Resource allocation under uncertainty

Rarely is the informational data required for the analysis of resource allocation decisions known with precision; uncertainty in such data is the rule, and not the exception. The conventional approach relying on data averaging, even when coupled with sensitivity analysis, limits the insights obtainable. Concepts of decision analysis under uncertainty are applied here to a controllable, Markov resource-allocative model. Proper application of these concepts provide the preferred framework for resolving informational data uncertainties.