Building Decision Models that Modify Decision Systems

摘要

Two criticisms of decision analysis are that the amount of effort expended and time spent on modeling a problem is too burdensome and that the resulting model is applicable to only one specific problem. It can be helpful to use the information learned in one decision problem to solve another similar problem. Transferring information across problems, however, can require a time commitment comparable to analyzing that problem anew. These criticisms may be great enough to convince decision makers that practical application of decision analysis is infeasible. To address these criticisms, we designed an intelligent decision system that automatically builds decision models using an alternative form of decision trees known as influence diagrams. The system aids clinicians in evaluating alternatives in ventilator management. We believe that this intelligent decision system can decrease the amount of time users spend (and frustration they experience) in applying decision analysis and thus, it should increase the acceptance of decision analysis as a useful tool.