ABSTRACT
This paper describes research being carried out at The Ohio State University leading toward a general theory of information flow and analysis. The objectives of the National Science Foundation-sponsored research program† include the following: (1) to develop a theory of information flow and analysis; (2) to identify important parameters and variables in the information process which can be quantified and measured; (3) to develop relationships among the variables which describe their behavior and limitations; (4) to apply this theory to specific practical situations, particularly those involving science information; and (5) to develop both simulation and experimental models for quantification and validation of the theory.
The generalized model of information flow is shown to represent virtually any decision situation. Using this model we derive a number of measures. Two of the more important ones are: I = m Σ i = 1 m P ( a i ) 2 - 1 https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9780429050602/0fe33661-9c3b-419c-9156-035c581736e8/content/inline-math1.tif"/> and DME = [ Σ i = 1 m P ( a i ) E V i * ] / max https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9780429050602/0fe33661-9c3b-419c-9156-035c581736e8/content/inline-math2.tif"/> (EV* k ). With these measures we can quantify the amount of information (I) in a decision state and evaluate decision-maker effectiveness (DME). We can also determine the value of information through its 20effect on DME. Rules are derived for use by the decision-maker for assimilating new data in his estimates of values and for use in determining selection probabilities for various courses of action. These rules enable feedback, learning, and alternative selection to be modeled, measured, and evaluated. Research is underway to validate the model realistically and to apply it to practical situations.
A further possible result of this research is the development of a decision calculus which will establish guidelines for decision-making given certain situations. These guidelines should permit quantification of the importance of information in the decision-making process.
