tradeoffs
• Combining many objectives
• Evaluate and Compare Alternatives
• Developing scenarios
• Sensitivity analysis
• Select Best Choice
Challenges in Undertaking Decision Analysis
Estimating Risks
• Disagreements among experts on nature of risk
• Very poor dose-response curves
• Little historical data on risk
Other Dimensions of Risk
• Fear of knowing one has radon
• Anxiety of contracting cancer
• Fear of large losses (catastrophic)
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Concept 2: effect of exposure: uncertain effect--no cancer/cancer, deterministic effect--cancer. Exposure and effect translate into the standard elements of any decision problem. Actions to reduce risk of cancer from radon: modify the environment, modify exposure processes, modify effects processes. Structuring decision problems: nature of the problem, generate alternatives, specify objectives. Impact of different alternatives: specify uncertainties, specify consequences, determine preferences of decision makers, value tradeoffs, combining many objectives, evaluate and compare alternatives, developing scenarios, sensitivity analysis, select best choice. Estimating risks: disagreements among experts on nature of risk, very poor dose-response curves, little historical data on risk.