Decision technology

This review is about decision technology-the rules and tools that help us m ake wiser decisions. First, we review the three rules that are at the heart of most traditional decision technology-multi-attribute utility, Bayes' th eorem, and subjective expected utility maximization. Since the inception of decision research, these rules have prescribed how we should infer values and probabilities and how we should combine them to make better decisions. We suggest how to make best use of all three rules in a comprehensive 19-st ep model. The remainder of the review explores recently developed tools of decision technology. It examines the characteristics and problems of decisi on-facilitating sites on the World Wide Web. Such sites now provide anyone who can use a personal computer with access to very sophisticated decision- aiding tools structured mainly to facilitate consumer decision making. It s eems likely that the Web will be the mode by means of which decision tools will be distributed to lay users. But methods for doing such apparently sim ple things as winnowing 3000 options down to a more reasonable number, like 10, contain traps for unwary decision technologists. The review briefly ex amines Bayes nets and influence diagrams-judgment and decisionmaking tools that are available as computer programs. It very briefly summarizes the sta te of the art of eliciting probabilities from experts. It concludes that de cision tools will be as important in the 21st century as spreadsheets were in the 20th.