Decisions Under Generalized Uncertainty

"Essentials of Decision Making Under Generalized Uncertainty"

T. Whalen & C. Bro/nn, 1988

In Kacprzyk & Fedrizzi: Combining Fuzzy Imprecision With Probabilistic Uncertainty in Decision Making, Springer-Verlag, 1988

I. Introduction: Uncertainty, Fuzziness, and Optimization

A. Certainty
B. Risk:
           States, Actions, Preferences, and Probabilities known
C. Generalized Uncertainty:
           States and Actions Known,Weak Lnowledge of Preference & Propensity
D. Ignorance:
           States, Actions, and Preferences Known, No Usable Knowledge of Propensity
E. Ill-Structured: Weak Knowledge of States and Actions
II. Obstacles to Certainty
    A. Uncertainty about alternative courses of action
    B. Uncertainty about consequences
    C. Uncertainty about preferences
    D. Sequentiality (uncertainty about dealing with uncertainty)
III. Methodologies for Decision Making Under Generalized Uncertainty
    A. Decision analysis typology
    B. Decision analysis with numeric or fuzzy numeric utilities
      1. no relative possibility information
      2. ordinal possibilities
      3. fuzzy or crisp numeric probabilities
    C. Decision analysis with ordinal utility
      1. no relative possibility information
      2. ordinal possibilities
        complete dominance
        global riskiest-states dominance
        pairwise riskiest-states dominance
        other ordinal techniques


    IV. General Multiple Facet Optimization

Social Choice Satisfy you
Satisfy me

Games Do well if opponent attacks on the left
Do well if opponent attacks on the right

Risky Decisions Do well in rain
Do well in sunshine

Multiple Objectives My car should have a low price
It should also be fast, roomy, attractive, and reliable
    V. Conclusion
    The goal is to maximize the efficient use of whatever information is actually available while minimizing the need for introducing arbitrary assumptions of questionable precision.